LU88573A1 - Thrombopoietin. - Google Patents

Description

THROMBOPOIETIN

The present invention relates to the isolation, purification and recombinant or chemical synthesis of proteins which influence the survival, proliferation, differentiation or maturation of hematopoietic cells, in particular of platelet precursor cells. The present invention particularly relates to the cloning and expression of nucleic acids encoding a protein ligand that can bind to and activate mpl, a member of the cytokine receptor superfamily. The invention further relates to the use of these proteins alone or in combination with other cytokines for the treatment of immunological or hematopoietic impairments, including thrombocytopenia. I. The hematopoietic system

The hematopoietic system produces the mature, highly specialized blood cells that are known to be necessary for the survival of all mammals. These mature lines include erythrocytes, specialized in transport, of oxygen and carbon dioxide, T and B lymphocytan, responsible for cell and antibody mediated immune responses, platelets or platelets, specialized in forming blood clots, and granulocytes and macrophages , specializing in load cells and accessory lines to fight infections. Granulocytes are further subdivided into neutrophils, eosinophils, basophils and mast cells, whereby specialized cell types perform certain functions. Notably, all of these specialized mature blood cells come from a single common primitive cell type, referred to as the pluripotent (or totipotent) stem cell, found primarily in bone marrow (Dexer et al., Ann. Rev. Cell Biol., 3: 423- 441 [1987]).

The mature, highly specialized blood cells have to be produced in large numbers continuously throughout the life of a mammal. The vast majority of these specialized blood cells are designed to be functionally active for only a few hours to weeks (Cronkite et al., Blood Cells, 2: 263-284 [1976]). Thus, a continuous renewal of the mature blood cells, the primitive stem cells themselves, as well as any intermediate or progenitor cell lines located between the primitive and mature lines is necessary to meet the normal steady-state needs of the mammal Maintain blood cells.

The core of the hematopoietic system is based on the pluripotent stem cell or cell lines. These lines occur in relatively small numbers, and they undergo self-renewal through proliferation in order to produce daughter stem cells, or they are transformed, in a series of differentiation steps, into increasingly mature predecessor lines restricted to a certain lineage , ultimately forming the highly specialized mature blood cells.

For example, certain multipotent progenitor cells, called CFC mix, derived from stem cells, undergo proliferation (self-renewal) and development to form colonies, all of which are the various myeloid cells, erythrocytes, neutrophils, megakaryocytes (progenitors platelets), macrophages, basophils, eosinophils and mast cells. Other progenitor cells of the lymphoid lineage undergo proliferation and development to T cells and B cells.

Furthermore, between the CFC mix progenitor cells and myeloid cells, there is another rank of progenitor cells of an intermediate nature. on their offspring. These progenitor cells set on a lineage are classified based on the successors they generate. So are. the immediate known precursors of the myeloi lines: erythroid colony-forming units (CFU-E) for erythrocytes, granulocyte / macrophage colony-forming lines (GM-CFC) for neutrophils and macrophages, megakaryo-cytic colony-forming lines (Meg-CFC) for megakaryocytes, eosinophil colony-forming lines (Eos-CFC) for eosinophils, and basophil colony-forming lines (Bas-CFC) for mast cells. Other intermediate progenitor cells between the pluripotent stem cells and the mature blood cells are known (see below) or are likely to be found, with varying degrees of lineage and self-renewal ability.

That the principle underlying the normal hematopoietic cell system seems to be a reduced ability for self-renewal and the multipotency are lost and the establishment of a lineage is carried out and the degree of maturity is acquired. Thus, at one end of the hematopoietic cell spectrum is the pluripotent stem cell, which has the ability to self-renew and differentiate into all the different progenitor cells, which are fixed on a special lineage. This ability forms the basis of bone marrow transplantation therapy, where primitive stem cells rebuild the entire hematopoietic cell system. At the other end of the spectrum are the predecessors and their descendants, who are strongly committed to a lineage and have lost the ability to renew themselves but have acquired a mature functional activity.

The proliferation and development of stem cells and progenitor cells determined on a lineage is carefully controlled by a variety of hematopoietic growth factors or cytokines. The role of these growth factors in vivio is complex and incompletely understood. Some growth factors, such as interleukin-3 (IL-3), can both stimulate multipotent stem cells as well as established progenitor cells for different lineages, including e.g. Megakaryocytes. Other factors, such as the granulocyte / macrophage colony stimulating factor (GM-CSF), were originally thought to be limited in their effect on GM-CFC's. However, it was later found that GM-CSF also influences the proliferation and development of, among others, megakaryocytes. Thus, IL-3 and GM-CSF have been found to have overlapping biological activities with different potency. It has only recently been found that interleukin-6 (IL-6) and interleukin-11 (IL-11), although they have no obvious influence on meg colony formation alone, act synergistically with IL-3, which To stimulate maturation of megakaryocytes (Yonemura et al., Exp. Hematol., 20: 1011-1016 [1992]).

Thus, hematopoietic growth factors can affect the growth and differentiation of one or more lineages, they can overlap other growth factors in influencing a single progenitor cell line, or they can act synergistically with other factors.

It also appears that hematopoietic growth factors can exert their effects at various stages of cell development, from the totipotent stem cell to various progenitor cells, based on lineages, to the mature blood cell. For example, erythropoietin (epo) appears to promote proliferation only of mature erythroid progenitor cells. IL-3 appears to exert its effects earlier by influencing primitive stem cells and intermediate progenitor cells based on lineages.

It can be seen from the foregoing that new hematological-poetic growth factors affecting the survival, proliferation, differentiation or maturation of any blood cell or its progenitor would be useful, particularly to destroy the rebuilding of one to help the hematopoietic system caused by disease or after radiation or chemotherapy. II. Hegakaryocytopoiesis - platelet production

The regulation of megakaryocytopoiesis and platelet production is presented in an overview article: Mazur,

Exp. Hematol., 15: 248 [1987] and Hoffman, Blood, 74: 1196-1212 [1989]. Briefly stated, pluripotent bone marrow stem cells differentiate into megakaryocyte-like, erythrocyte-like and myelocyte-like cell lines. It is believed that there is a hierarchy of established megakaryocyte-like progenitor cells between the stem cells and the megakaryocytes. At least three classes of megakaryocyte-like progenitor cells have been identified, namely burst-forming-unit-megakaryocyres (BFU-MK), colony-forming-unit-megakaryocytes (CFU-MK) and light-density-mega-karyocyte progenitor cells (LD-CFU-MK). Megakaryocyte maturation itself is a continuum of development that has been resolved in stages based on standard morphological criteria. The earliest recognizable member of the megakaryocyte (MK or meg) family is the megakaryoblast.

These lines initially have a diameter of 20 to 30 μm with a basophilic cytoplasm and a slightly irregular nucleus, loose, somewhat reticular chromatin and several nucleoli. Later, megakaryoblasts can contain up to 32 nuclei (polyploid), but the cytoplasm remains scanty and immature. As the ripening process progresses, the nucleus becomes more lobular and pyknotic, the amount of cytoplasm increases and it has a more acidophilic and granular effect. The most mature lines of these families can take on the appearance of releasing platelets into their surroundings. Usually less than 10% of the mega-karyocytes are in your blast stage and more than 50% are mature. The arbitrary morphological classification that is usually applied to the megakaryocyte-like series gives the megakaryoblasts as the earliest form; Promegakaryocyte or basophilic megakaryocyte for the intermediate form; and mature (acidophilic, granular or platelet-producing) megakaryocyte for the late form. The mature megekaryocyte extends filaments of the cytoplasm into sinusoidal spaces, where they detach and fragment into individual platelets (Williams et al., Hematology, 1972).

Megakaryocytopoe Se is believed to include several regulatory factors (Williams et al., Br. J. Haematol., 52: 173 [198'2] and Williams et al., J. Cell Physiol., 110: 101 (1982 The early level of megakaryocytopoiesis is postulated as mitotic and affects cell proliferation and initiation of colony formation by CFU-MK, but is not affected by platelet counts (Burstein et al., J. Cell Physiol., 109: 333 [1981] and Kimura et al., Exp. Hematol., 13: 1048 [1985]). The late stage of maturation is non-mitotic, includes core polyploidization and cytoplasmic maturation, and is presumably regulated by a feedback mechanism by peripheral platelet count (Odell et al., Blood, 48: 765 [1976] and Ebbe et al., Blood, 32: 787 [1968]).

The existence of a specific and specific megakaryocyte colony stimulating factor (MK-CSF) has been controversially discussed (Mazur, Exp. Hematol., 15: 340-350 [1987]). However, most authors assume that such a vital survival process, like platelet production, is regulated by one or more cytokines that are solely responsible for this process. The hypothesis that megakaryocyte / platelet-specific cytokines or cytokines existed was the basis for more than 30 years of research - but to date such a cytokine has not been purified, sequenced, or proven by an assay to be a unique MK-CSF (TPO) been.

Although it has been reported that MK-CSF's have been partially purified from experimentally generated thrombocytopenia (Hill et al., Exp. Hematol., 14: 752 [1986] and human conditioned embryonic kidney medium [CM] (Mc Donald et al. , J. Lab. Clin. Med., 85:59 [1975] and from humans of patients with plastic anemia and with idiopathic thrombocytopenia-like purpura from urine extracts (Kawakita et al., Blood, 6: 556 [1983] and from plasma (Hoffman et al., J. Clin. Invest., 75: 11174 [1985]), their physiological function is so far in most cases still unknown.

The conditioned medium of pokeweed mitogen-activated spleen cells (pokeweed mitogen; PWM-SpCM) and the mouse myelomonocyte cell line WEHI-3 (WEHI-3CM) were used as megakaryocyte enhancers. PWM-SpCM contains factors that increase CFU-MK growth (Metcalf et al., Pro. Natl. Acad. Sci., USA, 72: 1744-1748 [1975]; Quesenberry et al., Blood, 65: 214 [1985]; and Iscove, NN, in Hematopoietic Cell Differentiation, ICN-UCLA Symposia on Molecular and Cellular Biology, 10, Golde et al., Eds. [New York, Academy Press] pages 37-52 [1978], by dtnen one is interleukin-3 (IL-3), a colony-stimulating factor for multiple lineages (multi-CSF [Burstein, Blood Cells, 11: 469 [1986]). The other factors in this medium have not yet been identified and isolated. WEHI -3 is a mouse myelomonocytic cell line that secretes relatively large amounts of IL-3 and smaller amounts of GM-CSF. IL-3 has been found to enhance the growth of a wide variety of hematopoietic cells (Ihle et al. , J. Immunol. 13: 282 [1983]). IL-3 has also been found to be syn. With many of the known hematopoietic hormones or growth factors ergistically interacts (Bartelmez et al., J. Cell Physiol., 122: 362-369 [1985] and Warren et al., Cell, 36: 667-674 [1988]) including with erythropoietin (EPO) and interleukin-1 ( IL-1) in the induction of very early multipotent precursors and the formation of very large mixed hematopoietic colonies.

Other sources of megakaryocyte enhancers have been found in the conditioned media of lungs, bones, macrophage cell lines, in mouse exudate cells and in human embryonic kidney cells. Despite some contradictory data (Mazur, Exp. Hematol., 15: 340-350 [1987]) there are some indications (Geissler et al., Br. J. Haematol., 60: 233-238 [1985]), that activated T-lymphocytes rather than monocytes play a reinforcing role in mega-karyocytopesis. These findings suggest that secretions of activated T lymphocytes, such as interleukins, can be regulatory factors in MK development (Geissler et al., Exp. Hematol., 15: 845-853 [1987]). A number of studies on megakaryocytopoiesis with purified erythropoietin EPO (Vainchenker et al., Blood, 54: 940 [1979]; McLeod et al., Nature, 261: 492-4 [1976]; and Williams et al ., Exp. Hematol., 12: 734 [1984]) indicate that this hormone has a reinforcing effect on MK colony formation. This has also been shown in serum-free and serum-containing kitsuren and in the absence of accessory lines (Williams et al., Exp. Hematol., 12: 734 [1984]). EPO has been claimed to be more involved in the single-cell and two-cell stages of megakaryocytopoiesis, as opposed to the effect of PWM-SpCM, which is involved in the four-cell stage of megakaryocytic development. The interaction of all these factors in the early and late phases of megakaryocytic development remains to be elucidated.

Data generated by various laboratories suggest that the only factors influencing multiple lineages that have individual MK colony stimulating activity are GM-CSF and IL-3 and to a lesser extent B cell stimulating factor IL-6 (Ikebuchi et al., Proc. Natl.

Acad. Sci. USA, 84: 9035 [1987]). Recently, several authors have reported that IL-11 and leukemia inhibiting factor (LIF) act synergistically with IL-3 to increase megakaryocyte size and ploidy (Yonemura et al.,

British Journal of Hematology, 84: 16-23 [1993]; Burstein et al., J. Cell. Physiol., 153: 305-312 [1992]; Metcalf et al., Blood, 76: 50-56 [1990]; Metcalf et al., Blood, 77: 2150-2153 [1991]; Bruno et al., Exp. Hematol., 19: 378-381 [1991]; and Yonemura et al., Exp. Hematol., 20: 1011-1016 [1992]).

Other important documents include: U.S. Patent No. 4,962,091; Chong, U.S. Patent No. 4,879,111; Fernandes et al., U.S. Patent No. 4,604,377; Wissler et al., U.S. Patent No. 4,512,971; Gottlieb, U.S. Patent No. 4,468,379; Bennett et al., U.S. Patent No. 5,215,895; Kogan et al., U.S. Patent No. 5,250,732; Kimura et al., Eur. J. Immunol., 20 (9): 1927-1931 [1990]; Secor et al., J. of Immunol., 144 (4): 1484-1489 [1990]; Warren et al., J. of Immunol., 140 (1): 94-99 [1988]; Warren et al., Exp. Hematol., 17 (11): 1095-1099 [1989]; Bruno et al., Exp. Hematol., 17 (10): 1038-1043 [1989]; Tanikawa et al., Exp. Hematol., 17 (8): 883-888 [1989]; Koike et al., Blood, 75 (12): 2286-2291 [1990]; Lotem, Blood, 75 (5): 1545-1551 [1989]; Rennick et al., Blood, 73 (7): 1828-1335 [1989]; and Clutterbuck et al., Blood, 73 (6): 1504-1512 [1989]. III. Thrombocytopenia

Platelets are critical elements in the blood coagulation mechanism. Depletion of the circulating platelet level, called thrombocytopenia, occurs under various clinical conditions and disorders. Thrombocytopenia is usually defined as a platelet count below 150 x 109 per liter. Major causes of thrombocytopenia can be broadly divided into three categories based on platelet life, namely: (1) impaired

Production of platelets by the bone marrow, (2) platelet sequestration in the spleen (splenomegaly) or (3) increased destruction of platelets in the peripheral circulation (e.g. autoimmune thrombocytopenia or chemotherapy and radiation therapy). Furthermore, patients who receive large volumes of rapidly administered low-platelet blood products may develop thrombocytopenia due to the dilution.

The clinical manifestations of bleeding in thrombocytopenia depend on the severity of thrombocytopenia, its cause and possibly related coagulation defects. In general, patients with platelet counts between 20 and 100 x 109 per liter run the risk of excessive post-traumatic bleeding, while those with platelet counts below 20 x 10 per liter may have spontaneous bleeding. The latter patients are candidates for platelet transfusion with an associated immunological and viral risk. For any given level of thrombocytopenia, bleeding appears to be worse if the cause is reduced production rather than increased platelet destruction. In the latter situation, the accelerated platelet turnover leads to the circulation of younger, larger and more heat-effective platelets. Thrombocytopenia can result from a variety of disorders, as briefly described below. A more detailed description can be found in Schaf-ner, AI, "Thrombocytopenia and Disorders of Platelet Function", Internal Medicine, 3rd Ed., John J. Hutton et al., Eds., Little Brown and Co., Boston / Toronto / London [1990]. (a) Thrombocytopenia due to impaired platelet production

Causes of congenital thrombocytopenia include constitutional aplastic anemia (Fanconi syndrome) and congenital amegakaryocytic thrombocytopenia, which may be associated with skeletal malformations. Acquired platelet production disorders are caused by either hypoplasia of megakaryocytes or ineffective thrombopoiesis. Megakaryocytic hypoplasia can result from a variety of conditions, including bone marrow aplasia (including idiopathic forms or myelosuppression or chemotherapy or radiation therapy), myelfibrosis, leukemia, and bone marrow invasion by metastatic tumors or granulomas. In some situations, toxins, infectious agents, or medications can interact relatively selectively with thrombopoiesis; Examples include transient thrombocytopenia caused by alcohol and certain viral infections, and mild thrombocytopenia associated with administration of thiazide diuretics. Finally, ineffective thrombopoiesis as a secondary consequence of megaloplastic processes (folate or Bi2-Man, 3el) can also cause thrombocytopenia, usually with coexisting anemia and leukopenia.

Current treatment for thrombocytopenia due to decreased platelet production depends on the identification and reversal of the underlying cause of bone marrow failure. Platelet transfusions are usually reserved for patients with serious bleeding complications or for protection during surgical interventions, since the isoimmunization of another platelet transfusion can offer increased resistance. Mucosal bleeding, which is the result of severe thrombocytopenia, can be attenuated by the oral or intravenous administration of antifibrinolytic agents. However, thrombotic complications can develop when antifibrinolytic agents are used in patients with widespread intravascular coagulation (DIC). (b) Thrombocytopenia due to splenic sequestration

Splenomegaly for any cause can be associated with mild to moderate thrombocytopenia. This is an essentially passive process (hypersplenia) of splenic platelet sequestration, in contrast to the active destruction of the platelets by the spleen in cases of the immune-mediated thrombocytopenia discussed below. Although the most common cause of hypersplenia is congestive splenomegaly due to portal hypertension due to alcoholic cirrhosis, other forms of congestion, infiltrative or lymphoproliferative splenomegaly are also associated with thrombocytopenia. The platelet counts typically do not fall below 50 x 109 per liter as a result of hypersplenia alone. (c) Thrombocytopenia due to non-immunologically mediated platelet destruction

Thrombocytopenia can result from the accelerated destruction of platelets in various nonimmunological processes. Disorders of this type include extensive intravascular coagulation, intravascular prosthesis objects, extracoporal blood circulation, and thrombotic microangiophilia, such as thrombotic thrombocytic purpura. In all of these situations, circulating platelets that are either exposed to the artificial surfaces or to an abnormal vascular intima are consumed or damaged at these locations and then removed by the reticuloendothelial system before maturation. Disease stages or disorders in which extensive intravascular coagulation (DIC) can occur are explained in detail in Braun-wald et al. (Editor), Harrison's Principles of Internal Medicine, 11th Edition, page 1478, McGraw Hill [1987]. Intra-vascular prosthetic devices, including cardiac valves and intraortal balloons, can cause mild to moderate destructive thrombocytopenia and transient thrombocytopenia in patients undergoing cardio-pulmonary bypass treatment or hemodialysis, resulting from the consumption or damage of platelets results in the extracorporeal circulation. (d) Drug-induced immune thrombocytopenia

More than 100 active substances have been associated with immunologically mediated thrombocytopenia. However, only quinidine, Chmin, gold, sulfonamides, cephalothm and heparin have been well characterized. Drug-induced thrombocytopenia is regularly very serious and usually occurs suddenly within days while the patient is taking the sensitizing medication. (e) Immunological (autoimmunological) thrombocytopenic purpura (ITP) ITP in adults is a chronic disease characterized by autoimmune platelet destruction. The autoantibody is typically IgG, although other immunoglobulins have also been reported. Although the autoantibody for ITP has been found to be associated with the GPIIbIIIa of the platelet membrane, in most cases the platelet antigen specificity has not been identified. Extravascular destruction of sensitized platelets occurs in the reticuloendotheial system of the spleen and liver. Although more than half of all cases of ITP are idiopathic, many patients have underlying rheumatic or autoimmune diseases (e.g. systemic lupus erythematosus) or lymphoproliferative disorders (e.g. chronic lymphocytic leukemia).

(f) HIV-induced ITP ITP is an increasingly common complication in HIV infection (Morris et al., Ann. Intern. Med., 96: 714-717 [1982]) and can occur at any stage of disease progression occur, both in patients for whom the acquired immunodeficiency syndrome (AIDS) has been diagnosed, as well as those with the AIDS-related complex, and those with an HIV infection but without AIDS symptoms. HIV infection is a communicable disease that is characterized in the final stages by a severe lack of cellular immune function as well as the appearance of opportunistic infections and cancer. The primary immunological abnormality that results from infection with HIV is the progressive depletion and functional impairment of T lymphocytes that express CD4 surface glycoprotein (Lane et al., Ann. Rev. Immunol., 3 : 477 [1985]). The loss of CD4 helper / T cell inductor function is believed to be due to the severe defects in cellular and humoral immunity, leading to the opportunistic infections and malignancies that are characteristic of AIDS (H Lane, see above).

Although the mechanism of HIV-associated ITP is unknown, it is believed to be different from the mechanism of ITP that is not associated with HIV infection (Walsh et al., N. Eng. J. Med., 311: 635-639 [1984]; and Ratner, Am. J. Med., 86: 194-198 [1989]). IV. Current therapy for thrombocytopenia

The therapeutic approach to treating thrombocytopenia patients is dictated by the severity and urgency of the clinical situation. Treatment is similar for HIV-associated and non-HIV-associated thrombocytopenia, and although a number of different therapeutic approaches have been used, therapy remains controversial.

The platelet count in patients diagnosed with thrombocytopenia has been successfully increased by glucocorticoid (e.g. prednisolone) therapy, however in most patients the response is incomplete ^ or relapse if the glucocorticoid dose is reduced or administration is discontinued, based on studies with patients with HIV-associated ITP, some researchers have suggested that glucocorticoid therapy can predispose to AIDS. Giucocorticoids are usually administered if the number of plates falls below 20 x 109 per liter or if there is spontaneous bleeding. For patients who do not respond to giucocorticoids, the compound is: 4- (2-chloropheny1) -9-methyl-2- [3- (4-morpholiny1) -3-propanone-l-yl] 6H-thieno [3,2, f] [1,2,4] triazolo [4,3, a,] [1,4] diazepm (WEB 2086) have been used successfully to treat a severe case of non-HIV-associated ITP. A patient with a platelet count of 37000-58000 / μ1 was treated with WEB 2086, and after 1-2 weeks of treatment the platelet count increased to 140000-190000 / μ1 (EP 361,077 and Lohman et al., Lancet, 1147 [1988]) .

Although optimal treatment for acquired amegakaryocytic thrombocytopenia purpura (AATP) is uncertain, antithymocyte globulin (ATG), a horse antiserum to human thymus tissue, has been shown to produce prolonged complete remission (Trimble et al., Am. J. Hematol., 37: 126-127 [1991]). However, a recent report suggests that the hematopoietic effect of ATG is due to thio-mersal, where the protein is believed to act as a mercury carrier (Panelle et al., Cancer Research, 50: 4429-4435 [1990]).

Good results have been reported with splenectomy. Splenectomy removes the major site of platelet destruction and a major source of autoantibody production in many patients. This procedure leads to prolonged treatment-free remission in a large number of patients. However, since surgical procedures are generally to be avoided in an immunocompromised patient, splenectomy is only recommended in severe cases of thrombocytopenia (e.g. worse HIV-associated ITP), and in patients who fail to receive 2 to 3 weeks of glucocorticoid treatment or who fail to get a sustained response after stopping glucocorticoid administration. Based on current scientific knowledge, it is unclear whether splenectomy predisposes patients to AIDS.

In addition to prednisolone therapy and splenectomy, certain cytotoxic agents, e.g. Vincristine and azidothymidine (AZT, zidovudine) benefits in the treatment of HIV-induced ITP; however, the results are still preliminary.

It can be seen from the foregoing that one way to treat thrombocytopenia is to obtain an agent that can accelerate the differentiation and maturation of megakaryocytes or precursors thereof to the platelet-producing form. Considerable efforts have been made to identify such agents, commonly referred to as "thrombopoietin" (TPO). Other names for TPO commonly found in the literature include thrombocytopoiesis stimulating factor (TSF), megakaryocyte colony stimulating factor (MK-CSF), megakaryocyte stimulating factor, and megacryte enhancer. TPO activity was observed as early as 1959 (Rak et al., Med. Exp., 1: 125), and attempts to characterize and purify this agent have continued to this day. While there have been reports of partial purification of TPO-active polypeptides (see, e.g., Tayrien et al., J. Biol. Chem., 262: 3262 [1987] and Hoffmann et al., J. Clin, Invest. 75: 1174 [1985]) have postulated that TPO is not a specific unit itself, but rather simply the polyfunctional manifestation of a known hormone (IL-3, Sparrow et al., Prog. Clin. Bic. Res., 215: 123 [1986]). Regardless of its shape or origin, a molecule with thrombopoietic activity would be of considerable therapeutic value. Although no protein has yet been unequivocally identified as TPO, the recent discovery that mpl, a putative cytokine receptor, could transduce a thrombopoietic signal is of great interest. V. Mpl is a megakaryocytopoietic cytokine receptor

It is believed that the proliferation and maturation of hematopoietic lines is strictly regulated by factors that positively or negatively modulate the proliferation of the pluripotent stem cell and differentiation into multiple lineages. These effects are mediated by the binding of extracellular protein factors with high affinity to specific cell surface receptors. These cell surface receptors share considerable homology and are generally classified as members of the cytokine receptor superfamily. Members of this super family include receptors for: IL-2 (β and γ chains) (Hatakeyama et al., Science, 244: 551-556 [1989]; Takeshita et al., Science, 257: 379-382 [ 1991]), IL-3 (Itoh et al., Science, 247: 324-328 [1990]; Gorman et al., Proc. Natl.

Acad. Sci. USA, 87: 5459-5463 [1990]; Kitamura et al., Cell, 66: 1165-1174 [1991a]; Kitamura et al., Proc. Natl. Acad.

Sci. USA, 88: 5082-5086 [1991b]), IL-4 (Mosley et al., Cell, 59: 335-348 [1989]), IL-5 (Takaki et al., EMBO J., 9: 4367- 4374 [1990]; Tavernier et al., Cell, 66: 1175-1184 [1991]), IL-6 (Yamasaki et al., Science, 241: 825-828 [1988]; Hibi et al., Cell, 63 : 1149-1157 [1990]), IL-7 (Goodwin et al., Cell, 60: 941-951 [1990]), IL-9 (Renault et al., Proc. Natl. Acad.

Sci. USA, 89: 5690-5694 [1992], granulocyte macrophage colony stimulating factor (GM-CSF) (Gearing et al., EMBO J., 8: 3667-3676 [1991]; Hayashida et al., Proc. Natl. Acad. Sci. USA, 244: 9655-9659 [1990], Granulocyte colony stimulating factor (G-CSF) (Fukunaga et al., Cell, 61: 341-350 [1990a]; Fukunaga et al., Proc. Natl. Acad. Sci. USA, 87: 8702-8706 [1990b]; Larsen et al., J. Exp. Med., 172: 1559-1570 [1990]), EPO (D'Andrea et al., Cell, 57: 277-285 [1989]; Jones et al., Blood, 76: 31-35 [1990], Leukemia Inhibiting Factor (LIF) (Gearing et al., EMBO J., 10: 2839-2848 [1991]), Oncostatin M (OSM) (Rose et al., Proc. Natl. Acad. Sci. USA, 88: 8641-8645 [1991]) and also receptors for prolactin (Boutin et al., Proc Natl. Acad. Sci. USA, 88: 7744-7748 [1988]; Edery et al., Proc. Natl. Acad. Sci. USA, 86: 2112-2116 [1989]), growth hormone (GH) (Leung et al., Nature, 330: 537-543 [1987]) and Zilliar Neurotrophic Factor (CNTF) (Davis et al., Science, 253: 59-63 [1991]).

Members of the cytokine receptor superfamily can be grouped into three functional categories (for an overview, see Nicola et al., Cell, 67: 1-4 [1991]). The first class includes receptors with a chain, such as the erythropoietin receptor (EPO-R) or the granulocyte colony stimulating factor (G-CSF-R) receptor, which have a ligand with high affinity for the extracellular domain bind and also generate an intracellular signal. A second class of receptors, called a-subunits, include interleukin-6 receptor (IL6-R), receptor for the granulocyte-macrophage colony stimulating factor (GM-CSF-R), interleukin-3 receptor ( IL3-Ra) and other members of the

Super cytokine receptor family. These a subunits bind low affinity ligands, but they cannot generate an intracellular signal. A high affinity receptor capable of signaling is generated by a heterodimer between an α subunit and a third class member of cytokine receptors, better known as β subunits, e.g. ßc, the common ß subunit for the three a subunits IL3-Ra and GM-CSF-R.

An indication that mpl is a member of the cytokine receptor superfamily comes from sequence homology (Gearing, EMBO J., 8: 3-667-3676 [1988]; Bazan, Proc. Natl. Acad. Sci. USA, 87: 6834-6938 [1990]; Davis et al., Science, 253: 59-63 [1991] and Vigon et al., Proc. Natl. Acad. Sci. USA, 89: 5640-5644 [1992]) and his ability to generate proliferative signals.

The protein sequence closed from the molecular cloning of the Mâuse-c-mpl shows that this protein is homologous to other cytokine receptors. The extracellular domain contains 465 amino acid residues and consists of two subdomains, each with four highly conserved cysteines and a special motif in the N-terminal subdomain and in the C-terminal subdomain. The extracellular domains that bind the ligand presumably have similar double β-sheet structure geometries. This duplicated extracellular domain is highly homologous to the signaling chain that is common to IL-3, IL-5 and GM-CSF receptors as well as the low affinity binding domain of LIF (Vigon et al., Oncogene, 8: 2607 -2615 [1993]). Thus, mpl can belong to the class of low affinity cytokine receptors for ligand binding.

A comparison of mouse mpl and mature human mpl P shows that these two proteins show 81% sequence identity. In particular, the N-terminal and C-terminal extracellular subdomains show 75% and 80% sequence identity, respectively. The most conserved mpl region is the cytoplasmic domain, which shows 91% amino acid identity, with a sequence of 37 residues near the transmembrane domain being identical in both species. As a result, mpl is reported to be one of the most conserved members of the cytokine receptor superfamily (Vigon, see above).

Evidence that mpl is a functional receptor that can transmit a proliferative signal comes from the construction of chimeric receptors that contain an extracallular domain from a cytokine receptor with high affinity for a known cytokine and that contain mpl cytoplasmic table domain. Since no known ligand for mpl has yet been reported, it was necessary to construct the chimeric extracellular domain with high affinity for ligand binding from the first class of a cytokine receptor such as IL-4R or G-CSFR. Vigon et al., Supra, merged the extracellular domain of G-CSFR with the transmembrane and cytoplasmic domain of c-mpl. An IL-3 dependent cell line, BAF / B03 (Ba / F3) was transfected with the G-CSFR / mpl chimer together with a full (length) G-CSFR control. The lines transfected with the chimera grew equally well in the presence of cytokine IL-3 or G-CSF. Similarly, lines transfected with G-CSRF also grew well in IL-3 or G-CSF, respectively. All lines died in the absence of growth factors. A similar experiment was carried out by Skoda et al., EMBO J.-, 12 (7): 2645-2653 [1993], in which the extracellular and transmembrane domains of human IL-4 receptor (IL-4-R) with the Muse-mpl cytoplasmic domains were fused, and this construct was transfected into an IL-3-dependent BA / F3 mouse cell line. Ba / F3 lines transfected with the wild-type hIL-4-R increased normally in the presence of each of the species-specific IL-4 or IL-3. Ba / F3 cells transfected with hIL-4R / mpl proliferated normally in the presence of hIL-4 (in the presence or absence of IL-3), showing that in Ba / F3 cells the mpl cytoplasmic domains contains all the elements necessary to generate a proliferative signal.

These chimeric experiments show the ability to generate proliferation-promoting signals through the mpl-cytoplasmic domain, but they say nothing about it. whether the mpl extracellular domain can bind a ligand. These results are in agreement with at least two possibilities, namely mpl is a receptor with a single chain (class one), such as EPO-R or G-CSFR, or it is a signal-transmitting ß-subunit (class three), the one a subunit is required, such as IL-3 (Skoda et al. see above). VI. The mpl ligand is a thrombopoietin (TPO)

As described above, it has been suggested that the serum contains a unique factor, sometimes referred to as thrombopoientin (TPO), which works synergistically with various other cytokines to promote megakaryocyte growth and maturation. No such natural factor has ever been isolated from serum or any other starting material, although considerable effort has been made in this direction by numerous groups. Although it is not known whether mpl can directly bind a mega-karyocyte stimulating factor, recent experiments show that mpl is involved in the transmission of a proliferative signal from one or more factors in the serum of aplastic patients Bone marrow can be found (Methia et al., Blood, 82 (5): 1395-1401 [1993]).

An indication that a unique colony stimulating factor from serum, which differs from IL-la, IL-33, IL-4, IL-6, IL-11, SCF, EPO, G-CSF and GM-CSF, a proliferative signal generated by mpl comes from examining the distribution of c-mpI expression in primitive and fixed hematopoietic cell lines and from mpl antisense studies in one of these cell lines.

Using reverse transcriptase (RT) -PCR in immunologically purified human hematopoietic cells, Methia et al., See above, showed that strong mpI-mRNA messages were only found in CD34 + purified cells, megakaryocytes and platelets. CD34 + cells, purified from bone marrow (BM), represent approximately 1% of all BM cells and are enriched with primitive and fixed progenitor cells for all lineages (e.g. erythroid-like, granulomacrophage-like and megakaryocyte-like).

Mpl antisense oligodeoxynucleotides have been shown to suppress megakaryocyte colony formation from pluripotent CD34 + cells cultured in serum from patients with aplastic bone marrow (a rich source of megakaryocyte colony stimulating activity [MK-CSA]) .

These same antisense oligodeoxynucleotides had no effect on erythroid or granuloma macrophage colony formation.

It was still unknown whether mpl directly bound a ligand and whether the serum factor, which was shown to cause megakaryocytopoiesis, acted via mpl. However, it has been suggested that if mpl bound a ligand directly, its amino acid sequence was likely to be highly conserved and show cross-reactivity between species due to the substantial sequence identity between human and mpl-ex-tracellular domains of the mouse (Vigon et al., See above [1993]). VII. Tasks

In view of the foregoing, it is apparent that there is a current and ongoing need to isolate and identify molecules that can stimulate the proliferation, differentiation, and maturation of hematopoietic cells, particularly megakaryocytes or their precursors for therapeutic Use in the treatment of thrombocytopenia. Such a molecule is believed to be an mpl ligand, and therefore there is a further need to isolate such a ligand to elucidate its role (s) in cell growth and differentiation.

Accordingly, an object of the present invention is to obtain a pharmaceutically pure molecule which can stimulate the proliferation, differentiation and / or maturation of megakaryocytes into the mature platelet-producing form.

Another object is to provide a molecule in a form suitable for therapeutic use in the treatment of a hematopoietic disorder, particularly thrombocytopenia.

It is another object of the present invention to isolate, purify, and particularly identify protein ligands that can bind in vivo to a cytokine superfamily receptor known as mpl to produce a proliferative signal.

It is a further object to provide nucleic acid molecules encoding such a protein ligand and to use these nucleic acid molecules to produce mpl-binding ligands in recombinant cell culture for diagnostic and therapeutic purposes.

It is another object to provide derivatives and modified forms of the protein ligand, including amino acid sequence variants, variant glycoprotein forms, and equivalent derivatives thereof.

Another task is to provide fusion polypeptide forms that combine an mpl ligand and a heterologous protein, and covalent derivatives thereof.

Another object is to provide variant polypeptide forms that combine an mpl ligand with amino acid additions and substitutions from the EPO sequence to produce a protein that promotes both platelet and red proliferation and growth Can regulate blood cell precursors.

It is another object to prepare immunogens for generating antibodies to mpl ligands or fusion forms thereof, as well as to obtain antibodies that can bind such ligands.

These and other objects of the invention will become apparent from the following description.

The problem underlying the invention are achieved by providing an isolated protein from mammals that promotes megakaryocytopoietic proliferation and maturation, referred to as the "mpl ligand" (ML) or "thrombopoietin" (TPO), which is responsible for proliferation, maturation and / or Differentiation of megakaryocytes to the mature platelet-producing form can stimulate.

The substantially homogeneous protein can be purified from a natural starting material by a method comprising: (1) contacting a plasma starting material containing the mpl ligand molecules to be purified with an immobilized receptor polypeptide, in particular mpl or an mpl fusion polypeptide, immobilized on one Carriers, under conditions in which the mpl ligand molecules to be purified are selectively adsorbed onto the immobilized receptor polypeptide, (2) washing the immobilized receptor polypeptide and its carrier to remove non-adsorbed material and (3) eluting the mpl-Li gand molecules from the immobilized receptor polypeptide to which they are adsorbed with an elution buffer. The natural starting material is preferably a mammalian plasma or urine containing the mpl ligand. The acid is possibly aplastic and the immobilized receptor is an mpl-IgG fusion.

Optionally, the preferred megakaryocytopoietic proliferation and maturation promoting prorein is an isolated, essentially homogeneous mpl ligand polypeptide made by synthetic or recombinant means.

The "mpI ligand" polypeptide or "TPO" of this invention preferably has at least 70% overall sequence identity with the amino acid sequence of the highly purified, substantially homogeneous porcine mpl ~ ligand polypeptide and at least 80% sequence identity with the "EPO domain" of the porcine mpl ligand polypeptide. Optionally, the mpl ligand of the invention is a mature human mpl ligand (hML) having the mature amino acid sequence shown in Fig. 1 (SEQ ID NO: 1), or a variant or post-transcriptionally modified form thereof, or a protein with approximately 80% sequence identity with mature human mpl ligand. If appropriate, the mpl ligand variant is a fragment, in particular an amino terminus or "EPO domain" fragment of the mature human mpl ligand (hML). Preferably, the amino terminal fragment contains substantially all of the human ML sequence between the first and fourth cysteine residues, but it may contain substantial additions, deletions, or substitutions outside of this region. According to this embodiment, the polypeptide fragment can be represented by the formula:

X-hML (7-151) "Y wherein hML (7-l51) represents the human TOP (hML) amino acid sequence from Cys7 to Cys151 including; X represents the amino group of Cys or one or more of the amino-terminal amino acid residues of the mature hML or amino acid residue extensions thereof such as Met, Tyr or leader sequences, containing e.g. proteolytic cleavage sites (e.g. factor Xa or thrombin); and Y represents the carboxy-terminal group Cys151 or one or more carboxy-terminal amino acid residues of the mature hML or extensions thereof .

Optionally, the mpl ligand polypeptide or a fragment thereof can be fused to heterologous polypeptide (chimera). A preferred heterologous polypeptide is a cytokine, colony-stimulating factor or interleukin or a fragment thereof, in particular kit ligand (KL), IL-1, IL-3, IL-6, IL-11, EPO, GM-CSF or LIF . An optionally preferred heterologous polypeptide is an immunoglobulin chain, in particular human IgGl, IgG2, IgG3, IgG4, IgA,

IgE, IgD, IgM or a fragment thereof, in particular comprising the constant domain of an IgG heavy chain.

In a further aspect of the present invention there is provided a composition comprising an isolated mpl agonist which is biologically active and which is preferably capable of incorporating labeled nucleotides (eg 3H-thymidine) into it Stimulate DNA from IL-3 dependent Ba / F3 cells transfected with human mpl. Optionally, the mpl agonist is a biologically active mpl ligand and is preferably capable of stimulating the incorporation of 35S into circulating platelets in a mouse platelet rebound assay. A suitable mpl agonist includes hMLi53, hML (R153A, R154A), hML2, hML3, hML4, mMl, mML2, mML3, pML and pML2 or fragments thereof.

In another embodiment of the invention, an isolated antibody is provided which can bind to the mpl ligand. The isolated antibody attached to the mpl

Ligand can optionally be fused to a second polypeptide and the antibody or fusion thereof can be used to isolate and purify the mpl ligand from a starting material as described above for immobilized mpl. In a further aspect of this embodiment, the invention provides a method for the detection of the mpl ligand in vitro or in vivo comprising contacting the antibody with an analysis sample, preferably a serum analysis sample. which is believed to contain the ligand and its detection when binding is complete.

In a further embodiment, the invention provides an isolated nucleic acid molecule which codes for the mpl ligand or fragments thereof, where the nucleic acid molecule can optionally be labeled with a detectable residue, and a nucleic acid molecule is provided with a sequence which is complementary to or under moderate or strongly stringent conditions with a nucleic acid molecule with a sequence which codes for an mpl ligand, hybridized. Preferred nucleic acid molecules are those encoding a huroan mpl ligand or from pig or mouse, and include RNA and DNA as well as genomic and cDNA. In a further aspect of this embodiment, the nucleic acid molecule is a DNA encoding the mpl ligand, and further comprising a replicable vector in which the DNA is operably placed under the control sequences recognized by a host that is associated with the Vector is transformed. Optionally, the DNA is a cDNA with the sequence as shown in Fig. 1, 5'-3 '(SEQ ID NO: 2), 3' -5 'or a fragment thereof. This aspect further includes host cells, preferably CHO cells, transformed with the vector, and a method using DNA to effect the production of an mpl ligand, preferably comprising expressing the cDNA encoding the mpl ligand , in a culture of the transformed host cells and recovery of the mpl ligand from the host cells or the host cell culture. The mpl ligand produced in this way is preferably a human mpl ligand.

The invention further includes a method of treating a mammal with a hematopoietic disorder, particularly thrombocytopenia, comprising administering to the mammal a therapeutically effective amount of an mpl ligand. If appropriate, the mpl ligand is administered in combination with a cytokine, in particular a colony-stimulating factor or interleukin. Preferred colony-stimulating factors or interleukins include kit ligand (KL), LIF, G-CSF, GM-CSF, M-CSF, EPO, IL-1, IL-3, IL-6 and IL-11.

The invention further comprises a method for isolating and purifying TPO (ML) from a TPO-producing microorganism, comprising: (1) breaking up or lysing lines containing TPO, (2) optionally separating soluble material from insoluble material containing TPO, (3) solubilizing TPO in the insoluble material with a solubilization buffer, (4) separating solubilized TPO from other soluble or insoluble material, (5) refolding TPO in a redox buffer, and (6) separating properly folded TPO of incorrectly folded TPO.

The method provides a chaotropic agent for solubilizing the insoluble material containing TPO, the chaotropic agent being selected from a salt of guanidine, sodium thiocyanate or urea. The method further provides that solubilized TPO is separated from other soluble or insoluble material by one or more steps selected from centrifugation,

Gel filtration and reverse phase chromatography. The refolding step of the method provides a redox buffer containing an oxidizing and reducing agent. Usually the oxidizing agent is oxygen or a compound containing at least one disulfide bond, and the reducing agent is a compound containing at least one free sulfhydryl group. Preferably the oxidizing agent is selected from oxidized glutathione (GSSG) and cystine, the reducing agent is selected from reduced glutathione (GSH) and cysteine. Oxidized glutathione (GSSG) and the reducing agent is reduced glutathione (GSH) are particularly preferred as the oxidizing agent. It is also preferred that the molar ratio of the oxidizing agent is equal to or greater than that of the reducing agent. The redox buffer additionally contains a detergent, preferably selected from CHAPS and CHAPSO, which is present in an amount of at least 1%. The redox buffer additionally contains NaCl, preferably at a concentration in the range of approximately 0.1-0.5 M, and glycerol, preferably at a concentration of greater than 15%. The pH of the redox buffer is preferably in the range of approximately pH 7.5 - pH 9.0, and the refolding step is carried out in 4 steps for 12-48 hours. The refolding step produces biologically active TPO, in which a disulfide bond is formed between the cysteine closest to the amino terminus with the cysteine closest to the carboxy terminus of the EPO domain.

The invention further comprises a method for purifying biologically active TPO from a microorganism comprising: (1) lysing at least the extracellular membrane of the microorganism, (2) treating the lysate containing TPO with a chaotropic agent (3) refolding the TPO, and (4) separating contaminants and misfolded TPO from properly folded TPO.

Fig.l shows the deduced amino acid sequence (SEQ ID NO: 1) from the human mpl ligand (hML) cDNA and the coding nucleotide sequence (SEQ ID.NO: 2). The nucleotides are numbered at the beginning of each line. The non-translated 5 'and 3' areas are indicated by lower case letters.

The amino acid residues are numbered above the sequence, starting at Ser 1 of the protein sequence of the mature mpI ligand (ML). The boundaries of the suspected exon 3 are indicated by the arrows and the potential N-glycosylation sites are framed by rectangles. Cysteine residues are indicated by a point above the sequence. The underlined sequence corresponds to the N-terminal sequence found for the mpl ligand isolated from porcine plasma.

Fig. 2 shows the method that was used for the

Assay for the detection of 3H-thymidine incorporation by the mpl ligand. To determine the presence of the mpl ligand in various sources, the mpl P-containing Ba / F3 lines were used

Withdraw IL-3 for 24 hours in a humidified incubator at 37 ° C in 5% CO2 and air. Following IL-3 withdrawal, the lines were plated in 96-well plates. with or without diluted analytical samples and for 24 hours in one

Cultivated cell culture incubator. 20 pl serum-free RPMI-Me-. . 3 dien containing 1 pCi H-thymid were added to each well for at least 6-8 hours. Then the rows were harvested on 96-well filter plates and washed with water. The filters were counted.

Figure 3 shows the effect of pronase, DTT and heat on the ability of APP to stimulate Ba / F3-mp2 cell proliferation. Pronase (Boehringer Mannheim) or bovine serum albumin was coupled to the pronase degradation of APP

Affi-Gel 10 (Biorad) and each incubated with APP for 18 h at 37 ° C. The resins were then removed by centrifugation and the supernatants were tested. APP was also warmed to 80 ° C for 4 min or set to 100 μΜ DTT, followed by dialysis against PBS.

Figure 4 shows the elution of mpl ligand activity from Phenyl-Toyopearl, Blue-Sepharose and Ultralink-mpl columns. Fractions 4-8 from the mpl affinity column were the peak activity fractions eluted from the column.

5 shows the SDS-PAGE of eluted ultralink mpl fractions. To 200 μΐ of each fraction 2-8 1 ml acetone was added containing 1 mM HCl at -20 ° C. After 3 hours at -20 ° C, the analysis samples were centrifuged and the pellets obtained were washed twice with acetone at -20 ° C. The acetone pellets were then dissolved in 30 μΐ SDS solubilization buffer, adjusted to 100 μΜ DTT and heated at 90 ° C. for 5 min. The analysis samples were then separated on a 4-20% SDS polyacrylamide gel and the proteins were visualized by silver staining.

Figure 6 shows the elution of mpl ligand activity from SDS-PAGE. Fraction 6 from the mpl affinity column was separated on a 4-20% SDS-polyacrylamide gel under non-reducing conditions. Following electrophoresis, the gel was cut into 12 equal strips and electroeluted as described in the examples. The electroeluted analysis sample was dialyzed in PBS and analyzed at a 1/20 dilution. The Mr standards used to calibrate the gel were Novex Mark 12 standards.

Figure 7 shows the effect of APP depleted in mpl ligand on human megakaryocytopoiesis. The mpl ligand-depleted APP was prepared by passing 1 ml through a 1 mpl affinity column (700 pg mpl-IgG / ml NHS-Superose, Pharmacia). Human peripheral stem cell cultures were adjusted to 10% APP or 10% mpl ligand-depleted APP and cultured for 12 days. Megakaryocytopoiesis was quantified as described in the examples.

Figure 8 shows the effect of mpl-IqG on the stimulation of human megakaryocytopoiesis by APP. Human peripheral stem cell cultures were adjusted to 10% APP and cultured for 12 days. On day 0, 2 and 4, mpl-IqG (0.5 pg) or ANP-R-IgG (0.5 pg) was added. After 12 days, megakaryocytopoiesis was quantified as described in the examples. The mean of duplicate analytical samples is shown graphically, with the actual duplicate data in parentheses.

Figure 9 shows both strands of a 390 bp fragment of human. genomic DNA encoding the mpl ligand. The inferred amino acid sequence for "Exon 3n (SEQ ID NO: 3), the coding sequence (SEQ ID NO: 4) and their complementary sequence (SEQ ID NO: 5) are shown.

Figure 10 shows the inferred amino acid sequence of the mature human mpl ligand (hML) (SEQ ID NO: 6) and mature human erythropoietin (hEPO) (SEQ ID NO: 7). The predicted amino acid sequence for the human mpl ligand is aligned with the human erythropoietin sequence. Identical amino acids are framed and gaps that are introduced for optimal alignment are indicated by dashes. Potential N-glycosylation sites are underlined with a solid line for hML and a broken line for hEPO. The two cysteines that are important for erythropoietin activity are marked by a large dot.

Figure 11 shows the inferred amino acid sequence of mature human mpl ligand isoforms hML (SEQ ID NO: 6), hML2 (SEQ ID NO: 8), hML3 (SEQ ID NO: 9) and hML4 (SEQ ID NO: 10).

Identical amino acids are framed and gaps that have been introduced for optimal alignment are indicated by dashes.

Figures 12A, 12B and 12C show the effect of human mpl ligands on Ba / F3 mpl cell proliferation (A), human megakaryocytopoiesis in vitro, guaranteed using a radiolabeled IgG monoclonal mouse antibody specific for the megakaryocyte glycoprotein GPIIbIIIa (B), and mouse thrombopoiesis measured in a platelet rebound assay (C). 293 cells were transfected by the CaPC> 4 method (Gorman, C in DNA Cloning: A New Approach 2: 143-190 [1985]) with pRK5 vector alone, pRK5-hML or with pRK5-ML153 overnight (pRK5 -ML153 was generated by inserting a stop codon after the rest 153 of hML by PCR). The media were conditioned for 36 h and tested for stimulation of cell proliferation of Ba / F3-mpl as described in Example 1 (A), or for human megakaryocytopoiesis in vitro (B). Megakaryocytopoiesis was quantified using a radiolabeled IgG monoclonal mouse antibody (HP1-1D) against the megakaryocyte-specific glycoprotein GPIIbHIa / as described (Grant et al., Blood 69: 1334-1339 [1987]). The effect of partially purified recombinant ML (rML) on in vivo platelet production (C) was determined using the rebound thrombocytosis assay as described by McDonald, T.P. Proc. Soc. Exp. Biol. Med. 144: 1006-10012 [1973]. Partially purified rML was made from 200 ml conditioned medium containing the recombinant ML. The medium was passed over a 2 ml Blue Sepa-rose column equilibrated in PBS and the column was washed with PBS and eluted with PBS containing 2M urea and NaCl each. The active fraction was dla3_ysiert in PBS and adjusted to 1mg / ml with endotoxin-free BSA. The analytical sample contained less than one unit of endotoxin / ml. Mice were injected with 64000, 32000 or 16000 units of rML or the excipient alone. Each group consisted of six mice. The mean and standard deviation of each group is shown. p-values were determined with the help of a 2-Tailed-T-Test comparing the values in the middle (medians).

Figure 13 compares the effect of human mpl ligand isoforms and variants in the Ba / F3 mpl cell proliferation assay. hML, Mock, hML2, hML3, hML (R153A, R154A) and hML153 were tested at different dilutions as described in Example 1.

FIGS. 14A, 14B and 14C show the inferred amino acid sequence (SEQ ID NO: 1) of human mp2 ligands (hML) or human TPO (hTPO) and the human genomic DNA coding sequence (SEQ ID NO: 11) . Nucleotides and amino acid residues are numbered at the beginning of each line.

Figure 15 shows an SDS-PAGE of purified 293-rhML332 and purified 293-rhMLi53.

Fig. 16 shows the nucleotide sequence: cDNA coding sequence (SEQ ID NO: 12) and deduced amino acid sequence (SEQ ID NO: 13) of the open reading frame of a ML isoform of the mouse. This mature mouse mpl ligand isoform contains 331 amino acid residues, four less than the putative full mML, and is therefore referred to as mML2. The nucleotides are numbered at the beginning of each line. Amino acid residues are numbered above the sequence, starting with Ser 1. The potential N-glycosylation sites are underlined. Cysteine residues are indicated by a point above the sequence.

Figure 17 shows the cDNA sequence (SEQ ID NO: 14) and the inferred protein sequence (SEQ ID NO: 15) of this ML mouse iso form (mML). The nucleotides are numbered at the beginning of each line. Amino acid residues are numbered above the sequence, starting with Ser 1. This mature mpl ligand isoform of the mouse contains 335 amino acid residues and is presumably the complete mpl ligand, designated mML. The signal sequence is indicated by an interrupted underline, and the probable point of cleavage is indicated by an arrow. The 5 'and 3' untranslated regions are identified by lower case letters. The deletions found as the result of alternative splicing (mML 2 and mML3) are underlined. The four cysteine residues are marked with a dot. The seven potential N-glycosylation sites are boxed.

Figure 18 compares the inferred amino acid sequence of the human ML isoform hML3 (SEQ ID NO: 9) and a mouse ML isoform, designated mML3 (SEQ ID NO: 16). The predicted amino acid sequence for the human mpl ligand is aligned with the mouse jnpl ligand sequence. Identical amino acids are framed and gaps that have been introduced for optimal alignment are indicated by dashes. Amino acids are numbered at the beginning of each line.

Figure 19 compares the predicted amino acid sequences of mature ML isoforms from mouse ML (SEQ ID NO: 17), porcine ML (SEQ ID NO: 18) and human ML (SEQ ID NO: 6). The amino acid sequences are aligned with gaps identified by dashes that have been introduced for optimal alignment. Amino acid residues are numbered at the beginning of each line, with identical residues framed. Potential N-glycosylation sites are indicated by a hatched box, and the cysteine residues are identified by a dot. The conserved dibasic amino acid motif, which represents a potential protease cleavage site, is underlined. The four amino acid long deletion. which occurs in all three species (ML2) is identified by a box in bold.

Figure 20 shows the cDNA sequence (SEQ ID NO: 19) and the predicted mature protein sequence (SEQ ID NO: 18) of a porcine ML isoform (pML). This porcine mpl ligand isoform contains 332 amino acid residues and is believed to be the complete porcine mpl ligand, designated pML. The nucleotides are numbered at the beginning of each line. The amino acid residues are numbered above the sequence, starting with Ser 1.

Figure 21 shows the cDNA sequence (SEQ ID NO: 20) and predicted mature protein sequence (SEQ ID NO: 21) of a porcine ML isoform (pML2). This porcine mpl ligand isoform contains 328 amino acid residues and is a form with a deletion of four residues of the complete porcine mpl ligand, designated pML2. The nucleotides are numbered at the beginning of each line. The amino acid residues are numbered above the sequence, starting with Ser 1.

Figure 22 compares the deduced amino acid sequence for the complete porcine ML isoform pML (SEQ ID NO: 18) and an porcine ML isoform designated pML2 (SEQ ID NO: 21). The predicted amino acid sequence for the pML is aligned with the pML2 sequence. Identical amino acids are framed and gaps that have been introduced for optimal alignment are indicated by dashes. The amino acids are numbered at the beginning of each line.

Figure 23 shows the relevant features of plasmid pSVI5.ID.LL.MLORF ("full length" or TPO332) used to transfect the CHO-DP12 host cells to produce CH0-rhTP0332.

Figure 24 shows the relevant features of plasmid pSVI5.ID.LL.MLEPO-D ("truncated" or TP0153) used to transfect the CH0-DP12 host cells to produce CH0-rhTP0153.

Figures 25A, 25B and 25C show the effect of E. coli rhTPO (Met_1, 153) on platelets (A), red blood cells (B) and (C) white blood cells in normal mice. Two groups of 6 female C57 B6 mice each were injected daily with either PBS buffer or 0.3 ml E. Coli-rhTPO (Met1, 153) (100 μ (sc.) · On day 0 and on day 3- 7 40 μΐ blood were taken from the orbital sinus. The blood was immediately diluted in 10 ml of a commercial diluent and the total blood count was obtained using a Serrono Baker Hematology Analyzer 9018. The data are given as means ± the standard deviation of the mean.

FIGS. 26A, 26B and 26C show the effect of E. Coli rhTPO (Met "1/153) on platelets (A), red blood cells (B) and (C) white blood cells in s.ublethal irradiated mice. Two

Groups of 10 female C57 B6 mice were sublethalized. 137 emits 750 cGy of gamma radiation from a Cs source and injected daily with either PBS buffer or 3.0 mg E. Coli-rhTPO (ΜΕΤ_1, i53) (100 μΐ sc.). On day 0 and at subsequent times, 40 μl of blood were taken from the orbital sinus. The blood was immediately diluted in 10 ml of a commercial diluent and the total blood count was obtained on a Serrono Baker Hematology Analyzer 9018. The data are given as mean ± standard deviation of the mean.

Figures 27A, 27B and 27C show the effect of CH0-rhTP0332 on (A) platelets (platelets), (B) red blood cells (erythrocytes) and (C) white blood cells (leukocytes) in normal mice. Two groups of 6 female C57 B6 mice were injected daily with either PBS buffer or 0.3 mg CH0-rhTP0332 (100 ml SC.). On day 0 and days 3-7, 40 ml of blood were withdrawn from the orbital sinus. The blood was immediately diluted in 10 ml of a commercial diluent and the total blood count was obtained using a Serrono Baker Hematology Analyzer 9018. The data are as

Mean values ± standard deviation of the mean value given. Figure 28 shows the dose response curve for different forms of rhTPO obtained from different cell lines. The dose response curves for rhTPO were generated from the following cell lines: hTP0332 from CHO (complete, from Chinese hamster ovary cells); hTPOMet_1, 153 (truncated form derived from E. Coli with an N-terminal methionine); hTP0332 (complete TPO from human 293 cells); Met-free 155 E-Coli (the shortened form [rhTP0155] without the N-terminal methionine from E. Coli). Groups of 6 female C57 B6 mice were injected with rhTPO daily for 7 days, depending on the group. Each day, 40 μΐ blood was drawn from the orbital sinus for the total blood count. The data shown above represent the maximum effect observed with the various treatments and with the exception of (Met 153 E-Coli) this followed on day 7 of treatment. In the aforementioned "Met 153 E-Coli" group, the maximum effect was observed on day 5. The data are given as mean values ± standard deviation of the mean value.

Figure 29 shows the dose response curves comparing the activity of complete and "cliped" forms of rhTPO · made in CHO cells with the truncated form from E. coli. Groups of 6 female C57 Ββ mice were injected daily with 0.3 jug of rhTPO of various types. On days 2-7, 40 μΐ blood was withdrawn from the orbital sinus for the total blood count. Treatment groups were TP0153, the truncated form of TPO from E. Coli; TP0332 (mixed fraction) full TPO containing approximately 80-90% complete and 10-20% trimmed forms; TP0332 (3 0K fraction) = cleaned, trimmed fraction?> Us of the original "Mix" preparation; TPO332 (70K fraction) = purified fraction with complete TPO from the original "mix" preparation. The data are given as means ± standard deviation of the mean.

Figure 30 is a carton showing the KIRA ELISA assay for measuring TPO. The figure shows the MPL / Rse.gD chimera and relevant parts of the output receptor as well as the final construct (right part of the figure) and a flow chart (left part of the figure) showing the relevant steps of the assay.

Figure 31 is a flow diagram for the KIRA ELISA assay showing each step of the procedure.

Figures 32A-32L give the nucleotide sequence (SEQ ID NO: 22). of the pSVI17.ID.LL expression vector used for the expression of Rse.gD in Example 17.

33 shows a schematic representation of the production of plasmid pMPl.

34 shows a schematic representation of the production of plasmid pMP21.

35 shows a schematic representation of the production of plasmid pMP151.

36 shows a schematic representation of the production of plasmid pMP202.

37 shows a schematic representation of the production of plasmid pMP172. /

38 shows a schematic representation of the production of plasmid pMP210.

Figure 39 shows a table of the five best TPO-expressing clones from the pMP210 plasmid bank (SEQ ID NOS: 23, 24, 25, 26, 27 and 28).

40 shows a schematic representation of the production of plasmid pMP4l.

41 shows a schematic representation of the production of plasmid pMP57.

42 shows a schematic representation of the production of plasmid pMP251. I. Definitions

In general, the following words and phrases have the meaning given when used in the description, examples, and claims. "Chaotropic agent" means a compound that, in an aqueous solution and at an appropriate concentration, can cause a change in the spatial configuration or conformation of a protein by at least partially destroying the forces responsible for maintaining normal secondary and tertiary structure of the protein. Such connections include e.g. Urea, guanidine HCl and sodium thiocyanate. High concentrations, usually 4-9 M, of these compounds are normally required to influence the conformation of the protein. "Cytokine" is a general term for proteins released by a cell population that act on another cell as intercellular mediators. Examples for . such cytokines are lymphokines, monokines and traditional polypeptide hormones. Be included among the cytokines

Growth hormone, insulin-like growth factor, human growth hormone, human N-methionyl growth hormone, rhin-growth hormone, parathyroid hormone, thyroxine, insulin, proinsulin, relaxin, prorelaxin, glycoprotein hormones such as follicle-stimulating hormone (FSH), thyroid stimulating hormone (TSH) and luteinizing hormone (LH), hematopoietic growth factor, hepatic growth factor, fibroblast growth factor, prolactin, placental tactogen, tumor necrosis factor-a (TNF-a and TNF-ß) mullerian inhibitory substance, gonadotropin-associated mouse peptide , Activin, vascular endothelial growth factor, integrin, nerve growth factors such as NGF-ß, platelet growth factor, transforming growth factors (TGFs) such as TGF-a and TGF-ß, insulin-like growth factors I and II, erythropoietin (EPO), osteoinductive factors , Interferons such as interferon-cc, ~ ß, ~ y, colony-stimulating factors (CSFs), such as macrophage-CSF (M-CSF), granulocyte macro pha-gene-CSF (GM-CSF) and granulocyte-CSF (G-CSF), interleukins (IL1s) such as IL-1, IL-la, IL-2, IL-3, IL-4, IL-5, IL -6, IL-7, IL-8, IL-9, IL-11, IL-12 and other polypeptide factors including LIF, SCF and kit ligand. As used herein, the aforementioned terms mean that they include proteins from natural sources or from recombinant cell cultures. Similarly, the terms are meant to include biologically active equivalents; e.g. which differ in the amino acid sequence by one or more amino acids or in the type or extent of glycosylation. "mpl ligand", "mpl ligand polypeptide", "ML", "thrombopoietin" or "TPO" are used interchangeably herein and include any polypeptide that has the ability to bind to mpl, a member of the cytokine receptor superfamily , and which has a biological property of the ML as defined above. An exemplary biological property is the ability to stimulate the incorporation of labeled nucleotides (e.g. 3H-thymidine) into the DNA of IL-3 dependent Ba / F3 cells, transiected with human mpl P. Another exemplary biological property is the ability to stimulate the incorporation of 35S into circulating platelets in a mouse platelet rebound assay. This definition encompasses the polypeptide isolated from an mpl ligand source, such as from an aplastic porcine plasma described herein, or from another source (starting material), such as another animal species, including humans, or produced by recombinant or synthetic methods. and includes variant forms including functional derivatives, fragments, alleles, isoforms and analogues thereof.

An "mpl ligand fragment" or "TPO fragment" is part of a naturally occurring mature complete mpl ligand or TPO sequence with one or more amino acid residues or carbohydrate units deleted. The deleted amino acid residues can occur anywhere in the peptide, including either at the N-terminal end or the C-terminal end or within the sequence. The fragment shares at least one biological property with the mpl ligand. Mpl ligand fragments have a sequential sequence of at least 10, 15, 20, 25, 30 or 40 amino acid residues that are identical to the sequence of the mpI ligand isolated from a mammal including the ligand isolated from aplastic porcine plasma or the human or mouse ligand, especially the EPO domain thereof. Representative examples of the N-terminal fragment are hML153 or TPO (Met 1 1-153) '. "Mpl ligand variants" or "mpl ligand sequence variants" as defined herein mean a biologically active mpl ligand, as defined below, with less than 100% sequence identity with the mpl ligand isolated from recombinant cell culture or porcine aplastic or that humane

Ligands with the inferred sequence as described in Fig. 1 (SEQ 'ID NO: 1). Typically, a biologically active mpl ligand variant has an amino acid sequence with at least about 70% amino acid sequence identity with the mpl ligand isolated from pig aplastic plasma or the mature mouse or human ligand or fragments thereof (see Fig. 1 [SEQ ID NO: 1] ), preferably at least about 75%, most preferably at least about 80%, even more preferably at least about 85% and even more preferred are at least about 90% and most preferred are at least about 95% identity.

A "chimeric mpI ligand" is a polypeptide comprising the complete mpl ligand or one or more fragments thereof that are fused or bound to a second heterologous polypeptide or one or more fragments thereof. The chimera has at least one biological property in common with the io.pl ligand. The second polypeptide will typically be a cytokine, immunoglobulin, or fragments thereof. "Isolated mpl ligand", "highly purified mpl ligand" and "substantially homogeneous mpl ligand" are interchangeable and mean an mpl ligand that has been purified from an mpl ligand source or that has been prepared by recombinant or synthetic methods , and sufficiently free of other peptides or proteins, (1) by at least 15 and preferably 20 amino acid residues on the N-terminal or an internal amino acid sequence using a spin-cup sequencer or the best commercially available amino acid sequencer on the Market or by modified methods published on the filing date of the present invention, or (2) which is purified to homogeneity by SDS-PAGE under non-reducing or reducing conditions using Coomassie Blue or preferably Sil staining. Homogeneity here means one

Contamination of less than 5% with other proteins of the starting material. "Biological property" when used in conjunction with either the "mpl ligand" or "isolated mpl ligand" means that it has thrombopoietic activity or an in vivo effector or antigen function or activity that directly or indirectly causes or exerts is by an mpl ligand (either in its native or denatured conformation) or by a fragment thereof. Effector functions include mpl binding and any carrier binding activity, agonism or antagonism of mpl, in particular the generation (transduction) of a proliferative signal, including replication, DFA regulating function, modulation of the biological activity of other cytokines, receptor (in particular Cytokine) activation, deactivation, up and down regulation, cell growth or differentiation and the like. Antigenic function means possessing an epitope or site capable of cross-reacting with antibodies directed against the native mpl ligand. The basic antigenic function of an mpl ligand polypeptide is that it binds with an affinity of at least about 106 l / mol to an antibody that is produced against the mpl ligand isolated from porcine aplastic plasma. The polypeptide usually binds with an affinity of at least approximately 107 l / mol. Particularly preferred is the antigenically active mpl ligand polypeptide which binds to an antibody which is directed against the mpl ligand with one of the effector functions described above. The antibodies used to define "biological activity" are rabbit polyclonal antibodies obtained by formulating the mpl ligand isolated from recombining cell culture or aplastic porcine plasma, in Freund's complete adjuvant, subcutaneously injecting the formulation and Boost the immune response by intraperitoneal injection of the formulation until the titer of the mpl ligand antibody reaches a plateau. "Egologically active", when used in conjunction with either the "mpl ligand" or "isolated mpl ligand" means an mpl ligand or polypeptide which shows the thrombopoietic activity or an effector function with the mpl ligands isolated from porcine aplastic plasma or expressed in recombinant cell culture as described herein. A principle known effector function of the mpl ligand or of the folypeptide is binding to mpl and stimulating the incorporation of labeled nucleotides (3H-thymidine) into the DNA of IL-3-dependent Ba / F3 cells, transfected with human mpl P. Another known effector function of the mpl ligand or polypeptide herein is the ability to stimulate the incorporation of 35S into circulating platelets in a mouse platelet rebound assay. Another known effector function of the mpl ligand is the ability to stimulate human megacaryocytopoiesis in vitro, which can be quantified using a radiolabeled monoclonal antibody specific for the megakaryocyte glycoprotein GPIIbIIIa. "Percent amino acid sequence identity" with respect to the mpl ligand sequence is defined herein as the percentage of amino acid residues in the sequence in question that are identical to the residues in the mpl ligand sequence isolated from aplastic sham plasma or the mouse or human ligand with the inferred amino acid sequence, as described in Figure 1 (SEQ ID NO: 1) after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percentage sequence identity, and conservative exchanges are not considered part of the sequence identity. None of the N-terminal, C-terminal or internal extensions, deletions or insertions into the mpl ligand sequence are considered to influence the sequence identity or homology. Thus, exemplary biologically active mpl ligand polypeptides, of which an identical sequence is assumed, include prepro-mpl ligand, pro-mpl ligand and mature mpl ligand. "Mpl ligand microsequencing" can be performed by any suitable standard procedure, provided the procedure is sensitive enough. In such a process, highly purified polypeptide obtained from SDS gels or a final HPLC step is directly sequenced by automated Edman (phenylisothiocyanate) degradation using a Model 470A Applied Biosystems gas phase sequencer equipped with a 12OA phenylthiohydantion ( PTH) amino acid analyzer. Furthermore, mpl ligand fragments produced by chemical (for example CNBr, hydroxylamine, 2-nitro-5-thiocyanobenzoate) or enzymatic (for example trypsin, clostripain, staphylococcus protease) digestion, followed by fragment purification (for example HPLC), can be sequenced in a similar manner will. PTH amino acids are analyzed using the ChromPerfect data system (Justice Innovations, Palo Alto, CA). The sequence interpretation is carried out on a VAX 11/785 Digital Equipment Co. Computer, as described by Henzel et al., J. Chromatography, 404: 41-52 [1987]. If necessary, aliquots of the HPLC fractions can be subjected to electrophoresis on 5-20% SDS-PAGE, electrotransfer to a PVDF membrane (ProBlott, AIB, Foster City, CA) and stained with Coomassie Brilliant Blue (Matsurdiara, J. Biol. Chem., 262: 10035-10038 [1987]). A specific protein identified by the staining is cut out of the blot and the N-terminal sequencing is carried out using the gas phase sequencer described above. For internal protein sequences, HPLC fractions are dried under vacuum (SpeedVAC), resuspended in a suitable buffer and digested with cyanogen bromide, the Lys-specific enzyme Lys-C (Wako Chemicals, Richmond, VA) or Asp-N (Boehringer Mannheim, Indianapolis , IN). To

Digestion, the peptides obtained are sequenced as a mixture or after HPLC separation on a C4 column, which is developed with a propanol gradient in 0.1% TFA before gas phase sequencing. "Thrombocytopenia" is defined as the number of platelets below 150X109 per liter of blood. "Thrombopoietic activity" is defined as biological activity that consists of accelerating the proliferation, differentiation and / or maturation of megakaryocytes or precursor megakaryocytes into the platelet-producing forms of these lines. This activity can be measured in various assays, including an in vivo mouse platelet rebound synthesis assay, by induction of a platelet surface antigen assay as measured by an anti-platelet immunoassay (anti-GPIIbIIIa) for a human megakaryoplastic leukemia cell line (CMK), and induction of polyploidization in a megakaryoplastic cell line (DAMI). "Thr.ombopoientin" (TPO) is defined as a compound with thrombopoietic affinity or which can increase the number of serum platelets in a mammal. TPO is preferably able to increase the endogenous platelet count by at least 10%, particularly preferably by 50%, and most preferably it can increase the platelet count in a human being to more than 150 x 109 per liter of blood. "Isolated mpl ligand nucleic acid" is RNA or DNA containing more than 16 and preferably 20 or more nucleotide bases in sequence encoding a biologically active mpl ligand or a fragment thereof, it is complementary to the RNA or DNA or hybridized to the RNA or DNA and remains stably bound under moderate to stringent conditions. This RNA or DNA is free of at least one contaminating nucleic acid of the starting material with which it is normally associated in the natural starting material, and is preferably essentially free of any other mammalian RNA or DNA. The expression "free of at least one contaminating nucleic acid of the starting material with which it is normally associated" includes the case where the nucleic acid is present in the starting material or in the natural cell, but it is at another chromosomal site or is flanked by other nucleic acid sequences that are not normally found in the starting material cell. An example of isolated mpl ligand nucleic acid is RNA or DNA encoding a biologically active mpI ligand that has at least 75% sequence identity, more preferably at least 80%, even more preferably at least 85% and even more preferably 90% and most preferably 95 % Sequence identity with the human, mouse or porcine jnpl ligand. "Control sequences" when reference is made to expression means DNA sequences which are necessary for the expression of a functionally linked coding sequence in a specific host organism. The control sequences suitable for prokaryotes include e.g. a promoter, possibly an operator sequence, a ribosome binding site and possibly further sequences which have hitherto been poorly understood. Eukaryotic lines are known to use promoters, polyadenylation signals and enhancers. "Functionally linked" when reference is made to nucleic acid means that the nucleic acids have a functional relationship with other nucleic acid sequences. For example, DNA for a presequence or secretory leader is operably linked to DNA for a polypeptide when expressed as a preprotein that participates in the secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; Or, a ribosome binding site is operably linked to a coding sequence if it is arranged to facilitate translation. Generally, "operably linked" means that the DNA sequences are continuous and, in the case of a secretory leader, continuous and linked in reading frame. However, enhancers do not have to be sequential.

The linking is effected by ligation at suitable restriction sites. If such sites do not exist, synthetic oligonucleotide adapters or linkers are used according to conventional methods. "Exogenous" when referring to an element means a nucleic acid sequence that is foreign to the cell or homologous to the cell, but is located in a position in the host cell nucleic acid where the element is not normally found. "Cell", "cell line" and "cell culture" are used interchangeably, and such terms include all descendants of a cell or cell line. Thus, for example, terms such as "transformants" and "transformed rows" encompass the primary cell and configurations derived from it, regardless of the number of transfers. It is also assumed that not all offspring have to be exactly identical in DNA content due to deliberate or unavoidable mutations. Mutant progeny that have the same function or biological activity after the originally transformed cell was screened are included. Where different names are intended, this becomes clear from the context. 4L · "Plasmids" are autonomously replicating circular DNA molecules that have independent origins of replication and are identified herein by a lowercase letter "p" preceded by capital letters and / or numbers.

The starting plasmids are either commercially available, publicly available in an unrestricted manner, or they can be prepared from such available plasmids according to published methods. Furthermore, other equivalent plasmids are known in the art and are familiar to those of ordinary skill in the art. "Restriction enzyme digestion (cleavage)" when reference is made to DNA means the catalytic cleavage of internal phosphodiester bonds of DNA with an enzyme which only acts at certain positions and locations in the DNA sequence. Such enzymes are called "restriction endonucleases". Each restriction endonuclease recognizes a specific DNA sequence, called "restriction site (cleavage site)", which shows double symmetry. The various restriction enzymes used herein are commercially available and their reaction conditions, cofactors and other requirements as indicated by the enzyme manufacturer are used. Restriction enzymes are usually referred to by abbreviations consisting of a capital letter followed by other letters which represent the microorganism from which the individual restriction enzyme was originally obtained, and then followed by a number which indicates the specific enzyme. In general, 1 ßq of plamide or DNA with approximately 1-2 units of enzyme in approximately 20 μΐ of buffer solution is used. Suitable buffers and substrate amounts for special restriction enzymes are specified by the manufacturer. Typically, an incubation for approximately 1 hour at 37 ° C is used, but it can vary in accordance with the manufacturer's instructions. After incubation, protein or polypeptide is removed by extraction with phenol and chloroform, and the split nucleic acid is recovered from the aqueous fraction by precipitation with ethanol. Cleavage with a restriction enzyme can be followed by hydrolysis of the terminal 5'-phosphates with bacterial alkaline phosphatase to avoid the two ends of the restriction-cleaved DNA of a DNA fragment "circularizing" or forming a closed loop, which is what Inserting another DNA fragment into the restriction site would make it more difficult. Unless otherwise stated, the cleavage of the plasmids is not followed by 5'-terminal dephosphorylation. The methods and reagents for dephosphorylation are conventional, as described in sections 1.56-1.61 of Sam-brook et al., Molecular Cloning: A Laboratory Manual [New York: Cold Spring Harbor Laboratory Press 19891. "Win" or "Isolating" a given DNA fragment from a restriction cleavage means separating the cleavage batch on a polyacrylamide or agarose gel by electrophoresis, identifying the fragment of interest by comparing its mobility with that of marker DNA fragments of known molecular weight, removing the gel section containing that desired fragment, and separating the gel from DNA. This method is generally known. For example, see Lawn et al., Nucleic Acids Res., 9: 6103-6114 [1981], and Goeddel et al., Nucleic Acids Res., 8: 4057 [1980]. "Southern analysis" or "Southern blotting" is a method in which the presence of DNA sequences in a restriction endonuclease cleavage batch of DNA or DNA-containing composition is confirmed by hybridization to a known labeled oligonucleotide or DNA fragment. Southern analysis typically involves electrophoretically separating a DNA cleavage batch on agarose gels, denaturing the DNA after electrophoretic separation, and transferring the DNA to nitrocellulose, nylon or other suitable membrane support for analysis with a radiolabeled, biotinylated or enzyme-labeled sample, such as described in Sections 9.37-9.52 of Sambrook et al., see above. "Northern analysis" or "Northern blotting" is a method used to identify RNA sequences that are linked to a known sample, such as an oligonucleotide, a DNA fragment, cDNA or a fragment thereof, or to an RNA

Hybridize fragment. The sample is with a radioisotope, 32 .... . P, or marked by biotmylation or with an enzyme. The RNA to be analyzed is usually electrophoresed on an agarose or polyacrylic amide gel, transferred to nitrocellulose, nylon or another suitable membrane and hybridized with the sample using standard techniques known in the art are well known, such as those described in Sections 7.39-7.52 of Sambrook et al., supra. "Ligation" is the process of forming phosphodiester bonds between two nucleic acid fragments. To ligate these two fragments, the ends of the fragments must be compatible with each other. In some cases, the ends can be directly compatible after endonuclease cleavage. However, it may first be necessary to convert the overhanging ends typically created after endonuclease cleavage to blunt ends in order to make them compatible for ligation. To create blunt ends, the DNA is treated in a suitable buffer for at least 15 minutes at 15 ° C with approximately 10 units of the Klenow fragment of DNA polymerase I or T4 DNA polymerase in the presence of the four deoxyribonucleotide triphosphates. The DNA is then purified by phenol-chloroform extraction and ethanol precipitation. The DNA fragments to be ligated with one another are introduced into a solution in approximately equimolar amounts. The solution also contains ATP, ligase buffer and a ligase such as T4 DNA ligase with approximately 10 units per 0.5 µg per DNA. If the DNA is to be ligated into a vector, the vector is first linearized by cleavage with the appropriate restriction endonuclease (s). The linearized fragment is then treated with bacterial alkaline phosphatase or intestinal calf phosphatase to prevent self-ligation during the ligation step. “Preparing” DNA from rows means isolating the plasmid DNA from a culture of the host cells. Typically used methods for DNA production are the large and small scale plasmid production processes as described in Sections 1.25-1.33 of Sambrook et al., Supra. After the DNA has been prepared, it can be purified using methods known in the art, as described in section 1.40 of Sambrook et al., See above. "Oligonucleotides" are short chain, single or double stranded polydeoxynucleotides that are chemically synthesized using known methods such as phosphotriester, phosphite or phosphoramidite chemistry using solid phase techniques as described in EP 266,032, published May 4, 1988, Or via deoxynucloside H-Phos-phonate intermediates as described by Froehler et al., Nucl. Acids Res., 14: 5399-5407 [1986]. Other methods include the polymerase chain reaction as described below and other self-priming (autoprimeric) methods and solid support oligonucleotide syntheses. All of these methods are described in Engels et al., Angew. Chem. Int. Ed. Engl., 28: 716-734 [1989]. These methods are used when the complete nucleic acid sequence of the gene is caressed or when the nucleic acid sequence complementary to the coding strand is available. Alternatively, if the target amino acid sequence is known, one can deduce potential nucleic acid sequences using known and preferred coding residues for each amino acid residue. The oligonucleotides are then purified on polyacrylic amide gels. "Polymerase chain reaction" or "PCR" refers to a method or technique in which minute amounts of a specific piece of nucleic acid, RNA and / or DNA are multiplied, as described in U.S. Patent No. 4,683,195, issued July 28 1987. Ira general, sequence information must be available from the ends of the region of interest or beyond, so that oligonucleotide primers can be designed. These primers are identical or similar in sequence to the counterstrands of the template to be multiplied. The 5'-terminal nucleotides of the two primers can match the ends of the material to be multiplied. PCR can be used to multiply specific RNA sequences, specific DNA sequences from whole genomic DNA and cDNA, transcribed from whole cellular RNA, bactriopha gene or plasmid sequences, etc. See generally Mullis et al., Cold Spring Harbor Symp. Quant. Biol., 51: 263 [1987]; Erlich, ed., PCR Technology, (Stockton Press, NY, 1989). As used herein, PCR is considered to be, but not the only, example of a nucleic acid polymerase chain reaction method for multiplying a nucleic acid test assay sample that includes using a known nucleic acid as a primer and a nucleic acid polymerase to replicate or generate a specific piece of nucleic acid. "Stringent conditions" are those that (1) use low ionic strength and high temperature for washing, e.g. 0.015 M NaCl / 0.0015 M sodium citrate / 0.1% NaDodS04 (SDS) at 50 ° C, or (2) using a denaturing agent such as formamide during hybridization, e.g. 50% (vol / vol) formamide with 0.1% bovine serum albumin / O, 1% Ficoll / 0.1% polyvinylpyrrolidone / 50 mM sodium phosphate buffer at pH 6.5 with 750 mM NaCl, 75 mM sodium citrate at 42 ° C.

Another example is the use of 50% formamide, 5 X SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt's solution, sonicated Salmon sperm DNA (50 mg / ml), 0.1% SDS and 10% dextran sulfate at 42 ° C with washing at 42 ° C in 0.2xSSC and 0.1% SDS. "Moderately stringent conditions" are described in Sam-brook et al., Supra, and include the use of a washing solution and hybridization conditions (e.g. temperature, ionic strength and% SDS) that are less stringent than described above. An example of moderately stringent conditions are conditions such as overnight incubation at 37 ° C. in a solution comprising: 20% formamide, 5 x SSC (150 mM NaCl, 15 mM trisodium citrate), 50 mM sodium phosphate (pH 7.6), 5 x Denhardt's solution, 10% dextran sulfate and 20 μΐ / ml denatured sheared salmon sperm DNA, followed by washing the filter in 1 x SSC at approximately 37-50 ° C. One of ordinary skill in the art will recognize how to adjust temperature, ionic strength, etc. to adjust for factors such as sample length and the like. "Antibodies" (Abs) and "Immunoglobulins" (Igs) are glycoproteins with the same structural properties. While antibodies show binding specificity for a specific antigen, immunoglobulins include both antibodies and other antibody-like molecules that have no antigen specificity. Polypeptides of the latter type are e.g. produced in low levels by the lymphatic system and with increased levels by myelomas. "Native antibodies and immunoglobulins" are usually heterotetrameric glycoproteins of approximately 150,000 daltons, which are composed of two identical light (L) chains and two identical heavy (H) chains. Each light chain is linked to a heavy chain via a covalent disulfide bond, while the number of disulfide linkages varies between the heavy chains for different immunoglobulin isotypes. Every heavy and light chain also has evenly arranged disulfide bridges within the chain. Each heavy chain has a variable domain (VH) at one end followed by a number of constant doxnaenes. Each light chain has a variable domain at one end (V ^) and a constant domain at its other end; the light chain constant domain is aligned with the first heavy chain constant domain, and the light chain variable domain is aligned with the heavy chain variable mandrel. Special amino acid residues presumably form the interface between the variable domains of the light and heavy chain (Clothia et al., J. Mol. Biol., 1986: 651-663 [1985]; Novotny and Haber, Proc. Natl. Acad. Sci. USA , 82: 4592-4596 [1985]).

The term "variable" refers to the fact that certain portions of the variable domain differ widely in the sequence between antibodies, and are involved in the binding and specificity of each particular antibody for its particular antigen. However, the variability is not evenly distributed across the variable domains of antibodies. It is concentrated in three

Segments, so-called complementarity-determining regions (CDRs) or hypervariable regions, which occur both in the variable domains of the light chain and the heavy chain. The more conserved parts of the variable domains are called the so-called "framework" (FR). The variable domains of the native heavy and light chains each comprise four FR regions, which essentially assume a β-sheet configuration, connected to three CDRs, which form loops, which are connected to the β-sheet structure and in some cases part of it form. The CDRs in each chain are held in close proximity by the FR regions and, with the CDRs of the other chain, contribute to the formation of the antibody's antigen binding site (see Kabat et al., Sequences of Proteins of Immunological Interest, National Institute of Health, Bethesda, MD [1987]). The constant doraenes are not directly involved in the binding of an antibody to an antigen, but they show various effector functions, such as the participation of the antibody in the antibody-dependent cellular toxicity.

Papain cleavage of antibodies produces two identical anti-gene binding fragments called "Fab" fragments, each having a unique antigen binding site, and a remaining "Fc" fragment, the name of which reflects its ability to crystallize easily. Pepsin treatment produces an F (ab ') 2 fragment that contains two antigen binding sites and that can still crosslink antigen. "Fv" is the minimal antibody fragment that contains a complete antigen recognition and binding site. This region consists of a dimer from a variable domain of a heavy and a light chain in a close, non-covalent association. In this configuration, the CDRs of each variable domain interact with each other to define an antigen binding site on the surface of the VH-VL dimer. The six CDRs collectively confer antigen binding specificity to the antibody. However, even a single variable domain (or half of an Fv comprising only three CDRs specific for an antigen) has the ability to recognize and bind to an antigen, although with less affinity than the full binding site.

The Fab fragment also contains the light chain constant domain and the heavy chain first constant domain (CHI). Fab 'fragments differ from Fab fragments in that a few residues are added to the carboxy terminus of the CHI heavy chain domain, including one or more cysteines from the hinge region of the antibody. Fab'-SH is the name for Fab ', in which the cysteine residue (s) of the constant domain carries a free thiol group. F (ab ') 2 antibody fragments were originally made as pairs of Fab' fragments with hinge cysteines between them. Other chemical coupling variants of antibody fragments are also known.

The "light chains" of antibodies (immunoglobulins) of any vertebrate type. can be assigned to one of two distinctly different types, called kappa and lambda (λ), based on the amino acid sequences of the constant domains.

Depending on the amino acid sequence of the constant domain of their heavy chains, immunoglobulins can be assigned to different classes. There are five main classes of immunoglobulins: IgA, IgD, IgE, IgG and IgM, and some of them can be further divided into subclasses (isotypes), e.g. IgG-1, IgG-2, IgG-3 and IgG-4; IgA-1 and IgA-2. The constant domains of the heavy chain that the different

Corresponding classes of immunoglobulins are called a, delta, epsilon, γ and μ. The subunit structures and the three-dimensional configurations of the different classes of immunoglobulins are well known.

The term "antibody" is used in the broadest sense and relates in particular to individual monoclonal antibodies (including agonist and antagonist antibodies), antibody compositions with polyepitopic specificity, and also antibody fragments (eg Fab, F (ab ') 2 and Fv) , As long as they show the desired biological activity. "Monoclonal antibody" as used herein refers to an antibody obtained from a population of essentially homogeneous antibodies, i.e. the individual antibodies included in the population are identical except for any natural mutations that may occur, which may be present in small amounts. Monoclonal antibodies are highly specific and are directed against a single antigenic site. Furthermore, in contrast to conventional (polyclonal) antibody preparations, which typically contain different antibodies which are directed against different determinants (epitopes), each monoclonal antibody is directed against a single determinant on the antigen. In addition to their specificity, the monoclonal antibodies are advantageous in that they are synthesized by the hybridoma culture and are not contaminated by other immunoglobulins. The modifying term "monoclonal" indicates the character of the antibody that it is obtained from an essentially homogeneous population of antibodies, and it does not mean that the production of the antibody requires a special procedure. For example, the monoclonal antibodies to be used in accordance with the present invention can be made by the hybridoma method first described by Kohler & Milstein, Nature, 256: 495 (1975), or they can be made by recombinant DNA methods (see e.g., U.S. Patent No. 4,816,567 [Cabilly et al.]).

The monoclonal antibodies herein specifically include "chimeric" antibodies (immunoglobulins) in which a portion of the heavy and / or light chain is identical to or homologous to the corresponding sequences in antibodies that are of a particular type or to a particular anti body class or subclass, while the rest of the chain (s) are identical or homologous to corresponding sequences in antibodies obtained from other species or belong to a different antibody class or subclass, as well as fragments of such antibodies, as long as they have the desired biological Show activity (U.S. Patent No. 4,816,567 (Cabilly et al.); And Morrison et al., Proc. Natl. Acad. Sci. USA, 81: 6851-6855 [1984]. "Humanized" forms of non-human (e.g., mice ) Antibodies are chimeric immunoglobulins, immunoglobulin chains or

Fragments thereof (such as Fv, Fab, Fab ', F (ab') 2 or other antigen-binding sub-sequences of the antibodies) which contain a minimal sequence derived from a non-human immunoglobulin. Most often, humanized antibodies are human immunoglobulins (recipient antibodies) in which residues of a complementarity-determining region (CDR) of the recipient are replaced by residues of a CDR from a non-human species (donor antibodies) such as a mouse, rat or rabbit, which the desired Has specificity, affinity and capacity. In some cases, Fv framework residues of human immunoglobulin are replaced by corresponding non-human residues. Furthermore, a humanized antibody can comprise residues which do not occur either in the recipient antibody or in the introduced CDR or the framework sequences. These modifications are made to further refine and optimize the performance of the antibody. In general, the humanized antibody will comprise substantially all of the portion of at least one, and typically two, variable domains in which all or substantially all of the CDR regions correspond to those of a non-human immunoglobulin, and · the complete or substantially complete FR regions are from a human immunoglobulin consensus sequence.

The humanized antibody will optimally include at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. For further details see: Jones et al., Nature, 321: 522-525 [1986]; Reichmann et al., Nature, 332: 323-329 [1988] and Presta, Curr. Op. Struct. Biol., 2: 593-596 [1992]). "Non-immunogenic in humans" means that when the polypeptide is brought into contact in a pharmaceutically acceptable carrier and in a therapeutically effective amount with the appropriate human tissue, there is no state of sensitivity or resistance to the polypeptide on the second administration of the polypeptide is detectable after a suitable latency period (eg 8 to 14 days). II. Preferred embodiments of the invention

Preferred polypeptides according to the invention are essentially a homogeneous polypeptide or polypeptides, designated as mpl ligand (s) or thrombopoietin (TPO), which have the property of binding to mpl, a member of the receptor cytokine superfamily, and which have the biological property of Stimulating the incorporation of labeled nucleotides (3H-thymidine) into the DNA of IL-3-dependent Ba / F3 cells, transfected with human mpl P. More preferred mpl ligand (s) are isolated mammalian protein (s) with hematopoietic, in particular megakaryocytopoietic or thrombocytopoietic activity - namely they can proliferate, mature and / or differentiate immature mega-karyocytes or their precursors into the mature platelets. stimulate production form. Most preferred polypeptides according to the invention are human mpl ligand (s) including fragments thereof with hematopoietic, megakaryocytopoietic or thrombopoietic activity. If appropriate, these human mpl ligand (s) may not have any glycosylation. Other preferred human mpl ligands are the "EPO domain" of hML, designated hMLis3 or I1TOP153, a truncated form of hML, designated hML245 or hTP0245, and the mature complete polypeptide with the amino acid sequence as shown in Fig. 1 (SEQ ID NO: 1), referred to as hML, hML332 or hTP0332 / and the biologically active substituted variant - hML (R153A, R154A).

Optionally preferred polypeptides according to the invention are biologically or immunologically active mpl ligand variants selected from hML2, hML3, hML4, mML, mML2, mML3, pML and pML2.

Optionally preferred preferred polypeptides according to the invention are biologically active mpl ligand variant (s) which have an amino acid sequence with at least 70% amino acid sequence identity with the human mpl ligand (see FIG. 1 [SEQ ID NO: 1]), the mouse mpl ligand (see Fig. 16 [SEQ ID NOS: 12 & 13]), the recombinant porcine mpI ligand (see Fig. 19 [SEQ ID NO: 18]) or the porcine mpl ligand Aplastic porcine plasma, preferably at least 75%, more preferably at least 80%, even more preferably at least 85%, most preferably at least 90% and most preferably at least 95% identity.

The mpI ligand isolated from aplastic porcine plasma has the following properties: (1) the partially purified ligand elutes from a gel filtration column run either in PBS, PBS containing 0.1% SDS or PBS containing 4M MgCl2 with Mr of 60000-70000 ; (2) the ligand activity is destroyed by pronase; (3) the ligand is stable up to the low pH (2.5), SDS up to 0.1% and 2M urea; (4) the ligand is a glycoprotein based on its binding to various lectin columns; (5) the highly purified ligand eluted from non-reducing SDS-PAGE with a molecular weight of 25000-35000. Smaller amounts of activity also elute with molecular weights of 18,000-22,000 and 60,000; (6) the highly purified ligand is separated on reducing SDS-PAGE as a doublet with Mr of 28000 and 31000; (7) the amino terminal sequence of the 18000-22000, 28000 and 31000 bands is the same - SPAPPACDPRLLNKLLRDDHVLHGR (SEQ ID NO: 29); and (8) the ligand binds and elutes from the following affinity columns

Blue-Sepharose, CM Blue-Sepharose, MONO-Q, MONO-S,

Lentil-lectin-Sepharose, WGA-Sepharose,

Con A-Sepharose,

Ether 650m Toyopearl,

Butyl 650m Toyopearl,

Phenyl 650m Toyopearl and Phenyl Sepharose.

More preferred mpl ligand polypeptides are those which are encoded by the human genomic or cDNA, with an amino acid sequence as described in FIG. 1 (SEQ ID NO: 1).

Other preferred naturally occurring biologically active inpl ligand polypeptides according to the invention include pre-pro mpl ligand, pro mpl ligand, mature mpI ligand, mpl ligand fragments and glycosylation variants thereof.

Still further preferred polypeptides of the invention include mpl ligand sequence variants and chimeras. Preferred preferred mpl ligand sequence variants and chimeras are biologically active mpl ligand variants which have an amino acid sequence with at least 70% amino acid sequence identity with the human mpl ligand or the mpl ligand isolated from aplastic porcine plasma, preferably with at least 75% with at least 80%, more preferably with at least 85%, even more preferably with at least 90% and very particularly preferably with at least 95% identity. An exemplary preferred mpl ligand variant is an N-terminal domain hML variant (referred to as the "EPO domain" because of its sequence homology to erythropoietin). The preferred hML EPO domain comprises approximately the first 153 amino acid residues of mature hML and is referred to as hMLi53. An optionally preferred hML sequence variant comprises one in which one or more of the basic or dibasic amino acid residue (s) in the C-terminal domain is substituted with a non-basic amino acid residue (s) (eg hydrophobic, neutral, acidic) aromatic

Gly, Pro and similar). A preferred hML-C-terminal domain sequence variant comprises one in which Arg residues 153 and 154 are replaced with Ala residues. This variant is called hML332 (R153A, R154A). An alternatively preferred hML variant comprises either hML332 or hML153 in which amino acid residues 111-114 (QLPP or LPPQ) are deleted or replaced with a different tetrapeptide sequence (e.g. AGAG or the like). The deletion mutants mentioned above are called Ä4hML332 or A4hML153.

A preferred chimera is a fusion between the mpl ligand or a fragment (defined below) thereof with a heterologous polypeptide or a fragment thereof. For example, hMLis3 can be fused to an IgG fragment to improve its serum half-life or to IL-3, G-CSF or EPO to produce a molecule with increased thrombopoietic or chimeric hematopoietic activity.

An alternative preferred human mpl ligand chimera is an "ML-EPO domain chimera" consisting of the N-terminal 153 to 157 hML residues substituted with one or more, but not all, residues of human EPO, approximately aligned as in Fig. 10 (SEQ ID NO: 7). In this embodiment, the hML chimera is approximately 153-166 residues long, in which individual or blocks of residues from the human EPO sequence are added or residues in the hML sequence are substituted at positions corresponding to the orientation, as in Figure 10 (SEQ ID NO: 6). Exemplary block-like sequence insertions in the N-terminal portion of hML include one or more of the N-glycosylation sites at positions (EPO) 24-27, 38-40 and 83-85; one or more of the four predicted amphipatic a-helical bundles at positions (EPO) 9-22, 59-76, 90-107 and 132-152; and other highly conserved regions including the N-terminal and C-terminal regions and residue positions (EPO) 44-52 (see, e.g., Wen et al., Blood, 82: 1507-1516 [1993] and Boissel et al., J. Biol Chem., 268 (21): 15983-15993 [1993]). It is evident that this "ML-EPO domain chimera" has mixed thrdmbopoietic, erythropoietic (TEPO) biological activity.

Other preferred polypeptides of the invention include mpl ligand fragments having a contiguous sequence of at least 10, 15, 20, 25, 30 or 40 amino acid residues that are identical to the sequences of the mpl ligand isolated from porcine aplastic plasma or the human mpl ligand as described herein. (see table 14, example 24). A preferred mpl ligand fragment is human ML [1-X], where X is 153, 164, 191, 205, 207, 217, 229 or 245 (see FIG. 1 (SEQ ID NO: 1) for the sequence of residues 1 -X). Other preferred mpl ligand fragments include those made as a result of chemical or enzymatic hydrolysis or cleavage of the purified ligand.

Other preferred aspects of the present invention are a method for purifying mpl ligand molecules, comprising contacting an mpl ligand starting material containing the mpl ligand molecules with an immobilized receptor polypeptide, in particular with mpl or an mpl fusion polypeptide, under conditions , under which the mpl ligand molecules to be purified are selectively adsorbed onto the immobilized receptor polypeptide, washing the immobilized carrier to remove non-adsorbed material, and eluting the molecule to be purified from the immobilized receptor polypeptide with an elution buffer. The starting material containing the mpl ligand can be a plasma, the immobilized receptor preferably being an mpl-IgG fusion.

Alternatively, the starting material containing the mpl ligand is a recombinant cell culture, the concentration of mpl ligand in the culture medium as well as in the cell lysates being generally higher than in plasma or other natural starting materials. In this case, the mpl-IgG immunoaffinity method described above will still be useful, but is usually not required, and more traditional protein purification methods known in the art. can be applied. Briefly stated, the preferred purification method for providing substantially homogeneous mpl ligands comprises: removing particulate debris, either host cells or lysed fragments, by e.g. Centrifugation or ultrafiltration; if necessary, the protein can be concentrated using a commercially available protein concentration filter; followed by separation of the ligand from other contaminants by one or more steps selected from: immunoaffinity, ion exchange (eg DEAE or materials containing carboxymethyl or sulfopropyl groups), Blue-Sepharose, CM Blue-Sepharose, MONO-Q, MONO-S, Lentil-lectin-Sepharose, WGA-Sepharose, Con A-Sepharose, Ether-Toypearl, Butyl-Toypearl, Phenyl-Toypearl, Protein-A-Sepharose, SDS-PAGE, reverse phase HPLC (e.g. silica gel with attached aliphatic groups) or Sephadex -Molecular sieve or size exclusion chromatography and ethanol or ammonium sulfate precipitation. A protease inhibitor such as methylsulfonyl fluoride (PMSF) can be included in any of the above steps to inhibit proteolysis.

In another preferred embodiment, the present invention provides an isolated antibody that can bind to the mpl ligand. A preferred anti-inpl ligand antibody is monoclonal (Kohier and Milstein, Nature, 256: 495-497 [1975]; Campbell, Laboratory Techniques in Biochemistry and Molecular Biology, Burdon et al., Eds, Volume 13, Elsevier Science Publisrers, Amsterdam [1985]; and Huse et al., Science, 246: 1275-1281 [1989]). A preferred isolated anti-mpl ligand antibody is one that binds to the mpl ligand with an affinity of at least about 106 1 / mol. The antibody binds more preferably with an affinity of at least about 10 l / mol. The most preferred is an antibody against the ιαρί ligand, which has one of the effector functions described above. The isolated antibody that can bind to the rapl ligand can optionally be fused to a second polypeptide, and the antibody or fusion product can be used to isolate and purify mpl ligand from a starting material as described above for Immobilize mpl polypeptide. In a further aspect of this embodiment, the invention provides a method for detecting the mpl ligand in vitro or in vivo, comprising contacting the antibody with an analysis sample, in particular a serum analysis sample, which is assumed to be the Contains ligands, and its detection if binding has occurred.

In yet another preferred embodiment, the invention provides an isolated nucleic acid molecule encoding the mpI ligand or fragments thereof, which nucleic acid molecule may or may not be labeled with a detectable residue, and a nucleic acid molecule provided with a sequence , which is complementary to or under stringent conditions or moderately stringent conditions, hybridizes with a nucleic acid molecule which has the sequence which codes for an mpl ligand. A preferred mpl ligand nucleic acid is RNA or DNA which encodes a biologically active mpl ligand which has at least 75% sequence identity, more preferably at least 80%, more preferably at least 85% and even more preferably 90% and very particularly preferably 95% sequence ides shows the human mpl ligand. More preferred isolated nucleic acid molecules are DNA sequences encoding biologically active mpl ligands selected from: (a) DNA based on the coding region of a mammalian mpl ligand gene (e.g. DNA comprising the nucleotide sequence as shown in Fig. 1 ( SEQ ID NO: 2), or a fragment thereof); (b) DNA which can hybridize to a DNA from (a) under at least moderately stringent conditions; and (c) DNA degenerate from a DNA as defined in (a) or (b), resulting from the degeneration of the genetic code. It is believed here that the new mpl ligands described herein are members of a family of ligands or cytokines that have an appropriate sequence identity so that their DNA matches the DNA of FIG. 1 (SEQ ID NO: 2) ( or the complementary sequence or fragments thereof) can hybridize under low to moderately stringent conditions. Thus, another aspect of the present invention includes DNA that hybridizes to DNA encoding the mpl ligand polypeptides under low to moderately stringent conditions.

In a further preferred embodiment according to the invention, the nucleic acid molecule is a cDNA which encodes the mpl ligand and which further comprises a replicable vector in which the cDNA is operably linked to control sequences which are recognized by a host which transforms with the vector is. This aspect further includes host cells transformed with the vector and a method using the cDNA to effect the production of an mpl ligand comprising expressing the cDNA encoding the mpI ligand in a culture of the transformed Host cells and recovery of the mpl ligand from the host cell culture. The mpl ligand produced in this way is preferably essentially a homogeneous human mpl ligand. A preferred host cell for producing the mpl ligand is Chinese hamster ovary (CHO) cells.

The invention further comprises a preferred method for treating a mammal with an immunological or hematopoietic disorder, particularly thrombocytopenia, comprising administering to the mammal a therapeutically effective amount of an mpl ligand. The mpl ligand is optionally administered in combination with a cytokine, in particular a colony-stimulating factor or interleukin. Preferred colony stimulating factors or interleukins include: kit ligand, LIF, G-CSF, GM-CSF, M-CSF, EPO, IL-1, IL-2, IL-3, IL-5, IL-6, IL -7, IL-8, IL-9 or IL-11. III. production method

Platelet production has long been believed by many authors to be controlled by several lineage-specific humoral factors. It has been postulated that two particular cytokine activities, termed megakaryocyte colony stimulating factor (meg-CSF) and thrombopoietin, regulate megakaryocytopoiesis and thrombopoiesis (Williams et al., J. Cell Physiol., 110: 101-104 [1982 ]; Williams et al., Blood Cells, 15: 123-133 [1989]; and Gordon et al., Blood, 80: 302-307 [, 1992]). According to this hypothesis, meg-CSF stimulates the proliferation of precursor megakaryocytes, while thrombopoietin primarily affects the maturation of differentiated lines and finally the release of platelets. Since the 1960s, the induction and occurrence of meg-CSF and thrombopoietin activities in the plasma, serum and urine of animals and humans, following thrombocytopenia-like episodes, have been well documented (Odell et al., Proc. Soc. Exp. Biol.

Med., 108: 428-431 [1961]; Nakeff et al., Acta Haematol., 54: 340-344 [1975]; Specter, Proc. Soc. Exp. Biol., 108: 146-149 [1961]; Schreiner et al., J. Clin. Invest., 49: 1709-1713 [1970]; Ebbe, Blood, 44: 605-608 [1974]; Hoffman et al., N. Engl. J. Med., 305: 533 [1981]; Straneva et al., Exp. Hematol., 17: 1122-1127 [1988]; Mazur et al., Exp. Hematol., 13: 1164 [1985]; Mazur et al., J. Clin. Invest., 68: 733-741 [1981]; Sheiner et al., Blood, 56: 183-188 [1980]; Hill et al., Exp. Hematol., 20: 354-360 [1922]; and Hegyi et al.,

Int. J. Cell Cloning, 8: 236-244 [1990]). These activities have been reported to be lineage-specific and to differ from known cytokines (Hill RJ et al., Blood 80: 346 [1992]; Erickson-Miller CL et al., Brit. J Haematol., 84: 197-203 [1993]; Straneva JE et al., Exp. Hematol. 20: 4750 [1992]; and Tsukada J. et al., Blood 81: 866-867 [1993]). Attempts to purify meg-CSF or thrombopoietin from thrombocytopenia plasma or urine have so far been unsuccessful. '

In accordance with the observations regarding thrombocytopenia plasma described above, we have found that aplastic porcine plasma (APP) obtained from irradiated pigs stimulates human megakaryocytopoiesis in vitro. We have found that this stimulating activity is suppressed by the soluble extracellular domain of c-mpl, which confirms APP as a possible starting material for the putative mpl ligand (ML). We have now successfully isolated the mpl ligand from APP and the amino acid sequence information has been used to isolate mouse, porcine and human ML cDNA. These MLs have sequence homology to erythropoietin and have meg-CSF and thrombopoietin-like activities. 1. Purification and identification of the mpl ligand from plasma

As stated above, aplastic plasma from a variety of species has been reported to contain activities that stimulate hematopoiesis in vivo, but no hematopoietic stimulating factor isolated from plasma has previously been reported. A starting material for aplastic plasma is that obtained from irradiated pigs.

Aplastic pig plasma (APP) stimulates human hematopoiesis in vitro. In order to determine whether APP contains the mp2 ~ ligand, its effect was examined by measuring the 3H-thymidine incorporation in Ba / F3 cells, transfected with human mpl P (Ba / F3-mp2), according to the processes shown in FIG. APP stimulated 3H-thymidine incorporation in Ba / F3 mpl cells, but not in Ba / F3 control cells (i.e., lines not transfected with human mpl P). Furthermore, such activity was not observed in normal pig plasma. These results indicate that APP contains a factor or factors that produce a proliferation signal via the mpl receptor and which could therefore be the natural ligand for this receptor. This was further confirmed by the finding that treatment of APP with soluble mpl-IgG blocked the stimulating effects of APP on Ba / F3-jnpl cells.

The activity in APP appeared to be a protein since pronase, DTT or 'warmth destroyed the activity in APP (Fig. 3). The activity was also not dialysable. However, the activity was stable at low pH (pH 2.5 for 2 hours) and bound and eluted from several lectin affinity columns, indicating that it is a glycoprotein. In order to further elucidate the structure and identity of this activity, it was affinity-purified from APP using an mpl-IgG chimeras. APP was treated according to your protocol as set out in Examples 1 and 2. Briefly stated, the mpl ligand was purified using hydrophobic interaction chromatography (HIC), immobilized dye chromatography, and mpl affinity chromatography. The recovery of activity from each step is shown in Figure 4 and the purification factor is given in Table 1. The total yield of activity over the mpl affinity column was approximately 10%. The fraction with the activity peak (F6) from the mpl affinity column has an estimated specific activity of 9.8 x 106 units / mg. The total purification from 5 liters of APP was approximately 4 x 106 times (0.8 units / mg to 3.3 x 106 units / mg) with an 83 x 106 times reduction in protein (250 g to 3 pg) . We calculated the specific activity of the ligand, eluted from the mpl-Af f column, as around 3 x 10 units / mg. TABLE 1

Purification of mpl ligand

Protein was determined using the Bradford assay. The protein concentration of the eluted mpl fractions 5-7 was calculated based on the color intensity of a silver-colored SDS gel. One unit is defined as that which causes 50% of the maximum stimulation of the Ba / F3 mpl cell proliferation.

Analysis of the eluted fractions from the mpl-Af f mitate column by SDS-PAGE (4-20%, Novex gel) performed under reducing conditions showed the presence of several proteins (Fig. 5). Proteins that gave the highest intensity after silver staining were resolved with an apparent molecular weight (Mr) of 66000, 55000, 30000, 28000 and 18000-22000. In order to determine which of these proteins stimulated the proliferation of Ba / F3-mpl cell cultures, the proteins were eluted from the gel, as described in Example 2.

The results of this experiment showed that most activity was eluted from a gel strip containing Mr 28000-32000 protein, with less activity eluting from the 18000-22000 region of the gel (Fig. 6). The only visible proteins in these regions had Mr from 30000, 28000 and 18000 to 22000. To identify and obtain the protein sequence of these proteins that were separated in this area of the gel (i.e., bands at 30, 28 and 18-22 kDa), these three proteins were transferred to PVDF by an electroblot method and sequenced as described in Example 3. The amino terminal sequences obtained are shown in Table 2. TABLE 2

Amino-terminal sequences of the mpl ligand

Computer aided analysis showed that these amino acid sequences were new. Since all three sequences were the same, it was assumed that the 30 kDa, 28 kDa and 18-22 kDa proteins were related and may represent different formats of the same protein. Furthermore, this protein or proteins was a probable candidate for the natural mpl ligand, since the activity in the same region (28000-32000) was separated on a 4-20% gel of the SDS-PAGE. Furthermore, the partially purified ligand migrated with a Mr of 17,000 to 30,000 when subjected to gel filtrate ion chromatography using a Superose 12 (Pharmacia) column. The various Mr forms of the ligand are believed to be the result of proteolysis or glvcosylation differences or other post- or pre-translational modifications.

As described earlier, human antisense mpl RNA suppressed megakaryocytopoiesis in human bone marrow cultures that were enriched with CD 34 + progenitor cells without affecting the differentiation from other hematopoietic lineages (Methia et al., See above). This result suggests that the mpl receptor could play a role in the differentiation and proliferation of mega-karyocytes in vitro. To further elucidate the role of the mpl ligand in megakaryocytopoiesis, the effects of APP and APP depleted on the mpl ligand on human in vitro megakaryocytopoiesis were compared. The effect of APP on human megakaryocytopoiesis was determined using a modification of the liquid suspension megakaryocytopoiesis assay described in Example 4. In this assay, human peripheral stem cells (PSC) were treated with APP before and after mpl-IgG affinity chromatography. The GPIIbIIIa-stl-emulation of the megakaryocytopoiesis was quantified with a 125J anti-IIbIIIa antibody (Fig. 7). Figure 7 shows that 10% APP elicited approximately three-fold stimulation while APP depleted of the mpl ligand had no effect. Significantly, the mpl ligand-depleted APP did not induce the proliferation of the Ba / F3 mpl cells.

In a further experiment, soluble human mpl IgG, added on days 0, 2 and 4, to kuituren containing 10% APP neutralized the stimulating effects of APP on human megakaryocytopoiesis (Fig. 8). These results indicate that the mpl ligand has a role in regulating human megakaryocytopoiesis and, therefore, may be useful in the treatment of thrombocytopenia. 2. Molecular cloning of the mpl ligand

Based on the amino terminal amino acid sequences obtained from the 30 kDa, 28 kDa and 18-22 kDa proteins (see Table 2 above), two degenerate oligonucleotide primer pools were designed and used to replicate the genomic pigs -DNA by PCR. It was assumed that if the amino terminal amino acid sequence is encoded by a single exon, then the correct PCR product is expected to be 69 base pairs long. A DNA fragment of this size was found and subcloned into pGEMT. The sequences of the oligonucleotide PCR primers and the three clones obtained are given in Example 5. The amino acid sequence (PRLLNKLLR [SEQ ID NO: 33]) of the peptide encoded between the PCR primers was identical to the sequence obtained by amino terminal protein sequencing of the pig ligand (see residues 9-17 for 28- and 30 kDa pig protein sequences above).

A synthetic oligonucleotide based on the sequence of the PCR fragment was used to screen a human genomic DNA library. A 45-mer oligonucleotide, designated pR45, was designed and synthesized based on the sequence of the PCR fragment. This oligonucleotide had the following sequence:

TCG 3 '(SEQ ID NO: 34)

The deoxyoligonucleotide was used to screen a human genomic DNA library in Xgeml2 under low stringent hybridization and washing conditions, according to Example 6. Positive clones were removed, plaque-purified and analyzed by restriction mapping and Southern blotting. A 390 bp EcoRI-Xbal fragment that hybridized to the 45-mer was subcloned into pBluescript SK-. DNA sequencing of this clone confirmed that DNA encoding the human homologue to the porcine mpl ligand had been isolated. The human DNA sequence and the deduced amino acid sequence are shown in Fig. 9 (SEQ ID NO: 3 & 4). The predicted positions of the introns in the genomic sequence are also indicated by arrows and they define a suspected exon ("exon 3").

Based on the human "exon 3" sequence (example 6), oligonucleotides were synthesized which correspond to the 3 'and 5' ends of the exon sequence. These two primers were used in PCR reactions using a template cDNA made from different human tissues. The expected size of the correct PCR product was 140 bp. After analysis of the PCR products on a 12% polyacrylamide gel, a DNA fragment of the expected size was detected in cDNA libraries made from adult human kidney, Fötaien 293 kidney cells and cDNA made from human fetal liver.

A fetal liver cDNA library (7 x 106 clones) in Lambda-DR2 was next screened with the same 45-mer oligonucleotide used to screen the human genomic library, and the fetal liver cDNA library was low stringent hybridization conditions searched. Positive clones were removed, plaque cleaned, and the insert size was determined by PCR. A clone with a 1.8 kb insert was selected for further analysis. Using the procedure described in Example 7, the nucleotide and inferred amino acid sequences of the human mpl ligand (hML) were obtained. These sequences are shown in Fig. 1 (SEQ ID NO: 1 & 2). 3. structure of the human mpl ligand (hML)

The human mpl ligand (hML) cDNA sequence (Fig. 1 [SEQ ID NO; 2]) comprises 1774 nucleotides followed by a poly (A) tail. It contains 215 nucleotides of 5'-untranslated sequence and a 3'-untranslated region of 498 nucleotides. The assumed start codon at nucleotide position (216-218) lies within a consensus sequence that is preferred for the initiation of eukaryotic translation. The open reading frame is 1059 nucleotides long and codes for a 353 amino acid residue-long polypeptide, starting at nucleotide position 220. The N-terminus of the predicted amino acid sequence is highly hydrophobic and presumably corresponds to a signal peptide. Computer analysis of the predicted amino acid sequence (by Heijne et al., Eur. J. Biochem., 133: 17-21 [1983]) indicates a potential cleavage site for signal peptidase between residues 21 and 22. Cleavage at this position would produce a mature polypeptide of 332 amino acid residues starting with the amino terminal sequence obtained from the mpl ligand purified from porcine plasma. The predicted non-glycosylated molecular weight of the 332 amino acid residue ligand is approximately 38 kDa. There are 6 potential N-glycosylation sites and 4 cysteine residues.

A comparison of the mpl ligand sequences with the sequence database Genbank showed 23% identity between the amino terminal 153 residues of mature human mpl ligand and human erythropoietin (FIG. 10 [SEQ ID NO: 6 & 7]). If conservative substitutions are taken into account, this region of hML shows 50% similarity to human erythropoietin (hEPO). hEPO and the hML contain four cysteines. Three of the four cysteines are conserved in hML, including the first and last cysteine. Sequence-specific (site-directed) mutagenesis experiments have shown that the first and last cysteine of erythropoietin form a disulfide bond that is required for their function (Wang, F.F. et al., Endocrinology 116: 2286-2292 [1983]). Analogously, the first and last cysteine from hML can also form a critical disulfide bond. None of the glycosylation sites are conserved in hML. All potential N-dependent glycosylation sites in hML are located in the carboxy-terminal half of the hML polypeptide. Similar to hEPO, the hML mRNA does not contain the Konsen-sus polyadenylation sequence AAUAAA, nor does it contain the regulatory element AUUUA, which is present in the 3'-untranslated region of many cytokines and which is believed to be that it affects mRNA stability (Shaw et al., Cell, 46: 659-667 [1986]). Northern blot analysis shows low levels of a single 1.8 kb hML RNA transcript in fetal and adult liver. After prolonged exposure, a weaker band of the same size was found in the adult kidney. In comparison, human erythropoietin is expressed in fetal liver and, in response to hypoxia, in the adult kidney and liver (Jacobs et al., Nature, 313: 804-809 [1985] and Bondurant et al., Molec. Cell. Biol., 6: 2731-2733 [1986]).

The importance of the C-terminal region of the hML is still to be clarified. Based on the presence of the six potential sites for N-dependent glycosylation and the ability of the ligand to bind to lectin affinity columns, this region of the hML is presumably glycosylated. In some gel elution experiments we observed an activity which was separated to a Mr of about 60,000, which can be the 'complete glycosylated molecule. The C-terminal region can therefore serve to stabilize and extend the half-life of the circulating hML. In the case of erythropoietin, the non-glycosylated form has full in vitro biological activity, but it has significantly reduced plasma half-life compared to glycosylated erythropoietin (Takeuchi et al., J. Biol. Chem., 265: 12127-12130 [ 1990]; Narhi et al., J. Biol. Chem., 266: 23022-23026 [1991] and Spivack et al., Blood, 7: 90-99 [1986]). The C-terminal domain of hML contains two dibasic amino acid sequences (Arg-Arg motifs at positions 153-154 and 245-246) that can serve as potential processing sites. The cleavage at these sites can be responsible for the generation of the 30, 28 and 18-22 kDa forms of the ML isolated from APP. Significantly, the Argi53-Ârgi54 sequence immediately follows the erythropoietin-like domain of the ML. These observations indicate that the complete ML can be a precursor protein that undergoes limited proteolysis to form the mature ligand. 4. Isoforms and variants of the human mpl ligand

Isoforms and alternatively spliced forms of human mpl ligand were detected by PCR in human adult liver. In short, primers were synthesized that corresponded to each end as well as selected internal regions of the coding sequence of hML. These primers were used in RT-PCR to amplify human adult liver RNA as described in Example 10. In addition to the full form, designated hML, three other forms, designated hML2, hML3 and hML4, were observed or inferred. The mature inferred amino acid sequence of all four isoforms is shown in Figure 11 (SEQ ID NO: 6, 8, 9 & 10). hML3 has a 116 nucleotide deletion at position 700, which leads to both an amino acid deletion and a raster shift. The cDNA now encodes a mature polypeptide that is 265 amino acids long and in which the amino acid residue 139 diverges from the hML sequence. Finally, hML4 has a 12 nucleotide deletion following nucleotide position 618 (also found in the mouse and pig sequences (see below)) and has the 116 bp deletion found in hML3. Although no clones with only the 12 bp deletion (hereinafter nucleotide 619) have been isolated in humans (designated hML2), this form probably exists because such isoforms have been identified in the mouse and in the pig (see below), and because it has been identified in connection with the 116 nucleotide deletion in hML4.

A substitution variant of hML in which the dibasic Argi53_Ar <? L54 ~ Se (3uence had been replaced by two alanine residues, and a MEPO domain "truncated form of hML were constructed to determine whether the complete ML was required. The substituted dibasic Arg ^ -Arg ^ sequence variant, designated hML (R153A, R154A), was constructed using PCR, as described in Example 10. The 'ΈΡΟ-domain' '- shortened form , hMLi53, was also produced by means of PCR by introducing a stop codon following Argl53 5. Expression of recombinant human mpl ligand (rhML) in transiently transfected human embryonic kidney (293) cells

To confirm that the cloned human cDNA encodes a ligand for mpl, the ligand was expressed in 293 mammalian cells under the. Control of the cytomegalovirus immediate early promoter using the expression vectors pRK5-hML or pRK5-hML153. Supernatants from the transiently transfected human embryonic kidney 239 cells were found to stimulate 3H-thymidine incorporation in Ba / F3-mp2 cells but not in the parent Ba / F3 cells (Fig. 12A). Media from the 293 cells transfected only with the pRK vector alone did not show this activity. Addition of mpl-IgG to the media destroyed the stimulation (data not shown). These results show that the cloned cDNA encodes a functional human ML (hML).

To determine whether the "EPO domain" could bind to and activate mpl alone, the truncated form of hML, rhML153, was expressed in 293 cells. Supernatants from transfected lines were found to have similar activity to that found in supernatants from lines expressing the full hML (Figure 12A), indicating that the C-terminal domain of ML is not required is for the binding and activation of c-mpl. 6. The mpl ligand stimulates megakaryocytopoiesis and thrombopoiesis

The complete rhML and shortened rhML153 form of recombined hML stimulated human megakaryocytopoiesis in vitro (Fig. 12B). This effect has been observed in the presence of other exogenously added hematopoietic

Growth factors. With the exception of IL-3, ML was the only hamatopoietic growth factor tested that showed this activity. IL-11, IL-6, IL-1, erythropoietin, G-CSF, IL-9, LIF, kit ligand (KL), M-CSF, OSM and GM-CSF had no effect on megakaryocytopoiesis when in our assay was tested separately (data not shown). This result shows that ML has megakaryocyte stimulating activity and shows the role of ML in regulating megakaryocytopoiesis.

Thrombopoietic activities present in the plasma of thrombocytopenia animals have been shown to stimulate platelet production in a mouse rebound thrombocytosis assay (McDonald, Proc. Soc. 'Exp. Biol. Med., 14: 1006-1001 [1973] and McDonald et al., Scand. J. Haematol., 16: 326-334 [1976). In this model, acute thrombocytopenia is induced in mice using a specific anti-platelet serum, which leads to predictable rebound thrombocytosis. Such immunothrombocythemic mice respond better to exogenous thrombopoietin-like activities than normal mice (McDonald, Proc. Soc. Exp. Biol. Med., 14: 1006-1001 [1973]), just as exhypoxic mice are more sensitive to erythropoietin than normal Mice (McDonald et al., J. Lab. Clin. Med., 77: 134-143 [1971]). To determine whether rML stimulates platelet production in vivo, mice with rebound thrombocytosis were injected with partially purified rhML. The number of platelets and the incorporation of 35S into platelets was then quantified. Injection of mice with 64,000 and 32,000 units of rML significantly increased platelet production, as evidenced by an approximately 20% increase in the number of platelets (p = 0.0005 and 0.0001) and an approximately 40% increase in 35S incorporation in platelets (p = 0.003) in the treated mice compared to the control mice which were injected with the auxiliary alone (FIG. 12C). This level of stimulation is comparable to the result we observed with IL-6 in this model (data not shown). Treatment with 16,000 units of rML showed no significant stimulation of platelet production.

These results show that ML stimulates platelet production in a dose dependent manner and therefore has thrombo-poietin-like activity. 293 cells were also transfected with the other hML isoform constructs described above and the supernatants were tested using the Ba / F3-mp2 proliferation assay (see Figure 13). hML2 and hML3 showed no detectable activity in this assay, but the activity of hML (R153A, R154A) was similar to hML and hML153, showing that possession of the Argi53-Argi54 dibasic site is neither necessary nor necessary for the activity is harmful. 7. Megakaryocytopoiesis and the mpl ligand

Megakaryocytopoiesis has been suggested to be regulated at a variety of cellular levels (Williams et al., J. Cell Physiol., 110: 101-104 [1982] and Williams et al., Blood Cells, 15: 123-133 [1989]). This is largely based on the observation that certain hematopoietic growth factors stimulate the proliferation of megakaryocyte precursors, while others appear to primarily affect maturation. The results presented here suggest that ML acts as both a proliferative and a maturation factor. The finding that ML stimulates the proliferation of megakaryocyte precursors is supported by several clues. First, APP stimulates the proliferation as well as the maturation of human megakaryocytes in vitro, and this stimulation is completely inhibited by mpl-IqG · (Figures 7 and 8). Furthermore, the inhibition of megakaryocyte colony formation by c-mpl antisense oligonucleotides (Methia et al., Blood, 82: 1395-1401 [1993]) and the finding that c-mpl can produce a proliferative signal in lines also show into which it is transfected (Skoda et al., EMBO, 12: 2645-2653 [1993] and Vigon et al., Oncogene, 8: 2607-2615 [1993]) that ML stimulates proliferation. The apparent expression of c-mpl during all phases of megakaryocyte differentiation (Methia et al., Blood, 82: 1395-1401 [1993]) and the ability of recombinant ML to rapidly stimulate platelet production in vivo show that ML also Ripening affects. The availability of recombinant ML makes it possible to investigate carefully its role in regulating megakaryocytopoiesis and thrombopoiesis as well as its potential to influence other hematopoietic lineages. 8. Isolation of the gene for the human mpl ligand (TPO)

Human genomic DNA clones of the TPO gene were isolated by screening a human genetic library in X-Geml2 with pR45 under low stringency conditions or under highly stringent conditions with a fragment corresponding to the 3'-half of the human cDNA required for the encoded human mpl ligand. Two 35 kb overlapping lambda clones were isolated. Two overlapping fragments (BamHI and EcoRI) containing the complete TPO gene were subcloned and sequenced (Fig. 14A, 14B and 14C).

The structure of the human gene consists of six exons within 7 kb of genomic DNA. The boundaries of all exon / intron connections agree with the consensual motif that has been established for infant genes (Shapiro, M.B. et al., Nucl. Acids Res. 15: 7155 [1987]).

Exon 1 and exon 2 contain 5 'untranslated sequence and the initial four amino acids of the signal peptide. The rest of the secretory signal and the first 26 amino acids of the mature protein are encoded by exon 3. The full carboxyl domain and the 31 non-translated

The region as well as around 50 amino acids of the erythropoietin-like domain are encoded by exon 6. The four amino acids involved in the deletion observed in hML-2 (hTPO-2) are encoded by the 5 'end of exon 6.

Analysis of Huiaan genomic DNA by a Southern blot showed that the gene for TPO is in a single copy. The chromosomal location of the gene was determined by fluorescent in situ hybridization (FISH), which was assigned to chromosome 3q27-28. 9. Expression and purification of TPO from 293 cells

The preparation and purification of ML or TPO from 293 cells is described in detail in Example 19. Briefly stated, cDNA, which corresponds to the complete open reading frame of TPO, was obtained by PCR using pRK5-hmp! I. The PCR product was purified and cloned between the restriction sites Clal and Xbal of the plasmid pRKStkneo (a vector derived from pRK5, which was modified to express a neomycin resistance gene under the control of the thymidine kinase promoter), to the vector pRK5tkneo.ORF ( a vector that encodes the full open reading frame).

A second vector encoding the EPO homologous domain was generated in the same way, but using different PCR primers to obtain the final construct, called pRK5-tkneoEPO-D.

These two constructs were transfected in human embryonic kidney cells by the calcium phosphate method, and neomycin-resistant clones were selected and cultured to confluent density. Expression of ML153 or ML332 in the conditioned media of these clones was determined using the Ba / F3-mpi proliferation assay.

The purification of rhML332 was carried out as described in Example 19. Briefly stated, 293-rhML332 conditioned media were placed on a Blue Sepharose (Pharmacia) column, which was then washed with a buffer containing 2M urea. The column was eluted with a buffer containing 2M urea and 1M NaCl. The Blue Sepha rose elution pool was then placed directly on a WGA Sepharose column, washed with 10 column volumes of buffer containing 2M urea and 1M NaCl, and eluted with the same buffer containing 0.5M N-acetyl-D -glucosamine. The WGA-Sepharose eluate was applied to a C4-HPLC column (Synchrom,

Ine.) And eluted with a discontinuous propanol gradient. In SDS-PAGE, the purified 293-rhML332 migrates as a broad band in the 68-80 kDa range of the gel (see Fig. 15).

Purification of rhML153 was also performed as described in Example 19. Briefly, 293-rhML153-conditioned media were separated on Blue-Sepharose as described for rhML332. The Blue-Sepharose eluate was applied directly to an mpI affinity column given as described above. RhML153 eluted from the mpl-hffinity column was purified to homogeneity using a C4 HPLC column run under the same conditions as used for rhML332. In SDS-PAGE the purified rhML153 separates into two main and two minor bands with Mr of around 18000-22000 (see Fig. 15). 10. The mouse mpl ligand

A DNA fragment corresponding to the coding region of the human mpI ligand was obtained by PCR, gel-purified and labeled in the presence of P-dATP and P-dCTP. This sample was used to screen 106 clones of a mouse liver cDNA library in λΰΤΙΟ. A mouse clone (Fig. 16 [SEQ ID NO: 12 & 13)) containing a 1443 base pair insert was isolated and sequenced. The suspected start codon at nucleotide position 138-141 was within a consensus sequence that is advantageous for eukaryotic translation initiation (Kozak, M., J. Cell Biol., 108: 229-241 [1989]). This sequence defines an open reading frame of 1056 nucleotides that predicts a primary translation product of 352 amino acids. This open reading frame is flanked by 137 nucleotides 5'- and 247 nucleotides of a 31 non-translatated sequence. There is no poly (A) tail following the 3 'untranslated region, indicating that the clone is probably not complete. The N-terminus of the predicted amino acid sequence is highly hydrophobic and presumably represents a signal peptide. Computer analysis (by Heijne, G., Eur. J. Biochem. 133: 17-21 (1983)) shows a potential cleavage point for the signals -peptidase between residues 21 and 22. Cleavage at this position would result in a mature 331 amino acid (35 kDa) polypeptide which is identified as mML33i (or mML2, for the reasons given below). The sequence contains four cysteines, all of which are conserved in the human sequence, and seven potential N-glycosylation sites, five of which are conserved in the human sequence. Again, as with hML, the seven potential N-glycosylation sites are localized in the C-terminal half of the protein.

When compared to the human ML, considerable identity for the nucleotide and inferred amino acid sequences is observed in the "EPO domains" of these MLs. However, when the inferred amino acid sequences of human and mouse ML's are aligned, the mouse sequence appears to carry a tetrapeptide deletion between residues 111-114 that corresponds to the 12 nucleotide deletion that follows nucleotide position 618 as observed in human (see above) and the porcine (see below) cDNA. As a result, further clones were examined to identify possible mouse ML isoforms. A clone encoded a deduced 335 amino acid polypeptide sequence containing the "missing" tetrapeptide LPLQ. This form is believed to represent the complete ML of the mouse and is referred to as mML or 111ML335. The nucleotide and inferred araino acid sequence for mML are given in Figure 17 (SEQ ID NO: 14 & 15) . These cDNA clones consist of 1443 base pairs followed by a poly (A) tail. It has an open reading frame of 1068 bp, flanked by 134 bases of a 5f and 241 bases of a 3 'untranslated sequence. The suspected start codon is at nucleotide position 138-140. The open reading frame encodes a predicted protein of 356 amino acids, the first 21 of which are highly hydrophobic and probably act as a secretion signal.

Finally, a third mouse clone was isolated, sequenced and found to contain the 116 nucleotide deletion corresponding to hML3. This mouse isoform is therefore called mML3. A comparison of the inferred amino acid sequences of these two isoforms is shown in Fig. 18 (SEQ ID NO: 9 & 16).

The total amino acid sequence identity between human and mouse ML (Fig. 19 [SEQ ID NO: 6 & 17]) is 72%, but this homology is not evenly distributed. The region referred to as the "EPO domain" (amino acids 1-153 for the human sequence and 1-149 for the mouse) is more strongly conserved (86% homology) than the carboxy-terminal region of the protein (62% homology). This may further suggest that only the "EPO domain" is important for the biological activity of the protein. Interestingly, of the two dibasic amino acid motifs found in hML, only the dibasic motif that immediately follows the "EPO-Do-mâne" (residue position 153-154) in the human sequence is present in the mouse sequence. This is consistent with the possibility that the complete ML can be a precursor protein that undergoes limited proteolysis to generate the mature ligand. Alternatively, proteolysis between Argi53-Argi54 can facilitate hML clearance.

An expression vector containing the complete coding sequence for mML was transiently transfected into 293 cells as described in Example 1. Conditioned media from these lines stimulated 3H-thymidine incorporation into Ba / F3 cells that either the mouse or expressed human mpl, but they had no effect on the parent (mpl-free) cell line. This shows that the cloned mouse ML cDNA encodes a functional ligand capable of activating both the mouse and human ML receptors (mpl). 11. The mpl ligand to the pig

Porcine ML (pML) cDNA was isolated by RACE-PCR as described in Example 13. A 1342 bp PCR cDNA product was found in kidney and subcloned. Several clones were sequenced and were found to encode a porcine mpI ligand of 332 amino acid residues, designated pML (or pML332), which encoded the nucleotide and deduced amino acid sequence as shown in Figure 20 (SEQ ID NO: 18 & 19).

Again, a second form, designated pML2, encoding a protein with a 4 amino acid deletion (228 amino acid residues) was identified (see Figure 21 [SEQ ID NO. 21]). A comparison of the pML and pML2 amino acid sequences shows that the latter form is identical, with the exception that the tetrapeptide QLPP, corresponding to residues 111-114 inclusive, has been deleted, four amino acid long deletions (see Fig. 22 [ SEQ ID NO: 18 & 21]).

The four amino acid deletions observed in mouse and porcine ML cDNA occur in exactly the same position within the predicted protein.

A comparison of the predicted amino acid sequences of the mature human ML form. Mouse and pig (Fig. 19 [SEQ ID NO: 6, 17 & 18]) shows that the overall sequence identity is 72% between mouse and human, 68% between mouse and pig and 73% between pig and human. The homology is considerably larger in the amino terminal half of the ML (the EPO homologous domain). This domain is 80-84% identical between two types, whereas the carboxy-terminal half (carbohydrate domain) is only 57 to 67% identical. A dibasic amino acid motif, which could represent a protease cleavage site, is present at the carboxy end of the erythropoietin homology domain. This motif is preserved between the three species at this position (Fig. 19 [SEQ ID NO: 6, 17 & 18]). A second dibasic site present at positions 245 and 246 in the human sequence is absent in the mouse or pig sequence. The mouse and porcine ML sequences contain 4 cysteines, all of which are conserved in the human sequence. There are seven potential N-glycosylation sites within the mouse ligand and six within the porcine ML, five of which are conserved within the human sequence. Again, all potential N-glycosylation sites are present in the C-terminal half of the protein. 12. Expression and Purification of TPO from Lines of the Chinese Hamster (CHO)

The expression vectors used to transfect the CHO cells are designated as follows: pSV15.ID.LL.MLORF (complete or TPO332), and pSV15. ID.LL.MLEPO-D (shortened or TPOX53). The essential features of these plasmids are shown in FIGS. 23 and 24.

The transfection procedures are described in Example 20. Briefly, cDNA corresponding to the full open reading frame of TPO was obtained by PCR. The PCR product was purified and cloned between the restriction sites (Clal and Sali) of the plasmid pSV15.ID.LL to obtain the vector pSV15.ID.LL.ML0RF.

A second construct, corresponding to the EPO homologous domain, was generated in the same way, but using a different reverse primer (EPOD.Sal). The final construct for the vector encoding the EPO homologous domain of TPO becomes pSVl5. ID.LL.MLEPO-D.

These two constructs were linearized with Notl and transfected into Chinese hamster ovary cells (CHO-DP12 lines, EP 307,247, published March 15, 1989) by electroporation. 107 lines were electroporation in a BRL electroporation apparatus (350 volts, 330 mF, low capacitance) in the presence of 10, 25 or 50 mg of DNA as described (Andreason, GL, J. Tissue Cult. Meth. 15 , 56 [1993]). On the day following the transfection, the lines were divided into DHFR-selective media (high-glucose-DMEM-F12 50:50 without glycine, 2 mM glutamine, 2-5% dialyzed fetal calf serum). 10 to 15 days later, individual colonies were transferred to 96-well plates and cultivated to confluence. The expression of ML153 or ML332 in the conditioned media of these clones was measured using the Ba / F3-mpl proliferation assay (described in Example 1).

The procedure for cleaning and isolating TPO from harvested CHO cell culture fluid is described in Example 20. Briefly, harvested cell culture fluid (HCCF) is placed on a Blue Sepharose column (Pharmacia) at a ratio of approximately 100 liters of HCCF per liter of resin. The column is then washed with 3 to 5 times the column volume of buffer, followed by 3 to 5 column volumes of a buffer containing 2.0 M urea. TPO is then eluted with 3 to 5 column volumes of buffer containing 2.0M urea and 1.0M NaCl.

The eluted Blue Sepharose pool containing TPO is then applied to a wheat germ lectin Sepharose column (Pharmacia), equilibrated in the Blue Sepharose elution buffer, in a ratio of 8 to 16 ml of Blue Sepharose eluate per ml of resin given. The column is then washed with 2 to 3 column volumes of equilibration buffer. TPO is then eluted with 2 to 5 column volumes of a buffer containing 2.0M urea and 0.5M N-acetyl-D-glucosamine.

The wheat germ lectin eluate containing TPO is then acidified and C ^ It is added to a final concentration of 0.04%. The pool obtained is placed on a C4 reverse phase column, equilibrated in 0.1% TFA, 0.04% C ^ Eg, with a loading of approximately 0.2 to 0.5 mg protein per ml resin.

The protein is eluted with a two-phase linear gradient from acetonitrile containing 0.1% TFA and 0.04%

Ci2Eg, and a pool is created based on SDS-PAGE.

The C4 pool is then diluted and diafiltered against approximately 6 volumes of buffer on an Araicon YM or similar ultrafiltration membrane with a molecular weight cut-off at 10,000 to 30,000. The diafiltrate obtained can then be directly processed further or further concentrated by ultrafiltration. The dia-filtrate / concentrate is usually adjusted to a final concentration of 0.01% Tween-80.

All or part of the diafiltrate / concentrate, which is equivalent to 2 to 5% of the calculated column volume, is then placed on a Sephacryl S-300 HR column (Pharmacia), which is equilibrated in a buffer containing 0.01 % Tween-80, and was then chromatographed. The TPO-containing fractions, which are free of aggregates and proteolytic degradation products, are then combined on the basis of SDS-PAGE. The pool obtained is filtered and stored at 2-8 ° C.

13. Process for transforming and inducing TPO synthesis in a microorganism and isolating, cleaning and refolding the TPO produced therein

The construction of the E. coli TPO expression vectors is described in detail in Example 21. Briefly stated, plasmids pMP21, pMP151, pMP41, pMP57 and pMP202 were all designed to express the first 155 amino acids of TPO downstream of a small leader. that varies between different constructs. The leaders primarily serve a high level of translation initiation and quick cleaning. The plasmids pMP210-1, -T8, -21, -22, -24, -25 are designed to express the first 153 amino acids of the TPO downstream of an initiation methionine and differ only in the codon Usage for the first six amino acids of the TPO, while the plasmid pMP251 is a derivative of pMP210-l, in which the carboxy-terminal end of TPO is extended by two amino acids. All of the above plasmids produce high levels of intracellular expression of TPO in E. coli by induction of the tryptophan promoter (Yansura, DG et al., Methods in Enzymology (Goeddel, DV, ed.) 185: 54-60, Academic Press, San Diego [1990]). The plasmids pMPl and pMP172 are intermediate products in the construction of the above plasmids for intracellular TPO expression.

The above TPO expression plasmids were used to transform E. Coli using the CaCl2-Uitzeschock method (Mandei, M. et al., J. Mol. Biol., 53: 159-162, [1970]) and other methods as described in Example 21. Briefly stated, the transformed lines were first cultivated at 37 ° C until the optical density (600 nm) of the culture reached approximately 2-3. The culture was then diluted and acid was added after cultivation with aeration. The culture was then cultured with aeration for an additional 15 hours, after which time the rows were harvested by centrifugation.

The isolation, purification, and refolding procedures below for the production of biologically active, refolded huraane TPO or fragments thereof are described in Examples 22 and 23 and can be used to obtain any TPO variant, including finally N- and C-terminal elongated forms. Other suitable methods for refolding recombinant or synthetic TPO can also be found in the following patents: Builder et al., U.S. Patent 4,511,502, Jones et al., U.S. Patent 4,512,922, Olson U.S. Patent 4,518,526 and Builder et al ., U.S. Patent 4,620,948; for a general description of the recovery and refolding procedures for a variety of recombinant proteins expressed in an insoluble form in E. coli.

A. Obtaining Insoluble TPO

A microorganism, such as E. coli, which expresses TPO encoded by any plasmid, is fermented under conditions under which TPO is deposited as insoluble "retractile bodies". If necessary, the rows are first washed in a buffer which breaks the cell. Typically, about 100 g lines are resuspended in approximately 10 volumes of a line breaking buffer (e.g. 10mM Tris, 5mM EDTA, pH 8) with e.g. a Polytron homogenizer and the lines are centrifuged at 5000 x g for 30 min. The lines are then lysed using conventional techniques such as tonic shock, sonication, pressure cycling, chemical or enzymatic methods. For example, the above washed cell pellet can be resuspended in a further 10 volumes of the row-breaking buffer with a homogenizer, and the cell suspension is reduced by an LH-Cell Disrupter (LH Inceltech, Ine.) Or by a microfluidizer ( Microfluidics Interntatio-nal) according to the manufacturer's instructions. The particulate material containing TPO is then separated from the liquid phase and optionally washed with a suitable liquid. For example, a suspension of the cell lysate can be centrifuged at 5000 x g for 30 min, resuspended and optionally centrifuged a second time to give a washed pellet of refracting bodies. The washed pellet can be used immediately or, if necessary, stored frozen (e.g. at -70 ° C).

B. Solubilization and purification of monomeric TPO

Insoluble TPO in the pellet of the breaking body is then solubilized with a solubilization buffer. The solubilization buffer contains a chaotropic agent and is usually buffered at a basic pH and contains a reducing agent to improve the yield of monomeric TPO. Representative chaotropic agents include urea, guanidine HCl and sodium thiocyanate. A preferred chaotropic agent is guanidine HCl. The concentration of the chaotropic agent is usually 4-9M, preferably 6-8M. The pH of the solubilization buffer is maintained at a suitable pH in a pH range of approximately 7.5-9.5, preferably 8.0-9.0 and most preferably 8.0. Preferably, the solubilization buffer also contains a reducing agent to help form the monomeric form of TPO. Suitable reducing agents include organic compounds containing a free thiol group (RSH). Representative reducing agents include dithiothreitol (DTT), dithioerythritol (DTE), mercaptoethanol, glutathione (GSH), cysteamine and cysteine. A preferred reducing agent is dithiothreitol (DTT). Optionally, the solubilization buffer may contain a mild oxidizing agent (e.g. molecular oxygen) and a sulfite salt to form monomeric TPO via sulfitolysis. In this embodiment, the TPO-S sulfonate obtained is later folded back in the presence of the redox buffer (e.g. GSH / GSSG) to give the suitably folded TPO.

The TPO protein is usually further purified using, for example, centrifugation, gel filtration chromatography and reverse phase column chromatography.

To illustrate, the following procedure achieved suitable yields of monomeric TPO. The breaking body pellet is resuspended in approximately 5 volumes by weight in the solubilization buffer (20mM Tris, pH8, with 6-8M guanidine and 25mM DTT) and stirred for 1-3 hours or overnight at 4 ° C to achieve solubilization of the TPO protein. High concentrations of urea (6-8M) are also useful, but they usually result in a somewhat lower yield compared to guanidine. After solubilization, the solution is centrifuged at 30,000 x g for 30 min to produce a clear supernatant containing denatured, monomeric TPO protein. The supernatant is then chromatographed on a Superdex 200 gel filtration column (Pharmacia, 2.6 x 60 cm) at a flow rate of 2 ml / min and the protein is eluted with 20 mM sodium phosphate, pH 6.0 10mM DTT. Fractions with monomeric denatured TPO protein, which elute between 160 and 200 ml, are combined. The TPO protein is further purified on a semi-preparative C4 reverse phase column (2 x 20 cm VYDAC). The analysis sample is placed at 5 ml / min on a column which is equilibrated in 0.1% TFA (trifluoroacetic acid) with 30% acetonitrile. The protein is eluted with a linear gradient from acetonitrile (30-60% in 60 min). The purified reduced protein elutes at approximately 50% acetonitrile. This material is used for folding back in order to obtain a biologically active TPO variant. C. Folding back TPO creates the biologically active form

Following the solubilization and further purification of TPO, the biologically active form is obtained by refolding the denatured monomeric TPO form in a redox buffer. Because of the high potency of TPO (the half maximum stimulation in the Ba / F3 assay is achieved at around 3 pg / ml), it is possible to obtain biologically active material using many different buffers, detergents and redox conditions. However, only a small amount of properly folded material (<10%) is obtained under most conditions. For commercial manufacturing processes, it is desirable to obtain refolding yields of at least 10%, more preferably 30-50%, and most preferably> 50%. Many different detergents, including Triton X-100, dodecyl beta maltoside, CHAPS, CHAPSO, SDS, Sarkosyl, Tween 20 and Tween 80, Zwittergent 3-14, and others are suitable for making at least some properly folded material. Of these, however, the most preferred detergents were those of the CHAPS family (CHAPS and CHAPSO) which were found to work best in the refolding reaction and to limit protein aggregation and inappropriate disulfide formation. CHAPS levels in excess of approximately 1% were particularly preferred. Sodium chloride was required for the best yields, with optimal levels between 0.1M and 0.5M. The presence of EDTA (1-5 mM) in the redox buffer was preferred to limit the amount of metal-catalyzed oxidation (and aggregation) that was observed with some preparations. Gly-cerol concentrations of more than 15% created the optimal refolding conditions. For maximum yields, it was necessary to have a redox pair in your redox buffer, consisting of an oxidized and reduced organic thiol (RSH). Suitable redox pairs include mercaptoethanol, glutathione (GSH), cysteamine, cysteine and their corresponding oxidized forms. Preferred redox pairs were glutathione (GSH): oxidized glutathione (GSSG) or cysteine: cystine. The most preferred redox couple was glutathione (GSH): oxidized glutathione (GSSG). In general, higher yields were observed when the molar ratio of the oxidized component of the redox couple was equal to or greater than that of the reduced component of the redox couple. pH values between 7.5 and approximately 9 were optimal for the refolding of these TPO variants. Organic solvents (e.g. ethanol, acetonitrile, methanol) were tolerated at concentrations of 10-15% or less. High levels of organic solvents increased the amount of incorrectly folded shapes. Tris and phosphate buffers were generally useful. Incubation at 4 ° C produced higher levels of properly folded TPO. Refolding yields of 40-60% (based on the amount of reduced and denatured TPO used in the refolding reaction) are typical of preparations of TPO that have been cleaned by the first C4 step. Active material can be obtained if less pure preparations are used (e.g. immediately after the Superdex 200 column or after the initial extraction of the refractive bodies), although the yields are lower due to the extensive precipitation and interaction with non-TPO proteins during the TPO refolding process.

Since TPO contains four cysteine residues, it is possible to produce three different versions of this protein:

Version 1: disulfides between cysteine residues 1-4 and 2-3

Version 2: disulfides between cysteine residues 1-2 and 3-4

Version 3: disulfides between cysteine residues 1-3 and 2-4. During the initial investigation to determine the refolding conditions, several different peaks containing the TPO protein were separated by C4 reverse phase chromatography. Only one of these peaks had significant biological activity, as determined using the Ba / F3 assay. The refolding conditions were then optimized in order to preferably obtain this version. Under these conditions, the misfolded versions were less than 10-20% of the total monomeric TPO obtained from the solubilization step.

The disulfide pattern for the biologically active TPO has been determined as 1-4 and 2-3 using spectrometry and protein sequencing, the cysteines being numbered one after the other from the amino terminus. This cysteine cross-linking pattern is in line with the known disulfide binding pattern of the related molecule erythropoietin.

D. Biological Activity of Recombinant Refolded TPO Refolded and purified TPO has activity in the in vitro and in vivo assays. For example, in the Ba / F3 assay, half maximal stimulation of thymidine incorporation into the Ba / F3 cells by TPO (Met-1 1-153) was achieved at 3.3 pg / ml (0.3 pM). In the mpl receptor-based ELISA, the half-maximum activity occurred at 1.9 ng / ml (120 pM). In normal and myelosuprimized animals, produced by almost lethal X-rays, refolded TPO (Met 1 1-153) was highly potent (activity was observed at doses as low as 30 ng / mouse) to stimulate the production of new platelets. Similar biological activity was observed for other forms of TPO that had been refolded according to the procedures described above (see Figures 25, 26 and 28). 14. Method of measuring thrombopoietic activity

Thrombopoietic activity can be measured in various assays, including the Ba / F3-mpI ligand assay described in Example 1, an in vivo mouse platelet re-bound synthesis assay, an induction of platelet surface antigen induction. measured by an anti-platelet immunoassay (anti-GPIIbIIIa) for a human mega-karyoblastic leukemia cell line (CMK) (see Sato et al., Brit. J. Haematol., 72: 184-190 [1989] (see also the liquid suspension megakaryocytopoiesis assay described in Example 4) and by induction of polyploidization in a magakaryoblastic cell line (DAMI) (see Ogura et al., Blood, 72 (1): 49-60 [1988]) Megakaryocytes from immature, essentially non-DNA synthesizing lines to morphologically identifiable megakaryocytes comprise a process that includes the appearance of cytoplasmic organelles, the acquisition of membrane antigens (GPIIbIIIa), endoreplication and release of Tiles as described in the introduction. A lineage-specific promoter (i.e., the mpl ligand) of megakaryocyte maturation would be expected to induce at least some of these changes in immature megakaryocytes, thereby leading to platelet release and a reduction in thrombocytopenia. Thus, assays were designed to measure the occurrence of these parameters in immature megakaryocyte cell lines, i.e., CMK and DAMI cells. The CMK assay (Example 4) measures the appearance of a specific platelet marker, GPIIbIIIa / and the platelet shape formation. The DAMI assay (Example 15) measures endoreplication because the increase in ploidy is a hallmark of mature megakaryocytes. Recognizable megakaryocytes have ploidy values of 2N, 4N, 8N, 16N, 32N, etc. Finally, the in vivo mouse platelet rebound assay (Example 16) is useful for demonstrating that administration of the test compound (here the mpl- Ligands) leads to an increase in the number of platelets.

Two other in vitro assays have been developed to measure TPO activity. The first is a kinase receptor activation (KIRA) ELISA in which CHO cells are transfected with an mpl-Rse chimera, and the tyrosine phosphorylation of Rse is measured by the ELISA, after. the mpI portion of the chimera has been exposed to the mpI ligand (see Example 17). The second is a receptor-based Elisa in which an ELISA plate coated with rabbit anti-human IgG captures the human chimeric receptor mpl-IgG, which in turn binds the mpl ligand to be tested. A biotinylated rabbit polyclonal antibody to the mpl ligand (TP0155) is used to detect bound mpl ligand, which is measured using streptavidin peroxidase as described in Example 18. 15. In vivo biological response from normal and sublethal irradiated mice treated with TPO

Normal and sublethally irradiated MSuse were treated with shortened and complete TPO isolated from Chinese hamster ovary cells (CHO), E. coli and human embryonic kidney (293) cells. Both forms of TPO, generated in these three hosts, stimulated platelet production in mice, but the complete TPO, isolated from CHO, appeared to elicit a larger in vivo response. The results show that the correct glycosylation of the car-boxyterminal domain may be required for optimal in vivo activity. (a) E. Coli rhTPO (Met-1,153)

The "Met" form of the EPO domain (Met in the -1 position plus the first 153 residues of human TPO), made in E.

Coli (see Example 23) was injected daily into normal female C57 B6 mice as described in the legends to Figures 25A, B and C. These figures showed that the non-glycosylated, truncated form of TPO, made in E. Coli, and refolded as described above, is able to stimulate approximately a two-fold increase in platelet production in normal mice without affecting the red or white blood cell population.

The same molecule, injected daily into sublethally irradiated (Cs) female C57 B6-Muse, as described in the legends to Figures 26A, B and C, stimulated platelet production and reduced the low point, but had no effect on red blood cells or Leukocytes. (b) CH0-rhTP0332

The full form of CHO-produced TPO injected daily into normal female C57 B6 mice, as described in the legends to Figures 27A, B and C, produced about a five-fold increase in platelet production in normal mice without affect the erythrocyte or leukocyte population. (c) CH0-rhTP0332; E. Coli rhTPO (Met-1,153); 293-rhTP0332 and E. Coli-rhTPOi55

Dose-response curves were created for the treatment of normal mice with rhTPO from different cell lines (CHO-rhTP0332; E. Coli-rhTPO (Met-1,153); 293-rhTP0332 and E. Coli-rhTP0155) as described in the legend 28. This figure shows that all of the forms of the molecule tested stimulated platelet production, but the complete form produced in CHO has the greatest in vivo activity. (d) CH0-rhTP0153 / CHO-rhTPO "pruned" and CH0-rhTP0332

Dose-response curves were also established for the treatment of normal mice with various forms of rhTPO, made in CHO (CH0-rhTP0153, CHO-rhTPO "pricked" and CHO-rhTP0332), as described in the legend to FIG. 29. This figure shows that all CHO forms of the molecule tested stimulated platelet production, but that the full 70 Kda form has the greatest in vivo activity. 16. General recombinant production of mpl ligand and variants

Preferably, mpl ligand is made by standard recombinant methods, which include making the mpl ligand polypeptide by culturing transfected lines to express mpl ligand nucleic acid (typically by transforming the cell with an expression vector) and Extract the polypeptide from the lines. However, it is also contemplated that the mpl ligand can be made by homologous recombination, or by recombining production methods using control elements that are introduced into lines that already contain DNA necessary for the mpl ligand encoded. For example, an effective promoter / enhancer element, a suppressor, or an exogenous transcription-modulating element can be introduced into the genome of the intended host cell in the vicinity and orientation sufficient to affect the transcription of the DNA that is desired mpl ligand polypeptide encoded. The control element does not encode the jnpl ligand, rather the DNA is peculiar to the host cell genome. Next, look for lines that produce the receptor polypeptide of the invention, or look for increased or decreased levels of expression, if desired.

Thus, the invention comprises a method for producing an mpI ligand comprising introducing a transcription-modulating element into the genome of a cell, containing the mpI ligand nucleic acid molecule, in sufficient proximity and orientation to the nucleic acid molecule to transcribe it influence, with an optional further step, which comprises culturing the cell containing the transcription modulating element and the nucleic acid molecule. The invention also encompasses a host cell containing its own mpl ligand nucleic acid molecule, operably linked to exogenous control sequences which are recognized by the host cell. A. Isolate DNA encoding the mpl ligand polypeptide

The DNA encoding the mpI ligand polypeptide can be obtained from any cDNA library made from tissue that is believed to contain the mpl ligand mRNA and express it at a detectable level. The mpl ligand gene can also be obtained from a genomic DNA library or by in vitro oligonucleotide synthesis of the complete nucleotide or amino acid sequence.

The banks are screened with samples designed to identify the gene of interest or the protein encoded by it. For cDNA expression banks, suitable samples include monoclonal or polyclonal antibodies that recognize and specifically bind to the mpl ligand. Samples suitable for cDNA libraries include oligonucleotides of approximately 20 to 80 bases in length that encode known or accepted portions of the mpl ligand cDNA from the same or a different type; and / or complementary or homologous c-DNAs or fragments thereof which encode the same or a similar gene. Suitable samples for screening genomic DNA libraries include, but are not limited to, oligonucleotides, cDNAs or fragments thereof that encode the same or a similar gene, and / or homologous genomic DNAs or fragments thereof. Screening of the cDNA or genomic library with the selected sample can be performed using standard procedures as described in Sambrook et al., Chapters 10-12, see above.

An alternative means of isolating the gene encoding the mpl ligand is by using the PCR technique as described in section 14 of Sambrook et al., Supra. This method requires the use of oligonucleotide samples that hybridize to DNA encoding the mpl ligand. Strategies for selecting the oligonucleotides are described below.

A preferred method of practicing the invention is to use carefully selected oligonucleotide sequences to screen cDNA libraries from various tissues, preferably human or porcine (adult or fetal) or liver cell lines. For example, cDNA banks of a human fetal liver cell line are searched with the oligonucleotide samples. Alternatively, human genomic banks can be searched with the oligonucleotide samples. The oligonucleotide sequences selected as a sample should be long enough and sufficiently specific to minimize false positive results. The actual nucleotide sequence (s) is usually designed based on regions of the mpl ligand that have the least codon redundancy. The oligonucleotides can be degenerate at one or more positions. The use of degenerate oligonucleotides is of particular importance when searching a bank of a type of which the preferred codon usage is not known.

The oligonucleotide must be labeled so that it can be detected by hybridization to DNA in the bank to be searched. The preferred method of marking is 32

Use of ATP (e.g. γ P) and polynucleotide nose to radiolabel the 5 'end of the oligonucleotide. However, other methods of labeling the oligonucleotides can be used including, but not limited to, biotinylation and enzyme labeling.

Of particular interest is the mpl ligand nucleic acid that encodes the complete mpl ligand polypeptide. In some preferred embodiments, the nucleic acid sequence comprises the signal sequence of the native mpl ligand. A nucleic acid comprising the complete protein coding sequence is obtained by searching selected cDNA or genomic libraries with the aid of the deduced amino acid sequence. B. Amino acid sequence variants of native mpI ligand

Amino acid sequence variants of mpl ligand are made by introducing appropriate nucleotide changes into the mpl ligand DNA or by in vitro synthesis of the desired mpl ligand polypeptide. Such variants include e.g. Deletions or insertions or substitutions of residues within the amino acid sequence for the porcine mpI ligand. For example, carboxy-terminal portions of the mature full mpl ligand can be removed by proteolytic cleavage, either in vivo or in vitro, or by cloning and expressing a fragment or the DNA encoding the full mpl ligand to be a biologically active variant to create. Any combination of deletion, insertion and substitution is carried out to arrive at the final construct, provided that the final construct has the desired biological activity. The amino acid changes can also change the post-translational events on the mpl ligand, such as changing the number or position of the glycosylation sites. For the design of an amino acid sequence variant of the jnpl ligand, the location of the mutation site and the type of mutation depend on the mpl ligand property (s) to be changed. The locations for the mutation can be modified individually or in series, e.g. by (1) first substituting with a choice of conservative amino acids and then with more radical selections depending on the result to be achieved, (2) deleting the target residue, or (3) inserting residues of the same or a different class in proximity to the localized one Position, or combinations of options 1-3.

A useful method for identifying certain residues or regions of the mpl ligand polypeptide that are preferred sites for mutagenesis is called "alanine scanning mutagenesis" as described by Cunningham and Wells, Science, 244: 1081- 1085 [1989]. Here, a residue or a group of target residues is identified (e.g. charged residues such as arg, asp, his, lys and glu) and replaced by any, but preferably a neutral or negatively charged amino acid (most preferred are alanine or polyalanine) to influence the interaction of the amino acids with the surrounding aqueous environment inside or outside the cell. These domains, which prove to be functionally sensitive to the substitutions, are then further elucidated by inserting further or different variants at or for the sites to be substituted. Thus, while the location for the insertion of an amino acid sequence variation is predetermined, the nature of the mutation per se need not be predetermined. For example, to optimize the performance of a mutation at a given location, ala scanning or random mutagenesis is performed on the target codon or region, and the expressed mpl ligand variants are searched for the optimal combination of desired activity.

There are two principal variables in the construction of amino acid sequence variants: the location of the mutation site and the nature of the mutation. For example, mpI ligand polypeptide variants include those variants of the mpI ligand sequence that can be naturally occurring alleles (which do not require manipulation of the mpl ligand DNA) or pre-determined mutant forms are created by mutating the DNA to either to arrive at an allele or a variant that does not occur in nature. In general, the location and nature of the mutation chosen will depend on the mpl ligand property to be modified.

Amino sequence deletions are usually in the range of about 1-30 residues, preferably about Ι-ΙΟ residues, and are usually contiguous. Alternatively, the amino acid sequence deletions for the mpl ligand can capture part or all of the carboxy terminal glycoprotein domain. Amino acid sequence deletions can also include one or more of the first six amino terminal residues of the mature protein. Optionally, amino acid sequence deletions include one or more residues in one or more of the loop regions that exist between the "helical bundles". Contiguous deletions usually involve an even number of residues, but single or odd deletions are also included. Deletions can be introduced in regions of low homology among the mpl ligands that share the greatest sequence identity to modify the activity of the mpl ligand. Or, deletions can be introduced in regions with low homology between human mpl ligand and other mammalian mpI ligand polypeptides that have the greatest sequence identity to the human mpl ligand. Deletions from a mammalian mpl ligand polypeptide in areas of substantial homology with other mammalian mpl ligands are more likely to significantly alter the biological activity of the mpl ligand. The number of contiguous deletions is selected so that the tertiary structure of the mpl ligands is maintained in the domain concerned, e.g. a beta sheet or an alpha helix.

Amino acid sequence insertions include amino and / or carboxy terminal fusions that range in length from one residue to polypeptides containing 100 or more residues, as well as insertions within the sequence of one or more amino acid residues. Inserts within the sequence (intrasequence insertions) (i.e., insertions within the mature mpl ligand sequence) can usually comprise approximately 1-10 residues, preferably 1-5, most preferably 1-3. An exemplary preferred fusion is that of the mpl ligand or fragment thereof with another cytokine or fragment thereof. Examples of terminal insertions include the mature mpl ligand with an N-terminal methionyl residue, an artifact of direct expression of the mature mpl ligand in recombinant cell culture, and the fusion of a heterologous N-terminal signal sequence to the N-terminus of the mature mpl -Ligand Molecules, to facilitate the secretion of the mature mpl ligand from recombinant hosts. Such signal sequences are generally obtained from, and are thus homologous to, the intended host cell type. Suitable sequences include STII or Ipp for E. coli, alpha factor for yeast, and viral sequences such as herpes gD for mammalian cells.

Other insertion variants of the mpl ligand molecule include fusion to the N or C terminus of the mpl ligand from immunogenic polypeptides (i.e., they are not endogenous to the host to which the fusion is administered), e.g. bacterial polypeptides such as beta-lactamase or an enzyme encoded by the E. coli trp locus, or yeast protein, and C-terminal fusions with proteins with a long half-life, such as the constant regions of immunoglobulins (or other immunoglobulin regions), albumin or ferritin as described in WO 89/02922, published April 6, 1989.

A third group of variants are amino acid substitution variants. In these variants, at least one amino acid residue in the mpl ligand molecule has been removed and a different residue has been inserted in its place. The sites of greatest interest for substitutional mutagenesis include sites identified as the active site (s) of the mpI ligand and sites where the amino acids found in other analogues are considerably different in terms of the extent of the side chain, the charge or the hydrophobicity, but also where there is a high degree of sequence identity at the selected site between different mpl ligand types and / or between the different animal analogues of an mpl ligand member.

Other places of interest are those at which special residues of the mpI ligand obtained from different family members and / or animal species are identical for one member. These sites, in particular those that lie within a sequence of at least three other identical conserved sites, are substituted in a relatively conservative manner. Such conservative substitutions are listed in Table 3 under the heading of the preferred substitutions. If such substitutions lead to a change in the biological activity, then more significant changes, referred to as exemplary substitutions in Table 3 or as described further below with reference to amino acid classes, are introduced and the products are examined.

Significant modifications to the function or immunological identity of the mpl ligand are accomplished by selecting substitutions that differ significantly in their effect on maintaining (a) the structure of the polypeptide backbone in the area of substitution, e.g. as a leaflet or helical conformation, (b) the charge or the hydrophobicity of the molecule at the target site, or (c) your circumference of the side chain. Naturally occurring residues are divided into groups based on the common side chain properties: (1) hydrophobic norleucine, Met, Ala, Val, Leu, Ile; (2) neutral, hydrophilic Cys, Ser, Thr; (3) Acid Asp, Glu; (4) basic Asn, Gin, His, Lys, Arge (5) residues, the chain arrangement-influencing Gly, Pro; and (6) Aromatic: Trp, Tyr, Phe. • Non-conservative substitutions require an exchange of a member of one of these classes by a member of another class. Such substituted residues can also be introduced into the conservative substitution sites, but more preferably into the remaining (non-conserved) sites.

In one embodiment of the invention, it is desirable to inactivate one or more protease cleavage sites that are present in the molecule. These sites are identified by searching the encoded amino acid sequence in the case of trypsin, e.g. for an arginyl or lysinyl residue. If protease cleavage sites are identified, they are rendered inactive against proteolytic cleavage by substituting the target residue with another residue, preferably a basic residue such as glutamine or a hydrophobic residue such as serine; by deleting the rest; or by inserting a prolyl residue immediately after the resr.

In another embodiment, any methionyl residue that is not the starting methionyl residue of the signal sequence or any residue that is within about 3 residues N- or C-terminal to such a methionyl residue is substituted by another residue (preferably in accordance with Table 3 ) or deleted. Alternatively, about 1-3 residues are inserted adjacent to such sites.

Any cysteine residue that is not involved in maintaining the correct conformation of the mpl ligand can also be replaced, generally with serine, to prevent the stability of the molecule from oxidation and accidental crosslinking. It has been found that the first and fourth cysteine in the EPO domain, numbered from the amino terminus, are necessary to maintain the correct conformation, but the second and third are not required. As a result, the second and third cysteine of the EPO domain can be replaced.

Nucleic acid molecules which encode amino acid sequence variants of the jnpl ligand are produced by a number of methods known from the prior art. These methods include, but are not limited to, X isolation from a natural starting material (in the fail of naturally occurring amino acid sequence variants) or preparation by oligonucleotide-mediated (or site-directed) mutagenesis, PCR mutagenesis or cassette mutation Genesis of a previously made variant or a non-variant version of the mpl ligand polypeptide.

Oligonucleotide-mediated mutagenesis is a preferred method for generating substitution, deletion and insertion variants of mpl ligand DNA. This technique is known in the art as described by Adelman et al., DNA, 2: 183 [1983]. Briefly stated, mpl ligand DNA is changed by hybridizing an oligonucleotide encoding the desired mutation to a DNA template, the template being the single strand form of a plasmid or bacteriophage containing the unchanged or native DNA Sequence of the mpl ligand. After hybridization, a DNA polymerase is used to synthesize a complete second complementary strand of the template, which thus contains the oligonucleotide primer and which codes for the selected change in the mpl ligand DNA. Generally, oligonucleotides of at least 25 nucleotides in length are used. An optimal oligonucleotide has twelve to 15 nucleotides that are completely complementary to the template on each side of the nucleotide (s) encoding the mutation. This ensures that the oligonucleotide hybridizes properly to the single-stranded DNA template molecule. The oligonucleotides can be easily synthesized using techniques known in the art, as described by Créa et al., Proc. Natl. Acad. Sci. USA, 75: 5765 [1978].

The DNA template can be generated from those vectors which are either obtained from bacteriophage M13 vectors (the commercially available M13mpl8 and M13mpl9 vectors are suitable) or from vectors which contain a single-stranded phage replication origin, as described by Viera et al., Meth. Enzymol., 153: 3 (1987). Thus, the DNA to be mutated can be inserted into one of these vectors to create a single-stranded template. The preparation of the single-stranded template is described in Sections 4.21-4.41 of Sambrook et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Laboratory Press, NY 1989).

Alternatively, a single-stranded DNA template can be generated by denaturing a double-stranded plasmid (or other) DNA using standard techniques.

To change the native DNA sequence (e.g. to generate amino acid sequence variants), the oligonucleotide is hybridized to the single-stranded template under suitable hybridization conditions. A DNA polymerizing enzyme, usually the Klenow fragment of DNA polymerase I, is added to synthesize the complementary strand of the template using the oligonucleotide as a primer for synthesis. A herero duplex is thus formed such that one strand of the DNA encodes the mutated form of the mpl ligand, and the other strand (the original template) encodes the native, unchanged sequence of the mpl ligand. This heteroduplex molecule is then transformed into a suitable host cell, usually a prokaryote such as E. Coli JM101. After the lines are cultured, they are plated on agarose plates and screened using the 32-phosphorus labeled oligonucleotide primer to identify the bacterial colonies containing the mutated DNA. The mutated region is then removed and placed in a suitable vector for protein production, usually in an expression vector of the type that is commonly used to transform a suitable host.

The method described above can be modified to produce a homoduplex molecule in which both strands of the plasmid contain the mutation (s). The modifications are as follows: The single-stranded oligonucleotide is annealed to the double-stranded template as described above. A mixture of three deoxyribonucleotides, deoxyriboadenosine (dATP), deoxyriboguanosine (dGTP) and deoxyribothymidine (dTTP) is combined with a modified thio-deoxyribocytosine, dCTP- (aS) (which can be obtained from Amersham Corporation). This mixture is added to the template-oligonucleotide complex. When DNA polymerase is added to this mixture, a DNA strand identical to the template, except for the mutated base, is generated. This new DNA strand also contains dCTP- (aS) instead of dCTP, which serves to protect it from restriction endonuclease cleavage.

After the template strand of the double-stranded heteroduplex has been nicked with a suitable restriction enzyme, the template strand can be degraded with ExoIII nuclease or another suitable nuclease, along the region which contains the site (s) to be mutagenized. The reaction is then stopped to leave a molecule that is only partially single-stranded. A complete double-stranded DNA homoduplex is then formed with the aid of DNA polymerase in the presence of all four deoxyribonucleotide triphosphates, ATP and DNA ligase. This homoduplex molecule can be transformed into a suitable host cell, such as E. Coli JM101, as described above. DNA encoding mpl ligand mutants with more than one amino acid to be replaced can be produced in several ways. If the amino acids in the polypeptide chain are located close to each other, they can be mutated simultaneously using an oligonucleotide that encodes all of the desired amino acid substitutions. However, if the amino acids are spaced some distance apart (separated by more than ten amino acid residues), it is more difficult to generate a single oligonucleotide that encodes all of the desired changes. Instead, one of two alternative methods can be used.

The first method creates a separate oligonucleotide for each amino acid to be substituted. The oligonucleotides are then hybridized simultaneously to the single-stranded template DNA, and the second strand of DNA, which is synthesized from the template, codes for all desired amino acid substitutions.

The alternative method involves two or more rounds of mutagenesis to create the desired mutant. The first round corresponds to that described for the only mutation: wild-type DNA is used as a template, a

Oligonucleotide encoding the first desired amino acid substitution (s) is hybridized to the template and the heteroduplex molecule is then generated. The second round of mutagenesis uses the mutated DNA product from the first round of mutagenesis as the template. This matrix therefore already contains one or more mutations. The oligonucleotide encoding the additionally desired amino acid substitution (s) is then hybridized to the template, and the DNA strand obtained now codes for mutations from the first and second rounds of mutagenesis. This DNA obtained can be used as a template in a third round of mutagenesis, etc. PCR mutagenesis is also suitable for producing amino acid variants of the mpl ligand polypeptide. While the following discussion relates to DNA, it goes without saying that the technique also applies to RNA. The PCR technique generally relates to the following procedure (see Ehrlich, see above, the chapter by R. Higuchi, pp. 61-70). If small amounts of template DNA are used as starting material in a PCR, primers that differ slightly in sequence from the corresponding region in a template DNA, are used to generate relatively large amounts of a specific DNA fragment that differs from the template sequence only at the positions where the primer differs from the matrix differs. To introduce a mutation into a plasmid DNA, one of the primers is designed so that it overlaps with the position of the mutation and contains the mutation; the sequence of the other primer must be identical to a sequence portion of the counter strand of the plasmid, but this sequence can be located anywhere along three plasmid DNA. However, it is preferred that the sequence of the second primer be located within 200 nucleotides of that of the first primer so that in the end the complete, expanded region of the DNA delimited by the primers can be easily sequenced. The amplification by PCR using a pair of primers, such as the one just described, leads to a population of DNA fragments which differs at the mutation site defined by the primer and possibly at different positions since the copying of the template is somewhat error-prone .

If the ratio of template to product material is extremely low, the vast majority of product DNA fragments contain the desired mutation (s). The product material is used to replace the corresponding region in the plasmid that served as a PCR template using standard DNA techniques. Mutations at separate positions can be introduced simultaneously using either a mutated second primer or performing a second PCR with different mutant primers and simultaneously ligating the two PCR fragments obtained to the vector fragment in a three (or more) part ligation.

In a specific example of PCR mutagenesis, template plasmid DNA (1 μς) is linearized by cleavage with a restriction endonuclease which has a single recognition site in the plasmid DNA outside the region to be amplified. 100 ng of this material is added to a PCR mixture containing PCR buffer which contains the four deoxynucleotide triphosphates and which is contained in the GeneAmpO kits (obtained from Perkin-Elmer Cetus, Norwalk, CT and Emeryville, CA) and containing 25 pmol of each oligonucleotide primer, and adjusted to a final volume of 50 μΐ. The reaction mixture is covered with 35 μl mineral oil. The reaction mixture is denatured for 5 min at 100 ° C., briefly placed on ice, and then 1 μl Thermus aquaticus (Taq) DNA polymerase (5 units / μl, obtained from Perkin-Elmer Cetus) below the mine. added oil layer. The reaction mixture is then introduced into a DNA thermal cycler (obtained from Perkin-Elmer Cetus), which is programmed as follows: 2 min 55 ° C, 30 s 72 ° C, then 19 cycles as follows: 30 s 94 ° C, 30 s 55 ° C and 30 s 72 ° C.

At the end of the program, the reaction tube is removed from the thermal cycler, the aqueous phase is transferred to a new tube, extracted with phenol / chloroform (50/50 volume) and precipitated with ethanol, and the DNA is recovered according to standard procedures. This material is then subjected to the appropriate treatments for insertion into a vector.

Another method of producing variants, cassette mutagenesis, is based on the technique as described by Wells et al., Gene, 34: 315 [1985]. The starting material is the plasmid (or another vector) containing the mpl ~ ligand DNA to be mutated. The codon (s) to be mutated in the mpl ligand DNA are identified. There must be a unique restriction endonuclease cleavage site on each side of the identified mutation site (s). If such restriction cleavage sites do not occur, they can be generated using the oligonucleotide-mediated mutagenesis method described above, in order to introduce them into the mpl ligand DNA at suitable sites. After the restriction sites have been introduced into the plasmid, the plasmid is cut at these sites to linearize it. A double-stranded oligonucleotide encoding the DNA sequence between the restriction sites, but containing the desired mutation (s), is synthesized using standard procedures. The two strands are synthesized separately and then hybridized together using standard techniques. This double-stranded oligonucleotide is called the cassette. This cassette is designed to have 3 'and 5' ends that are compatible with the ends of the line arized plasmid so that they can be ligated directly to the plasmid. This plasmid now contains the mutated mpl ligand DNA sequence. C. Inserting a nucleic acid into a replicable vector

The nucleic acid (e.g. cDNA or genomic DNA) encoding a native or modified mpl ligand polypeptide is inserted into a replicable vector for further cloning (amplification) of DNA or for expression. Many vectors are available, and the selection of the appropriate vector depends on (1) whether it should be used for DNA amplification or expression, (2) the size of the nucleic acid to be inserted into the vector, and (3) that with the Vector host cell to be transformed. Each vector contains different components depending on its function (replication of DNA or expression of DNA) and the host cell with which it is compatible. The vector components generally include, but are not limited to, one or more of the following: a signal sequence, an origin of replication, one or more marker genes, an enhancer element, a promoter, and a transcription termination sequence. (i) Signal sequence component

The mpI ligand according to the invention can not only be expressed directly, but also as a fusion with a heterologous polypeptide, preferably a signal sequence or another polypeptide with a specific cleavage site at the N-terminus of the mature protein or polypeptide. In general, the signal sequence can form part of the vector, or it can represent part of the mpl ligand DNA that is introduced into the vector. The selected heterologous signal sequence should be a sequence that is recognized and processed by the host cell (i.e., cleaved by the signal peptidase). For prokaryotic host cells that do not recognize and do not process the native mpl ~ ligand signal sequence, the signal sequence is replaced by a prokaryotic signal sequence selected from e.g. the group of leaders of alkaline phosphatase, penicillinase, Ipp or heat-resistant enterotoxin II. For yeast secretion, the native signal sequence can be replaced by e.g. the leaders of yeast invertase, alpha factor or acid phosphatase, the C. albicans glucoamylase reader (EP 362,179, published April 4, 1990) or the signal as described in WO 90/13646, published on November 15, 1990. At expression in mammalian cells, the native signal sequence (ie, the mpl ligand presequence, which normally controls the secretion of the mpl ligand from its mammalian natural cells in vivo) is satisfactory, although other mammalian signal sequences may be suitable, such as signal sequences from other mpl ligand polypeptides or from the same mpl ligand from a different animal species, signal sequences from one mpl ligand, and signal sequences from secreted polypeptides of the same or related type, as well as viral secretory leaders, e.g. the herpes simplex gD signal. (ii) Replication origin component

Expression and cloning vectors contain a nucleic acid sequence that allows the vector to replicate in one or more selected host cells. In general, in cloning vectors, this sequence is a sequence that allows the vector to replicate independently of the chromosomal host DNA and includes origins of replication or autonomously replicating sequences. Such sequences are for a wide variety of bacteria, yeast and

Viruses known. The origin of replication of the plasmid pBR322 is suitable for most Gram-negative bacteria, the origin of replication of the 2 μ plasmid is suitable for yeast, and various viral origins of replication (SV40,

Polyoma, adenovirus, VSV or BPV) are useful for cloning vectors in mammalian cells. In general, the origin of replication component is not needed for acid expression vectors (the SV40 origin of replication is usually used only because it contains the early promoter).

Most expression vectors are "shuttle" vectors, i.e. they are capable of replication in at least one class of organisms, but they can be transfected into another organism for expression. For example, a vector is cloned into E. Coli, and then the same vector is transfected into yeast or mammalian cells for expression, although it is unable to replicate independently of the host cell chromosome. DNA can also be increased by inserting it into the host genome. This is easily achieved using Bacillus species as the host, e.g. by inserting a DNA sequence into the vector that is complementary to a sequence found in Bacillus genomic DNA. The transfection of

Bacillus with the vector leads to homologous recombination with the genome and the incorporation of the mpl ligand DNA. However, obtaining genomic DNA encoding the mpl ligand is more complex than obtaining an exogenously replicating vector because restriction enzyme cleavage is required to cut out the mpl ligand DNA. (iii) Selection gene component

Expression and cloning vectors should contain a selection gene, also called a selection marker. This gene codes for a protein that is necessary for survival

Or growth of the transformed host cells, which are cultivated in a selective culture medium. Host cells that are not transformed with the vector containing the selective gene will not survive in the culture medium. Typical selection genes encode proteins that (a) confer resistance to antibiotics or other toxins, e.g. Ampicillin, neomycin, methotrexate or tetra-cyclin, or (b) complement auxotrophic deficiency, or (c) provide critical foods that are not available in complex media, e.g. the gene coding for D-alanine racemase for Bacilli.

An example of a selection scheme uses an active ingredient to stop the growth of a host cell. Such lines, which have been successfully transformed with a heterologous gene, express a protein which confers resistance to an active substance and thus survive the selection process. Examples of such a dominant selection use the active substances neomycin (Southern et al., J.

Molec. Appl. Genet., 1: 327 [1982]), mycophenolic acid (Mulligan et al., Science, 209: 1422 [1980]) or hygromycin (Sugden et al., Mol. Cell. Biol., 5: 410-413 [1985] ). The three examples given above use bacterial genes under eukaryotic control to confer resistance to the appropriate active ingredient G418 or neomycin (geneticin), xgpt (mycophenolic acid) or hygromycin.

Examples of other suitable selectable markers for mammalian cells are those which allow the identification of lines which the mpl ligand nucleic acid can take up, such as dihydrofolate reductase (DHFR) or thymidine kinase. The transformed mammalian cells are placed under selection pressure in such a way that only the transformants are uniquely adapted to survival by incorporating the marker. Selection pressure is imposed by culturing the transformants under conditions where the concentration of the selection agent in the medium is successively changed, thereby amplifying both the selection gene and the DNA encoding the mpl ligand polypeptide. Amplification is the process by which genes that have a greater need to produce a protein critical for growth are repeated in tandem in the chromosomes of subsequent generations of the recombinant cells. Increased amounts of mpl ligand are synthesized from the amplified DNA.

For example, lines transformed with the DHFR selection gene are first identified by cultivating all transformants in a culture medium containing methotrexate (Mtx), a competitive DHFR antagonist. A suitable host cell if wild-type DHFR is used is the Chinese hamster ovary cell line (CHO), which is deficient in DHFR activity; produced and cultivated as described by Urlaub and Chasin, Proc. Natl. Acad. Sci. USA, 77: 4216 [1980]. The transformed lines are then exposed to increased levels of Mtx. This leads to the synthesis of multiple copies on the DHFR gene and at the same time multiple copies on other DNA that comprise the expression vectors, such as the DNA encoding the mpl ligand. This amplification technique can be used in any other suitable hosts, for example ATCC No. CCL61 CHO-K1, regardless of the presence of endogenous DHFR, e.g. a mutated DHFR gene is used which is highly resistant to Mtx (EP 117,060). Alternatively, host cells (especially wild-type hosts containing endogenous DHFR) can be transformed or cotransformed with DNA sequences encoding the mpl ligand, wild-type DHFR protein and other selectable markers such as aminoglycoside-31-phosphotransferase (ΑΡΗ) are selected via the cell growth in medium containing a selection agent for the selection marker, such as an aminoglycoside antibiotic, for example Kanamycin, Neomycin or G418. See also U.S. Patent 4,965,199.

A suitable selection gene for use in yeast is the trpl gene, which occurs in the yeast plasmid YRp7 (Stinchcomb et al., Nature, 282: 39 [1979]; Kingsman et al., Gene, 7: 141 [1979] or Tschemper et al., Gene, 10: 157 [1980]). The trpl gene provides a selection marker for a yeast mutant strain lacking the ability to grow in tryptophan, e.g. ATCC No. 44076 or PEP4-1 (Jones, Genetics, 85:12 [1977]). The presence of the trpl lesion in the yeast host cell genome then provides an effective environment for demonstrating transformation by growth in the absence of tryptophan. Similarly, yeast strains with a Leu2 deficiency (ATCC # 20,622 or 38,626) are complemented by known ones Plasmids containing the Leu2 gene. (iv) promoter component

Expression and cloning vectors usually contain a promoter which is recognized by the host organism and which is operably linked to the mpl ligand nucleic acid. Promoters are untranslated sequences located upstream (51) from the start codon of a structural gene (generally within about 100 to 1000 bp) that involve the transcription and translation of a particular nucleic acid sequence, such as the mpl ligand nucleic acid sequence of which they are operatively linked. Such promoters typically fall into two classes, inducible and constitutive. Inducible promoters are promoters that initiate the increased levels of DNA transcription under their control in response to a change in culture conditions, e.g. the presence or absence of a nutrient or a change in temperature. A large number of promoters that are recognized by a large number of potential host cells are currently known. These promoters are operably linked to a DNA encoding the mpl ligand by removing the promoter from the

Starting material DNA by restriction enzyme cleavage and insertion of the isollerten promoter sequence into the vector. Both the native mpl ligand promoter sequence and many heterologous promoters can be used to control the amplification and / or expression of the mpl ligand DNA. However, heterologous promoters are preferred because they generally provide stronger transcription and higher yields of the expressed mpI ligand compared to the native mpl ligand promoter. Promoters suitable for use in prokaryotic hosts include the β-lactamase and lactose promoter systems (Chang et al., Nature, 275: 615 [1978]; and Goeddel et al., Nature, 281: 544 [1979]) , alkaline phosphatase, a tryptophan (trp) promoter system (Goeddel, Nucleic Acids Res., 8: 4057 [1980] and EP 36,776) and hybrid promoters such as the tac promoter (deBoer et al., Proc. Natl. Acad. Sci. USA, 80: 21-25 [1983]). However, other known bacterial promoters are suitable. Their nucleotide sequences have been published, which enables the average person skilled in the art to ligate them operably with the DNA encoding the mpl ligand (Siebenlist et al., Cell, 20: 269 [1980]) Use of linkers and adapters to provide required restriction sites. Promoters for use in bacterial systems also contain a Shine-Dalgarno (S.D.) Sequence that is operably linked to the DNA encoding the mpl ligand poly-peptide.

Promoter sequences are known for eukaryotes. Virtually all eukaryotic genes have an AT-rich region located approximately 25 to 30 bases upstream of the point at which transcription is initiated. Another sequence found 70-80 bases upstream from the start of transcription in many genes is the CXCAAT region, where X can be any nucleotide. At the 3 'end of most eukaryotic genes there is an AATAAA sequence which can represent the signal for the attachment of the poly A tail to the 3' end of the coding sequence. All of these sequences are suitably located in eukaryotic expression vectors.

Examples of suitable promoter sequences for use in yeast hosts include the promoters for 3-phosphoglycerate kinase (Hitzeman et al., J. Biol. Chem., 255: 2073 [1980]) or other glycolytic enzymes (Hess et al., J. Adv Enzyme Reg., 7: 149 [1968] and Holland, Biochemistry, 17.4900 [1978]), such as from enolase, glyceraldehyde-3-phosphate dehydrogenase, hexokinase, pyruvate decarboxylase, phosphofructokinase, glucose-6-phosphate isomerase, 3-phosphoglycerate mutase, Pyru-vatkinase, triose phosphate isomerase, phosphoglucose isomerase and glucokinase.

Other yeast promoters that are inducible promoters with the further advantage of transcription controlled by the cultivation conditions are the promoter regions for alcohol dehydrogenase 2, isocytochrome c, acid phosphatase, degrading enzymes associated with nitrogen metabolism, metallothionine, glyceraldehyde-3 phosphate dehydrogenase, and enzymes for the use of maltose and galactose. Suitable vectors and promoters for use in expression in yeast are further described in Hitzeman et al., EP 73,657A. Yeast enhancers are also used advantageously with yeast promoters.

The mpl ligand transcription of vectors in mammalian host cells is, for example, controlled by promoters obtained from the genomes of viruses such as polyomavirus, chicken pox virus (UK 2,211,504, published July 5, 1989), adenovirus (such as adenovirus 2). , Bovine papilloma virus, avian sarcoma virus, cytomegalovirus, a retrovirus, hepatitis B virus and particularly preferably Simian Virus 40 (SV40), from heterologous mammalian promoters, e.g. the actin promoter or an immunoglobulin promoter, heat shock promoters and dein promoter that is normally associated with the mpl ligand sequence, provided that such promoters are compatible with the host cell system.

The early and late promoters (early and late) of the SV40 virus are easily obtained as an SV40 restriction fragment, which also contains the SV40 viral origin of replication. Fiers et al., Nature, 273: 113 [1978], Mulligan and Berg, Science, 209: 1422-1427 [1980]; Pavlakis et al., Proc. Natl., Acad. Sci. USA, 78: 7398-7402 [1981]. The immediate ear ly promoter of the human cytomegalovirus is suitably obtained as a HindII-E restriction fragment. Greenaway et al., Gene, 18: 355-360 [1982]. A system for expressing DNA in mammalian hosts using bovine papillora virus as a vector is disclosed in U.S. Patent 4,419,446. A modification of this system is described in U.S. Patent 4,601,978. See also Gray et al., Nature, 295: 503-508 [1982] for expressing immune interferon-encoding cDNA in monkey cells; Reyes et al., Nature, 297: 598-601 [1982] for expressing human β-interferon cDNA in mouse cells under the control of a herpes simplex virus thymidine kinase promoter; Canaani and Berg, Proc. Natl. Acad. Sci. USA, 79: 5166-5170 [1982] for the expression of the human interferon-β1 gene in cultured mouse and rabbit cells; and Gorman et al., Proc. Natl. Acad. Sci. USA, 79: 6777-6781 [1982] for the expression of bacterial CAT sequences in CV-1 monkey kidney cells, chicken embryo fibroblasts, Chinese hamster ovary cells, HeLa cells and mouse NIH-3T3 cells using the long terminal Repeats (LTR) of Rous Sarcoma virus as' a promoter. (v) enhancer element component

The transcription of a DNA encoding the mpl ligand according to the invention is increased in higher eukaryotes of trials by inserting an enhancer sequence into the vector. Enhancers are cis-acting elements of DNA, usually about 10-300 bp long, that affect a promoter to increase its transcription. Enhancers are relatively independent of orientation and position, and are 51 (Laimins et al., Proc. Natl. Acad. Sci. USA, 78: 993 [1981]) and 3 '(Lusky et al., Mol Cell Bio., 3: 1108 [1983]) from the transcription unit, within introns (Banerji et al., Cell, 33: 729 [1983]) as well as within the coding sequence itself (Osborne et al., Mol. Cell Bio., 4: 1293 [1984]). Many enhancer sequences are known from mammalian genes (globin, elastase, albumin, a-fetoprotein and insulin). Typically, however, one will use an enhancer from a eukaryotic cell virus. Examples include the SV40 enhancer on the late side of the origin of replication (bp 100-270), the cytomegalovirus early promoter enhancer, the latex polyoma enhancer on the origin of replication and adenovirus Enhancer. See also Yaniv, Nature, 297: 17-18 [1982] on enhancer elements for activating eukaryotic promoters. The enhancer can be inserted into a vector at a position 5 'or 3' to the mpl ligand coding sequence, but is preferably located at a 5 'position from the promoter. (vi) Transcription termination component

Expression vectors for use in eukaryotic host cells (yeast, fungi, insects, plants, animals, humans, or nucleated lines from other multicellular organisms) also contain sequences that are necessary for the termination of transcription and the stabilization of the mRNA. Such sequences are commonly available from the 5 'and sometimes the 3' untranslated regions of eukaryotic or viral DNAs or cDNAs. These regions contain nucleotide segments that are transcribed as polyadenylated fragments in the untranslated portion of the mRNA encoding the mpl ligand. (vii; Construction and analysis of the vectors

The construction of suitable vectors containing one or more of the components listed above uses standard ligation techniques. Isolated plasmid or DNA fragments are cleaved, tailored at the end and ligated again in the desired form in order to generate the required plasmid.

As analysis to confirm the correct sequences in the constructed plasmids, the ligation mixtures are used to transform E. Coli K12 strain 294 (ATCC No. 31,446), and successful transformants are selected by ampicillin or tetracycline resistance where appropriate. Plasmids from the transformants are prepared, analyzed by restriction endonuclease cleavage and / or sequencing according to the method of Messing et al., Nucleic Acids Res., 9: 309 [1981] or according to the method of Maxam et al., Methods in Enzymology, 65: 499 [1980]. (viii) Transient expression vectors

Expression vectors which provide transient expression in mammalian cells of DNA encoding the mpl ligand polypeptide are particularly useful in the practice of the present invention. In general, transient expression involves the use of an expression vector capable of replicating effectively in a host cell so that the host cell accumulates many copies of the expression vector and therefore high levels of a desired poly-peptide encoded by the expression vector , synthesized. Sambrook et al., Supra, pp. 16.17-16.22. Transient

Expression systems, including a suitable expression vector and a host cell, allow both the easy positive identification of polypeptides encoded by the cloned DNAs and the rapid screening of such polypeptides for the desired biological or physiological property. Thus, transient expression systems are particularly useful in the present invention for the purposes of identifying analogs and variants of the mpl ligand polypeptide that have the biological activity of the mpl ligand polypeptide. (ix) Suitable exemplary vectors for vertebrate cells

Other methods, vectors and host cells suitable for adapting to the synthesis of the mpl ligand in recombined vertebrate cell cultures are described in Gething et al., Nature, 293: 620-625 [1981]; Mantei et al., Nature, 281: 40-46 [1979]; Levinson et al .; EP 117,060; and EP 117.058. A particularly useful plasmid for the expression of mpl ligand in mammalian cell culture is pRK5 (EP 307,247, U.S. Patent 5,258,287) or pSV16B (PCT publication WO 91/08291). D. Selection and transformation of host cells

Suitable host cells for cloning and expressing the vectors herein are the prokaryotic, yeast or higher eukaryotic lines as described above. Suitable prokaryotes include eubacteria such as gram-negative or gram-positive organisms, e.g. E. Coli, Bacilli such as B. Subtilis, Pseudomonas species such as P. aeruginosa, Salmonella typhimurium oer Sèrratia marcescans. A preferred E. Coli cloning host is E. Coli 294 (ATCC No. 31,446), although other strains such as E. Coli B, E. Coli X1776 (ATCC No. 31,537) and E. Coli W3110 (ATCC No. 27325) are suitable. These examples are illustrative rather than restrictive. The host cell should preferably secrete minimal amounts of proteolytic enzymes. Alternatively, in vitro methods for cloning, e.g. PCR or other nucleic acid polymerase reactions are suitable.

In addition to prokaryotes, eukaryotic microbes such as filamentous fungi or yeast are suitable hosts for the mpl ligand-encoding vectors. Saccharomyces cerevisiae or the conventional baker's yeast are the most used organisms among the lower eukaryotic host microorganisms. However, a variety of other genera, species and strains are generally available and useful herein, such as Schizosaccharomyces pombe (Beach and Nurse, Nature, 290: 140 [1981], * EP 139,383, published May 2, 1985), Kluyveromyces hosts (US -PS 4,943,529) such as K. lactis (Louvenourt et al., J. Bacteriol., 737 [1983]), K. fragilis, K. bulgaricus, K. thermotolerans and K. marxianus, yarrowia (EP 402,226), Pichia pastoris (EP 183,070; Sreekrishna et al., J. Basic Microbiol., 28: 265-278 [1938]), Candida, Trichoderma reesia (EP 244,234), Neurospora crassa (Case et al., Proc. Natl. Acad. Sci. USA, 76: 5259- 5263 [1979]), and filamentous mushrooms such as Neurospora, Pénicillium, Toly-pocladium (WO 91/00357, published January 10, 1991), and Aspergillus hosts such as A. nidulans (Ballance et al., Biochem. Biophys. Res. Commun., 112: 284-289 [1983]; Tilburn et al., Gene, 26: 205-221 [1983], Yelton et al., Proc. Natl. Acad. Sci. USA, -81: 1470-1474 (1984)) and A. niger ( Kelly and Hynes, EMBO J. 4: 475-479 [1985]).

Suitable host cells for the expression of glycosylated mpI ligand are obtained from multicellular organisms. Such host cells are capable of complex processing and have glycosylation activities. Basically, any higher eukaryotic cell culture can be used, whether from a vertebrate or invertebrate culture. Examples of invertebrate cells include plant and insect cells. Numerous baculoviral strains and variants and corresponding permissive insect host cells from hosts such as Spodoptera frugiperda (caterpillar), Aedes aegypti (mosquito), Aedes albopictus (mosquito), Drosophila melano-gaster (fruit fly) and Bombyx mori have been identified. See e.g. Luckcw et al., Bio / Technology, 6: 47-55 [1988]; Miller et al., Genetic Engineering, Setlow et al., Ed., Vol. 8 (Plenum Publishing, 1986), pp. 277-279; and Maeda et al., Nature, 315: 592-594 [1985]. A variety of viral strains for transfection are publicly available, e.g. the L-1 variant of Autographa californica NPV and the Bm-5 strain of Bombyx mori NPV, and such viruses can be used as the virus according to the invention, in particular for the transfection of Spodoptera frugiperda cells.

Plant cell cultures from cotton, cereals, potatoes, soybeans, petunias, tomatoes and tobacco can be used as hosts. Typically, plant cells are transfected by incubation with certain strains of the bacterium Agrobacterium tumefaciens, which has previously been manipulated, to contain the mpl ligand DNA. During the incubation of the plant cell culture with A. tumefaciens, the DNA encoding the mpl ligand is transferred to the plant host cell so that it is transfected, and it will express the mpl ligand DNA under suitable conditions. Regulatory and signal sequences that are compatible with the plant cells are also available, such as the nopaline synthase promoter and polyadenylation signal sequences. Depicker et al., J. Mol. Appl. Gen., 1: 561 [1982]. Furthermore, DNA segments isolated from the upstream region of the T-DNA-780 gene can activate or increase the transcription levels of genes which can be expressed in plants in recombinant DNA-containing plant tissue. EP 321,196, published June 21, 1989.

However, interest in vertebrate cells was greatest, and the proliferation of vertebrate cells in culture (tissue culture) has become a routine procedure in recent years (Tissue Culture, Academie Press, Kruse and Patterson, ed. [1973]). Examples of useful mammalian host cell lines are the monkey kidney line CV1 transformed by SV40 (COS-7, ATCC CRL 1651), the human embryonic kidney line (293 or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen Virol ., 36:59 [1977]); Baby hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells / DHFR (CHO, Urlaub and Chasin, Proc. Natl. Acad. Sci. USA, 77: 4216 [1980]); Mouse Sertoli cells (TM4, Mather, Biol. Reprod., 23: 243-251 [1980]); Monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical cancer cells (HELA, ATCC CCL2); Canine kidney cells (MDCK, ATCC CCL 34); Buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2, HB 8065); Mouse breast tumor (MMT 060562, ATCC CCL51); TRI cells (Mather et al., Annals N.Y. Acad. Sci., 333: 44-68 (1982)); MRC 5 cells; FS4 cells; and human hepatoma cell line (Hep G2).

The host cells are transfected and preferably transformed with the above-described expression or cloning vectors according to the invention and cultivated in conventional nutrient media, suitably modified to induce the promoters, select the transformants or amplify the genes which code for the desired sequences.

Transfection refers to the uptake of an expression vector by a host cell, regardless of whether coding sequences are actually expressed. Numerous methods of transfection are known to those skilled in the art, e.g. CaPÜ4 and Elektropora'tion. Successful transfection is generally assumed if there is an indication of the functioning of the vector within the host cell.

Transformation means penetration of a DNA into an organism so that the DNA can be replicated, either as an ex-trachromosomal element or by chromosomal integration. Depending on the host cell used, the transformation is performed using standard techniques suitable for these rows. Calcium treatment using calcium chloride, as described in section 1.82 of Sambrook et al., Supra, is commonly used for prokaryotes or other cells that contain essential cell wall barriers. Infection with Agrobacterium tumefaciens is used for the transformation of certain plant cells, as described by Shaw et al., Gene, 23: 315 [1983] and WO 89/05859, published on June 29, 1989. Furthermore, plants can be transfected are using an ultrasound treatment as described in WO 91/00358, published January 10, 1991. For mammalian cells without cell walls, the calcium phosphate precipitation method of Graham and van der Eb, Virology, 52: 456-457 [1978] is preferred. General aspects of transformations in mammalian host cell systems are described by Axel in U.S. Patent 4,399,216, issued August 16, 1983. Transformations in yeast are typically carried out according to the method of Van Solingen et al., J. Bact., 130: 946 [1977] and Hsiao et al., Proc. Natl. Acad. Sci. (USA), 76: 3829 [1979]. However, other methods of inserting DNA into cells, such as injection into nuclei, electroporation or protoplast fusion, can also be used. E. Culturing the Confusional Cell

Prokaryotic lines used to prepare the mpl ligand polypeptide of the invention are cultured in suitable media as generally described in Sambrook et al., Supra.

The mammalian host cells used to prepare the mpI ligand of the invention can be cultured in a variety of media. Commercially available media such as Ham's F10 (Sigma), Minimal Essential Medium ((MEM), Sigma), RPMI-1640 (Sigma) and Dulbecco's Modified Eagle's Medium ((DMEM), Sigma) are suitable for culturing the host cells. Furthermore, any medium, as described in Ham and Wallace, Meth. Enz., 58:44 [1979], Barnes and Sato, Anal. Biochem., 102: 255. [1980], U.S. Patent 4,767,704; 4,657,866; 4,927,762; Or 4,560,655; WO 90/03430; WO 87/00195; U.S. Patent 30,985 (Reissue), or the pending U.S.S.N. 07 / 592,107 or 07 / 592,141, both filed October 3, 1990, can be used as culture media for the host cells. As required, each of these media can be supplemented with hormones and / or other growth factors (such as insulin, transferrin or epidermal growth factor), salts (such as sodium chloride, calcium, magnesium and phosphate), buffers (such as HEPES), nucleosides (such as adenosine and thymidine) , Antibiotics (such as Gentamycin ™ drug), trace elements (defined as inorganic compounds that are usually present in a final concentration in the micromolar range), and glucose or an equivalent energy source. Any other necessary supplement can also be included with suitable concentrations that are familiar to the person skilled in the art. The culture conditions, such as temperature, pH and the like, are those used previously for the host cell selected for expression and are familiar to those skilled in the art.

The host cells referred to in the present disclosure include cells in microculture as well as cells present in a host animal. F. Evidence of gene amplification / expression

Gene amplification and / or expression can be measured directly in an analysis sample, e.g. by conventional Southern blotting, Northern blotting to quantify the transcription of mRNA (Thomas, Proc. Natl. Acad. Sci. USA, 77: 5201-5205 [1980]), dot blotting (DNA analysis) or by in situ hybridization using a suitable labeled sample, based on the sequences provided directly. Different labels can be used, the most common are radioisotopes, in particular P. However, other techniques can also be used, such as the use of biotin-modified nucleotides for incorporation into a polynucleotide. The biotin then serves as a binding site for avidin or antibodies, which can be labeled with a large variety of labels, such as radionuclides, fluorescent substances, enzymes or the like. Alternatively, antibodies can be used which recognize specific duplex structures, including DNA duplexes, RNA duplexes or DNA-RNA hybrid duplexes or DNA-protein duplexes. The antibodies in turn can be labeled and the assay can be performed with the duplex bound to a surface so that the formation of a duplex on the surface can detect the presence of an antibody bound to the duplex.

Alternatively, gene expression can be measured using immunological methods, such as immunohistochemical staining of tissue sections, and an assay with cell culture or body fluids to directly quantify the expression of the gene product. In immunohistochemical staining techniques, a cell analysis sample is prepared, typically by dehydration and fixation, followed by reaction with labeled antibodies that are specific for the gene product, the labels typically being detectable by consideration, such as enzymatic labels, fluorescent labels, luminescent labels and similar. A particularly sensitive dyeing technique that is suitable for the purposes of the present invention is described by Hsu et al., Am. J. Clin. Path., 75: 734-738 [1980].

Antibodies useful for immunohistochemical staining and / or assay with analytical sample liquids can be either monoclonal or polyclonal and can be raised in any mammal. Typically, the antibodies can be raised against a native mpl ligand polypeptide or against a synthetic peptide based on the DNA sequences provided herein as further described below. G. Purification of mpl Ligand Polypeptide

Mpl ligand is preferably recovered from the culture medium as a secreted polypeptide, although it can also be recovered from host cell lysates if it is directly expressed with no secretory signal.

If mpl ligand is expressed in a recombinant cell other than that of human origin, the mpl ligand is completely free of proteins or polypeptides of human origin. However, it is usually still necessary to free the mpl ligand from other recombinant cell proteins or polypeptides in order to obtain preparations that are essentially homogeneous in terms of the mpl ligand per se. As a first step, the culture medium or lysate centrifuged to remove particulate debris. The membrane and soluble protein fractions are then separated. Alternatively, a commercially available filter for protein concentration (e.g. Amicon or Millipore Pellicon ultrafiltration units) can be used. The mpl ligand can then be purified from the soluble protein fraction and the membrane fraction of the culture lysate, depending on whether the mpl ligand is membrane bound. The mpl ligand is then freed from contaminating soluble proteins and polypeptides by salting out and exchange

Or by chroraatographic methods using different gel matrix materials. These matrix materials include acrylamide, agarose, dextran, cellulose and other common protein purification materials. Exemplary chroraatographic methods that are suitable for protein purification include affinity for affinity (eg anti-hmpl ligand Mab), receptor affinity (eg mpl-IqG or protein-A-Sepha-rose), hydrophobic interaction chromatography (HIC) (eg ether, Butyl- or Phenyl-Toyopearl), lectin chromatography (e.g. Con A-Sepharose, lentil-lectin-Sepharose), size exclusion (e.g. Sephadex G-75), cation and anion exchange columns (e.g. DEAE or carboxyraethyl and sulfo-propyl cellulose ), and reverse phase high pressure liquid chromatography (RP-HPLC) (see, for example, Urdal et al., J. Chroraatog., 295: 171 [1984], where two successive RP-HPLC steps are used to purify recombinant human IL -2). Other cleaning steps may include ethanol precipitation, ammonium sulfate precipitation, chromatofocusing, preparative SDS-PAGE and the like.

Mpl ligand variants in which residues have been deleted, inserted or substituted are obtained in the same way as native mpl ligand, taking into account their significant change in the properties caused by the variation. For example, facilitating mpl ligand fusion with another protein or poly peptide, e.g. a bacterial or viral antigen, the purification, "an immunoaffinity column containing antibodies to the antigen can be used to adsorb the fusion polypeptide. Immunoaffinity columns, such as a column containing rabbit polyclonal anti-mpl ligand serum, can be used to adsorb the mpl- Ligand variant by binding to at least one remaining immunepi-top. Alternatively, the mpI ligand can be purified by affinity chromatography using purified mpl-IqG coupled to (preferably) immobilized resin, such as Affi-Gel 10 (Bio-Rad ,. Richmond, CA) or the like, by means known in the art A protease inhibitor such as phenylmethylsulfonyl fluoride (PMSF) may also be useful in inhibiting proteolytic degradation during cleaning, and antibiotics may be used to promote the growth of to prevent accidental contaminating organisms. Those skilled in the art will recognize that Re Purification methods which are suitable for native mpl ligand may require modification in order to take into account the changes in the property of mpl ligand or its variants by expression in recombinant cell culture. H. Covalent Modifications of mpl Ligand Polypeptide

Covalent modifications of the mpl ligand polypeptide are also covered by the present invention. Both native jnpl ligand and amino acid sequence variants of the mpl ligand can be modified covalently. One type of covalent modification that is encompassed by the present invention is an mpl ligand fragment. Variant mDl ligand fragments with up to approximately 40 amino acid residues can be produced simply by chemical synthesis or by enzymatic or chemical cleavage of the complete or a variant mpl ligand polypeptide. Other types of covalent modifications of the mpl ligand or the fragments thereof are introduced into the molecule by reacting the amino acid residues in question of the mpl ligand or the fragments thereof with an organic derivatizing agent which is with selected side chains or the N or C-terminal residues can react.

Cvsteinyl residues are usually reacted with α-haloacetates (and corresponding amines), such as chloroacetic acid or chloroacetamide, to give carboxymethyl or carboxyamidomethyl derivatives. Cysteinyl residues are also derivatized by the reaction with bromotrifluoroacetone, cx-Biroin ß (5-imidazoyl) propionic acid, chloroacetyl phosphate, N-alkyl-maleimide, 3-nitro-2-pyridyl disulfide, methyl-2-pyridyl disulfide, p-chloro mercuribl , 2-chloro mercuri-4-nitrophenol or chloro-7-nitrobenzo-2-oxa-l, 3-diazole.

Histidyl residues are derivatized by reaction with diethyl pyrocarbonate at pH 5.5-7.0 because this agent is relatively specific for the histidyl side chain. Pairabroni phenacyl bromide is also useful; the reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0.

Lysinyl and amino terminal residues are reacted with succinic or other carboxylic acid anhydrides. The derivatization with these agents has the effect of reversing the charge of the lysinyl residues. Other suitable reagents for derivatizing amino-containing residues include imido esters, such as methylpicolinimidate; Pyridoxal phosphate; Pyridoxal; Chloroborohydride; Trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reaction with glyoxylate.

Arginyl residues are modified by reaction with one or more conventional agents, including phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione and ninhydrin.

The derivatization of arginine residues requires that the reaction be carried out under alkaline conditions because of the high pKa of the guanidine functional group. Furthermore, these reagents can react with the groups of the lysine as well as the epsilon amino group of the arginine.

The specific modification of tyrosyl residues can be carried out, the introduction of spectral markings into the tyrosyl residues being of particular interest, by reaction with aromatic diazonium compounds or

Tetranitromethane. Mostly N-acetylimidazole and tetra-nitroraethane are used to form O-acetyltyrosyl variants or 3-nitroderivatives. Tyrosyl residues are iodinated using 125J or 131J to produce labeled proteins for use in a radioimmunoassay using the chloramine T method described above.

Side carboxyl groups (aspartyl or glutamyl) are selectively modified by reaction with carbodiimides (RN = c = N-R '), in which R and R' represent different alkyl groups, such as i-cyclohexyl-3- (2-morpholinyl-4- ethyl) carbodiimide or l-ethyl-3- (4-azonia-4,4-dimethylpentyl) carbodiimide. Furthermore, aspartyl and glutamyl residues are converted to asparaginyl and glutaminyl residues by the reaction with ammonium ions.

Bifunctional derivatization is useful for crosslinking the mpI ligand with a water-insoluble support matrix or surface for use in the anti-mpl ligand antibody purification process and vice versa. Commonly used crosslinking agents include, e.g., 1,1-bis (diazoacetyl) -2-phenylethane, glutaraldehyde, N-hydroxysuccinimide ester, e.g. Esters with 4-acidosalicylic acid, homobifunctional imido esters including disuccinimidyl esters such as 3,3'-dithiobis (succinimidyl propionate) and bifunctional maleimides such as bis-N-maleimido-1,8-octane.

Derivatizing agents such as methyl-3 - [(p-azido-phenyl) dithio] propioimidate result in photoactivatable intermediates which can form crosslinking structures in the presence of light. Alternatively, reactive water-insoluble matrix materials such as cyanogen bromide activated carbohydrates and the reactive substrates can be described in U.S. Patent 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537 and 4,330,440 can be used for protein immobilization.

Glutaminyl and asparaginyl residues are regularly deamidated to the corresponding glutamyl and aspartyl residues.

These residues are deamidated under neutral or basic conditions. The deamidated form of these residues falls within the scope of the present invention.

Other modifications include the hydroxylation of proline and lysine, phosphorylation of hydroxyl groups from seryl or threonyl residues, methylation of the a-amino group from lysine, arginine and histidine side chains (TE Creighton, Proteins: Structure and Molecular Properties, WH Freeman & Co., San Francisco, Pp. 79-86 [1983]), acetylation of the N-terminal amine and amidation of any C-terminal carboxyl group.

Other types of covalent modification of the mpl ligand polypeptide that are encompassed by the present invention include changing the native glycosylation pattern of the polypeptide. By altering is meant deleting one or more carbohydrate residues found in the native mpl ligand and / or adding one or more glycosylation sites that are not present in the native mpl ligand.

The glycosylation of polypeptides is typically carried out either via an N bond or an O bond. N-linked refers to attaching the carbohydrate residue to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine, in which X represents any amino acid other than proline, are the recognition sequences for the enzymatic attachment of the carbohydrate residue to the asparagine side chain. Thus, the presence of one of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation means adding one of the sugars N-acetylgalactosamine, galacctose or xylose to a hydroxyamino acid, mostly serine or threonine, although 5-hydroxy-proline or 5-hydroxylysine can also be used.

Adding glycosylation sites to the mpl ligand polypeptide is easily accomplished by changing the amino acid sequence to include one or more of the tripeptide sequences (for N-linked glycosylation sites) described above. The change can also be made by adding or replacing one or more serine or threonine residues to the native mpl ligand sequence (for O-linked glycosylation sites). For the sake of simplicity, the mpl ligand amino acid sequence is preferably changed by changes at the DNA level, particularly by mutating the DNA encoding the mpl ligand polypeptide at preselected bases to generate codons that can be translated into the desired amino acids. The DNA mutation (s) can be generated using methods as described above under the heading "Amino acid sequence variants of the mpl ligand".

Another way to increase the number of carbohydrate residues on the mpl ligand is by chemical or enzymatic coupling of glycosides to the polypeptide.

These methods are advantageous in that they do not require the polypeptide to be made in a host cell that has glycosylation capabilities for N- or O-linked glycosylation. Depending on the coupling method used, the sugar (s) can be added to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups, such as those of cysteine, (d) free hydroxyl groups, such as those of serine , Threonine or hydroxyproline, (e) aromatic residues such as those of phenylalanine, tyrosine or tryptophan, or (f) the amide group of glutamine. These methods are described in WO 87/05330, published September 11, 1987, and in Aplin and Wriston, CRC Crit. Rev. Biochem., Pp. 259-306 [1981].

Removal of the carbohydrate residues that occur on the mpl ligand polypeptide can be effected chemically or enzymatically. Chemical deglycosylation requires exposure of the polypeptide to the trifluoromethane sulfonic acid compound or an equivalent compound. This treatment results in the elimination of most or all of the sugars, except the binding sugar (N-acetylglucosamine or N-acetylgalactosamine), while leaving the polypeptide intact v / ïrcj. Chemical deglycosylation is described in Hadimuddin, et al. Arch. Biochem. Biophys., 259: 52 [1987] and Edge et al., Anal. Bochem., 118: 131 [1981]. The enzymatic cleavage of carbohydrate residues from polypeptides can be achieved by using a variety of endo- and exoglycosidases as described by Thotakura et al., Meth. Enzymol., 138: 350 [1987].

Glycosylation at key glycosylation sites can be prevented by using the compound tunicamycin as described by Duskin et al., J. Biol. Chem., 257: 3105 [1982]. Tunicamycin blocks the formation of protein-N-glycoside bonds.

Another type of covalent modification of the mpl ligand involves attaching the mpl ligand polypeptide to a variety of non-proteinaceous polymers, e.g. Polyethylene glycol, polypropylene glycol or polyoxyalkylene, in the manner set forth in U.S. Patent 4,640,835; 4,496,689; 4301.144; 4,670,417; 4,791,192 or 4,179,337.

It can be seen that it is necessary to search the mpl ligand variants obtained in order to select the optimal variant for binding to an mpl and which has the immunological and / or biological activity defined above. One can search for the stability in recombinant cell culture or in plasma (e.g. against proteolytic cleavage), high affinity for an mpl member, oxidative stability, ability to secrete in increased yields and the like. For example, a change in the immunological property of the mpl ligand polypeptide, such as affinity for a given antibody, is measured by a competitive type immunoassay. Other potential modifications of the protein or polypeptide properties, such as redox or heat stability, hydrophobicity or sensitivity to proteolytic degradation, can be tested using known methods. 17. General procedures for the production of antibodies against the mpl ligand

Antibody Production (i) Polyclonal Antibodies

Polyclonal antibodies to mpl ligand polypeptides or fragments are usually generated in animals by multiple subcutaneous (sc) or intraperitoneal (ip) injections of the mpl ligand and an adjuvant. It may also be useful to conjugate the mpl ligand or a fragment containing the target amino acid sequence to a protein that is immunogenic in the manner to be immunized, e.g. Keyhole Limpet Hemocyanin (KLH), serum albumin, beef hyroglobulin or soybean trypsin inhibitor, using bifunctional or derivatizing reagents, e.g. Maleimido-benzoylsulfosuccinimide ester (conjugation over cysteine residues), N-hydroxysuccinimide (over lysine residues), glutaraldehyde, sucinic anhydride, SOCI2 or R1N = C = NR, where R and R1 represent different alkyl groups.

The animals are immunized against the mpl ligand polypeptide or a fragment, immunogenic conjugates or derivatives by combining 1 mg or 1 pg of the peptide or conjugate (for rabbits or mice) with 3 volumes of Freund's complete adjuvant, and injecting the solution intradermally at several locations. A month later, the animals are boosted with 1/5 to 1/10 of the original amount of peptide or conjugate in Freund's complete adjuvant by subcutaneous injection at multiple sites. Seven to fourteen days later, blood is drawn from the animals and the serum is tested for an mpl ligand antibody titer. They are boosted until the titer reaches a plateau. The animals are preferably boosted with the conjugate from the same mpl ligand, but conjugated to a different protein and / or linked to it via a different crosslinking agent. Conjugates can also be produced in recombinant cell culture as protein fusions. Aggregating agents such as alum are also used to increase the immune response.

N (ii) monoclonal antibodies

Monoclonal antibodies are obtained from a population of essentially homogeneous antibodies, i.e. the individual antibodies that make up the population are identical except for possible naturally occurring mutations that may be present in minor amounts. Thus, the modifying term "monoclonal" means that the antibody is not a mixture of individual antibodies.

For example, the monoclonal antibodies to mpl ligand according to the invention can be made using the hybridoma method first described by Kohier & Milstein, Nature, 256: 495 [1975], or they can by recombinant DNA Processes have been prepared (U.S. Patent No. 4,816,567 (Cabilly et al.)).

In the hybridoma method, a mouse or other suitable host, such as a hamster, is immunized as previously described to elicit lymphocytes that can produce or produce antibodies that are specifically directed against the protein used for immunization. Alternatively, lymphocytes can be immunized in vitro. The lymphocytes are then fused to myeloma cells using a suitable fusion agent, such as polyethylene glycol, to form a hybridoma cell line (Goding, Monoclonal Antibodies: Principles and Practice, pp. 59-103 (Academic Press, 1986)).

The hybridoma cells produced in this way are placed in a suitable culture medium and cultured, which preferably contains one or more substances which inhibit the growth or survival of the unfused parent myeloma cells. For example, if the parent myeloma cell lacks the enzyme hypoxanthine guanine phosphoribosyltransferase (HGPRT or HPRT), the culture medium for the hybridomas typically contains hypoxanthine, aminopterin and thymidine (HAT medium), these substances preventing the growth of cells that HGPRT is missing.

Preferred myeloma cells are those that fuse efficiently, support a high level of expression of the antibody by the selected antibody-producing cell, and are sensitive to a medium such as HAT medium. Among these, preferred myeloma cell lines are the mouse myeloma lines, such as those obtained from MOPC-21 and MPC-II mouse tumors, available from the Salk Institute Cell Distribution Center, San Diego, California, USA, and SP-2 cells, available from American Type Culture Collection, Rockville, Maryland, USA. Human myeloma and mouse human heteromyeloma cell lines have also been described for the production of human monoclonal antibodies (Kozbor, J. Immunol., 133: 3001 [1984]; Brodeur et al., Monoclonal Antibody Production Techniques and Applications, S 51-63, Marcel Dekker, Inc., New York, 1987).

The culture medium in which hybridoma cells are cultivated is tested for the production of monoclonal antibodies which are directed against the mpl ligand. The binding specificity of the monoclonal antibodies produced by the hybrido-raacelia is preferably determined by immunoprecipitation or by an in vitro binding assay, such as a radioimmunoassay (RXA) or an enzyme-linked immunoabsorbent assay (ELISA).

The binding affinity of the monoclonal antibody can e.g. are determined using the Scatchard analysis by Munson & Pollard, Anal. Biochem., 107: 220 [1980].

After hybridoma cells have been identified which produce antibodies with the desired specificity, affinity and / or activity, the clones can be subcloned by limiting dilution methods and cultured using standard methods (Goding, see above). Suitable culture media for this purpose include e.g. Dulbecco's Modified Eagle's Medium or RPMI-1640-Medium. Furthermore, the hybridoma cell lines can be cultivated in vivo as ascitic tumors in an animal.

The monoclonal antibodies secreted by the subclones are suitably separated from the culture medium, ascetic fluid or serum by conventional immunoglobulin purification methods such as e.g. Protein A-Sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis or affinity chromatography. DNA that co-encodes the monoclonal antibodies of the invention can be easily isolated and sequenced using conventional methods (e.g., using oligonucleotide samples that can specifically bind to genes that encode the light and heavy chain of mouse antibodies) . The hybridoma cells according to the invention serve as the preferred starting material for this DNA. Once the DNA is isolated, it can be inserted into expression vectors which are then transfected into host cells such as monkey COS cells, Chinese hamster ovary (CHO) cells or myeloma cells that do not otherwise produce immunoglobulin protein to aid synthesis of monoclonal antibodies in the recombinant host cells. The DNA can also be modified, e.g. by inserting the coding sequence for the constant domains of the human light and heavy chain in the place of the homologous mouse sequences (Cabilly et al., see above; Morrison et al., Proc. Nat. Acad. Sci., 81: 6851 [1984]), or by covalently linking the immunoglobulin coding sequence with a part or the complete coding sequence for a non-immunoglobulin polypeptide.

Typically, such non-immunoglobulin polypeptides replace the constant domains of the antibodies of the invention or they replace the variable domain of an antigen binding site of the body of the invention to produce a chimeric bivalent antibody comprising an antigen binding site specific for an mpI ligand and another antigen binding site with specificity for another antigen.

Chimeric or hybrid antibodies can also be made in vitro using known methods in the field of synthetic protein chemistry, including those using crosslinking agents. For example, immunotoxins can be made using a disulfide exchange reaction or by forming a thioether linkage. Examples of suitable reagents for this purpose include iminothiolate and methyl 4-mercaptobutyrimidate. For diagnostic applications, the antibodies according to the invention are typically labeled with a detectable one

Remainder. The detectable remainder can be any remainder that is able to generate a detectable signal either directly or indirectly. For example, the detectable residue may be a radioisotope such as H, C, P, S or 125J, or a fluorescent or chemiluminescent compound such as fluorescein isothiocyanate, rhodamine or luciferin; radioactive isotope labels, e.g. 125J, 32P, 14C or 3H, or an enzyme such as alkaline phosphatase, beta-galactosidase or horseradish peroxidase.

Any prior art method of separately conjugating the antibody to the detectable residue can be used, including the methods described by Hunter et al., Nature, 144: 945 [1962]; David et al., Biochemistry, 13: 1014 [1974]; Pain et al., J. Immunol. Meth., 40: 219 [1981] and Nygren, J. Histochem. and Cytochem., 30: 407 [1982].

The antibodies according to the invention can be used in any known assay method, such as a competitive binding assay, a direct or indirect sandwich assay and an immunoprecipitation assay. Zola, Monoclonal Antibodies: A Manual of Techniques, pp. 147-158 (CRC Press, Inc., 1987).

Competitive binding assays are based on the ability of a labeled standard (which may be an mpl ligand or an immunologically reactive part thereof) with the analyte (mpl ligand) of the test analysis sample to compete for binding with a limited amount of antibodies. The amount of mpl ligand in the test analysis sample is inversely proportional to the amount of standard that is bound to the antibodies. To facilitate the determination of the amount of bound standard, the antibodies are generally insolubilized before or after competition so that the standard and analyte bound to the antibodies are slightly different from the standard and analyte that are not are bound, can be separated.

Sandwich assays involve the use of two antibodies, each capable of binding to a different immunogenic part or epitope of the protein to be detected (mpl ligand). In a sandwich assay, the analyte of the test analysis sample is bound by a first antibody, which is immobilized on a solid support, and then a second antibody binds to the analyte, thus forming an insoluble three-part complex. David & Greene, U.S. Patent 4,376,110. The second antibody itself can be labeled with a detectable residue (direct sandwich assay), or it can be measured using an anti-immunoglobulin antibody which is labeled with a detectable residue (indirect sandwich assay). For example, one type of sandwich assay is an ELISA assay, in which case the detectable residue is an enzyme (e.g. horseradish peroxidase). (iii) Humanized and human antibodies

Methods for humanizing non-human antibodies are known in the art. In general, a humanized antibody has one or more amino acid residues introduced into it from a source that is not of human origin. These non-human amino acid residues are often referred to as "import" residues, which typically come from a variable "import" domain. Humanization can essentially be carried out according to the method of Winter and co-workers (Jones et al., Nature, 321: 522-525 [1986]; Riechmann et al., Nature, 332: 323-327 [1988]; Verhoeyen et al., Science, 239: 1534-1536 [1988]), by replacing a rodent's CDRs or CDR sequences with the corresponding sequences of a human antibody. Accordingly, such "humanized" antibodies are chimeric antibodies (Cabilly et al., See above), in which considerably less than a complete human variable domain has been replaced by the corresponding sequence from a non-human species. In practice, humanized Antibodies typically are human antibodies in which some CDR residues and possibly some FR residues are substituted by residues from analogous sites of rodent antibodies.

The choice of the human variable domains, both light and heavy, to be used for the production of the humanized antibodies is very important to reduce the antigenicity. According to the so-called "best-fit" method, the sequence of the variable domain of a rodent antibody is screened against a complete bank of known human variable domain sequences. The human sequence closest to the rodent sequence is then used as the human framework (FR) for the humanized antibody (Sims et al., J. Immunol., 151: 2296 [1993]; Chothia and Lesk, J. Mol Biol., 196: 901 [1987]).

Another method uses a special framework which is derived from the consensus sequence of all human antibodies of a special subgroup of light or heavy chains. The same framework can be used for several different humanized antibodies (Carter et al., Proc. Natl. Acad. Sci. USA, 89: 4285 [1992]; Presta et al., J. Imrnunol., 151: 2623 [ 1993]).

It is also important that the antibodies be humanized while maintaining high affinity for the antigen or other beneficial biological properties. To achieve this goal, according to a preferred method, a humanized antibody is produced by an analysis method of the parental sequences and various conceptual humanized products using three-dimensional models of the parental and huraanized sequences. Three-dimensional immunoglobulin models are generally available and familiar to the person skilled in the art. Computer programs are available that explain and graphically show the possible three-dimensional conformational structures of selected immunoglobulin sequence candidates. Examining these representations allows analysis of the likely role of the residues in the function of the desired immunoglobulin sequence, i.e. analysis of residues that affect the ability of the desired immunoglobulin to bind to its antigen. In this way, FR residues can be selected from the consensus sequence and combined with the import sequence, so that the desired anti-body property, such as an increased affinity for the target antigen (s), is achieved. In general, the CDR residues are directly and most heavily involved in the influence on antigen binding. For more details, see U.S. Application Serial Number 07 / 934,373, filed August 21, 1992, which is a continuation-in-part application of Serial Number 07 / 715,271, filed June 14, 1991.

Alternatively, it is now possible to produce transgenic animals (e.g. mice) that, after immunization, are able to produce a complete repertoire of human antibodies in the absence of endogenous immunoglobulin production. For example, it has been described that the homozygous deletion of the gene for the antibody heavy chain joining region (JH) leads to a complete inhibition of the endogenous antibody production in chimeric and germline mutated mice. The transfer of the immunoglobulin arrangement from human germline into such germline mutated mice leads to the production of human antibodies when exposed to antigen. See e.g. Jakobovits et al., Proc. Natl. Acad. Sci. USA, 90: 2551-255 [1993]; Jakobovits et al., Nature, 362: 255-258 [1993]; Bruggerman et al., Year in Immuno., 7:33 [1993]. Human antibodies can also be produced in phage banks (Hoogenboom and Winter, J.

Mol. Biol. 227, 381 [1991]; Marks et al., J. Mol. Biol. 222, 581 [1991]). (iv) Bispecific antibodies

Bispecific antibodies are monoclonal, preferably human or humanized, antibodies that have binding specificity for at least two different antigens. Methods for producing bispecific antibodies are known in the prior art.

Traditionally, the recombinant production of bispecific antibodies is based on the coexpression of two pairs of immunoglobulin heavy chains and light chains, the two heavy chains having different specificities (Millstein and Cuello, Nature, 305: 537-539 [1983]). Because of the random pairing of light and heavy immunoglobulin chains, these hybridomas (quadromas) generate a potential mixture of 10 different antibody molecules, only one of which has the correct bispecific structure. The cleaning of the correct molecule, which is usually done by affinity chromatography steps, is quite cumbersome and the product yields are low. Similar methods are disclosed. in PCT publication WO 93/08829 (published 13. Max 1993) and in Traunecker et al., ΞΜΒΟ, 10: 3655-3659 [1991].

According to another and more preferred approach, variable antibody domains with the desired binding specificities (antibody-antigen binding sites) are fused to immunoglobulin sequences for the constant domain. The fusion preferably takes place within a constant domain of the immunoglobulin heavy chain, comprising at least part of the hinge, CH2 and CH3 regions. It is preferred that the first heavy chain constant region (CHI), which contains the site necessary for light chain binding, be present in at least one of the fusions. DNAs encoding the fusions of the immunoglobulin heavy chain and, if desired, the immunoglobulin light chain are inserted into separate expression vectors and cotransfected into a suitable host organism. This allows great flexibility with regard to the setting of the individual portions of the three polypeptide fragments in embodiments, if the unequal ratio of the three polypeptide chains used for the construction enables the optimal exploitation. However, it is possible to insert the coding sequences for two or all three polypeptide chains into an expression vector if the expression of at least two polypeptide chains in the same ratio leads to high yields or if the ratios are of no particular importance. In a preferred embodiment of this approach, the bispecific antibodies are composed of a hybrid immunoglobulin heavy chain with a first binding specificity in one arm, and a hybrid immunoglobulin heavy chain light chain pair (the one second binding specificity) in the other arm. It has been found that this asymmetrical structure facilitates the separation of the desired bispecific compound from the undesired immunoglobulin chain combinations because the presence of only one immunoglobulin light chain in only one half of the bispecific molecule provides an easier route for the separation. This approach is disclosed in co-pending application serial number 07 / 931,811, filed August 17, 1992.

Further details on generating bispecific antibodies are apparently e.g. in Suresh et al., Methods in Enzymology, 121: 210 [1986]. (v) heteroconjugate antibody

Heteroconjugate antibodies are also covered by the present invention. Heteroconjugate antibodies are composed of two covalently linked antibodies. Such antibodies are e.g. have been claimed to direct the immune system cells to undesired cells (U.S. Patent 4,676,980) and to be useful for the treatment of HIV infections (PCT publications WO 91/00360 and WO 92/00373, EP 03089). Heteroconjugate antibodies can be made using any suitable cross-linking method. Suitable crosslinking agents are known in the art and are disclosed in U.S. Patent 4,676,980, along with a variety of crosslinking techniques. IV. Therapeutic use of the megakaryocytopoetic protein mpI ligand

The biologically active mpl ligand with hematopoietic effector function and referred to herein as megakaryocytopoietic or thrombocytopoietic protein (TPO) can be used in a sterile pharmaceutical preparation or formulation to stimulate megakaryocytopoietic or thrombopoietic activity in patients suffering from thioopenia suffer from impaired production, sequestration or increased destruction of platelets. Thrombocytopenia-associated bone marrow hyperplasia (eg aplastic anemia following chemotherapy or bone marrow transplantation) can be effectively treated with the compounds according to the invention, as well as disorders such as extensive intravascular coagulation (DIC), immune thrombocytopenia (including HIV-induced ITP or non-HIV-induced ones) ITP), chronic idiopathic thrombocytopenia, congenital thrombocytopenia, myelodyspla-she and thrombotic thrombocytopenia. Furthermore, these megakaryocytopoietic proteins can be useful for the treatment of myeloproliferative thrombocytotic diseases as well as thrombocytosis as a result of inflammatory conditions and iron deficiency.

Preferred uses of the megakaryocytopoetic or thrombocytopoietic protein (TPO) according to the invention are in connection with myelotoxic chemotherapy, myeloablative chemotherapy and thrombocytopenia due to bone marrow failure.

Other disorders that can be treated in a useful manner with the megakaryocytopoietic proteins of the invention include defects or damage to platelets resulting from drugs, poisoning or activation on artificial surfaces. In these cases the compounds can be used to stimulate the "shedding" of new "undamaged" platelets. For a more complete list of useful applications, see the "Introduction" above, in particular sections (a) - (f) and the references cited therein.

The megakaryocytopoietic proteins according to the invention can be used individually or in combination with other cytokines, hematopoietins, interleukins, growth factors or antibodies for the treatment of the disorders and conditions identified above. Thus, the present compounds can be used in combination with other proteins or peptides with thrombopoietic activity including: G-CSF, GM-CSF, LIF, M-CSF, IL-1, IL-3, erythropoietin (EPO), kit Ligand, IL-6 and IL-11.

The megakaryocytopoietic proteins according to the invention are prepared in a mixture with a pharmaceutically acceptable carrier. This therapeutic composition can be administered intravenously or through the nose or lungs. The composition can also be administered parenterally or subcutaneously, if desired. If the therapeutic composition is administered systemically, it should be pyrogen-free and in a parenterally compatible solution, taking into account the pH, the isotonicity and the stability. These conditions are known to the person skilled in the art. Briefly stated, dose formulations of the compounds according to the invention are prepared for storage and administration by mixing the compound with the desired degree of purity with physiologically compatible carriers, auxiliaries and stabilizers. Such materials are not toxic to the recipient in the doses and concentrations used and include buffers such as phosphate, citrate, acetate and other organic acid salts; Antioxidants such as ascorbic acid; Low molecular weight peptides (less than 10 residues) such as polyarginine, proteins such as serum albumin, gelatin or immunoglobulins; hydrophilic polymers such as polyvinyl pyrrolidone; Amino acids such as glycine, glutaric acid, aspartic acid or arginine; Monosaccharides, disaccharides or other carbohydrates, including cellulose or its derivatives, glucose, mannose or dextrins; Chelating agents such as EDTA; Sugar alcohols such as mannitol or sorbitol; Counterions such as sodium and / or non-ionic surfactants such as Tween, Pluronics or polyethylene glycol.

Approximately 0.5 to 500 mg of a compound or mixture of the megakaryocytopoietic protein in the form of the free acid or base or as a pharmaceutically acceptable salt is processed with a physiologically acceptable vehicle, carrier, auxiliary, binder, preservative, stabilizer, flavoring agent, etc., each according to what pharmaceutical practice demands. The amount of active ingredient in these compositions is chosen so that a suitable dosage is obtained in the range indicated.

Sterile compositions for injection can be formulated according to standard pharmaceutical practice. For example, dissolving or suspending the active compound in a vehicle such as water, or naturally occurring vegetable oils such as sesame, peanut, or construction seed oil, or a synthetic fat vehicle such as ethyl oleate or the like may be desired. Buffers, preservatives, antioxidants and the like can be introduced in accordance with permitted pharmaceutical practice.

Suitable examples of preparations with delayed release include semipermeable matrix materials made of solid hydrophobic polymers containing the polypeptide, the

Matrix materials are in the form of shaped bodies, e.g. Films or microcapsules. Examples of sustained release matrix materials include polyesters, hydrogels (e.g., poly (2-hydroxyethyl methacrylate), as described by Langer et al., J. Biomed. Mater. Res., 15: 167-277 [1981] and Langer, Chem Tech., 12: 98-105 [1982] or poly (vinylalkchol)), polylactide (US Pat. No. 3,773,919, EP 58,481), copolymers of L-glutamic acid and gammaethyl-L-glutamate (Sidman et al., Biopolymers, 22: 547-556 [1983]), non-degradable ethylene vinyl acetates (Langer et al., See above), degradable lactic acid-glycolic acid copolymers, such as the Lupron-Depot ™ (injectable microspheres, which are composed of lactic acid Glycolic acid copolymer and leuprolide acetate), and poly-D - (-) - 3-hydroxybutyric acid (EP 133,988). While polymers such as ethylene vinyl acetate and lactic acid glycolic acid allow the release of molecules over a period of more than 100 days, certain hydrogels release proteins in a shorter period of time. When encapsulated, the proteins remain in the body for an extended period of time, but they can denature or aggregate as the result of exposure to moisture at 37 ° C, resulting in a loss of biological activity and possibly changes results in immunogenicity. Rational strategies can be designed for protein stabilization depending on the mechanism involved. For example, if it is found that the aggregation mechanism is based on intermolecular SS bond formation due to disulfide exchange, the stabilization can be achieved by modifying the sulfhydryl residues, lyophilizing from acidic solutions, controlling the moisture content, using suitable additives and developing specific polymer matrix combinations. settlements.

Delayed release compositions of the megakaryocytopoietic protein also include megakaryocytopoietic frotein packaged in liposomes. Liposomes containing mega-karyocytopoietic protein are produced in a manner known per se: DE 3,218,121; Epstein et al., Proc. Natl. Acad. Sci. USA, 82: 3688-3692 [1985]; Hwang et al., Proc.

Natl. Acad. Sci. USA, 77: 4030-4034 [1980]; EP 52,322; EP 36,676; EP 88,046; EP 143,949; EP 142,641; JP-A-83-118008; U.S. Patents 4,435,045 and 4,544,545; and EP 102,324. Typically, the liposomes are of the small (about 200-800 angstroms) unilamellar type in which the lipid content is greater than about 30 mole% cholesterol, with the optimal level adjusted to be used for megakaryocytopoietic protein therapy is optimal.

The dosage is determined by the attending physician, who takes into account various factors known to affect the effects of active ingredients, including the severity and type of disease, body weight, gender, diet, time and the route of administration, other medications and other relevant clinical factors. Typically, the daily regimen is in the range of 0.1 to 100 mg / kg body weight. The dosage is preferably in the range of 0.1-50 μg / kg body weight. The dosage is particularly preferably in the range of 1-5 jng / kg / day. If necessary, the range of dosing can be the same as for other cytokines, in particular G-CSF, GM-CSF and EPO. Therapeutically effective doses can be determined using either in vitro or in vivo methods.

EXAMPLES

Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and illustrative examples, practice and use the invention to any extent. The following exemplary embodiments therefore explain specific embodiments according to the invention and are not intended to restrict the disclosure in any way. EXAMPLE 1

Partial purification of the Schweiner mpl ligand

Low platelet plasma was collected from normal and aplastically anemic pigs. The pigs were made aplastic by irradiation with 900 cGY for the entire body irradiation using a 4mEV linear accelerator. The irradiated pigs were assisted with intra-muscular injections of cefazolin for 6 to 8 days. The entire blood volume was then removed under general anesthesia, heparinized and centrifuged at 1800 x g for 30 min in order to produce platelet-poor plasma. Mega-karyocyte stimulating activity peaked six days after radiation.

Aplastic pig plasma, obtained from the irradiated pigs, is adjusted to 4M with NaCl and stirred for 30 min at room temperature. The precipitate obtained is removed by centrifugation at 3800 rpm in a Sorvall RC3B and the supernatant is loaded onto a phenyl-Toyopearl column (220 ml) equilibrated in 10 mM NaP04 containing 4M NaCl. The column is washed with this buffer until an A28O of <0.05 is reached and it is eluted with dH2Û. The eluted protein peak is diluted with d ^ O to a conductivity of 15 mS and loaded onto a Blue Sepharose column, equilibrated (240 ml) in PBS. Then the column with 5 column volumes of PBS and 10 mM. NaP04 (pH 7.4) containing 2M urea. The proteins are eluted from the column with 10 mM sodium phosphate (pH 7.4) containing 2M urea and 1M NaCl. The eluted

Protein peak is adjusted to 0.01% octyl glucoside (n-octyl-β-D-glucopyranoside) and 1 mM EDTA and Pefabloc (Boehringer Mannheim) each and directly to tandem-linked CD4-IgG- (Capon, DJ et al., Nature 337: 525-531 (1989)) and mpl-IgG Ultralink (Pierce) columns (see below). The CD4-IgG (2 ml) column is removed after the analysis sample is loaded and the mpl-IgG (4 ml) column is washed with 10 column volumes of PBS and PBS containing 2 M NaCl each and eluted with 0.1M glycine-HCl pH 2.25. The fractions are collected in 1/10 volume 1 M Tris-HCl (pH 8.0).

Analysis of the eluted fractions from the mpl affinity column by SDS-PAGE (4-20%, Novex gel), running under reducing conditions, showed the presence of several proteins (Fig. 5). Proteins that show the greatest intensity with the silver staining separate with an appealing Mr of 66000, 55000, 30000, 28000 and 14000. To determine which of these proteins stimulates the proliferation of Ba / F3-mpi cell cultures , these proteins were eluted from the gel as described in Example 2 below.

Ultralink affinity columns 10-20 mg mpl-lqG or CD4-IgG in PBS are coupled to 0.5 g Ultralink resin (Pierce) as described in the manufacturer's instructions.

Construction and expression of mpl-lqG

A chimeric molecule comprising the full extracellular domain of human mpl (amino acids 1-491) and the Fc region of a human IgGl molecule was expressed in 293 cells. A cDNA fragment encoding amino acids 1-491 of human mpl was obtained by PCR from a human mega-karyocytic CMK cell cDNA library and sequenced.

A Clal site was inserted at the S 'end and a BstEII site at the 3' end. This fragment was inserted upstream of the IgGl-Fc coding region in a bluescript vector between the Clal and BstEII sites after partial cleavage of the PCR product with BstEII because two additional BstEII sites in the DNA that cellular domain encoded by mpl were present. The BstEII site that was introduced at the 3 'end of the mpl PCR product was designed so that the Fc region was in the grid with the mpl extracellular domain. The construct was subcloned into the pRK5-tkneo vector between the Clal and Xbal sites and transfected into 293 human embryonic kidney cells by the calcium phosphate method. These lines were selected in 0.4 mg / ml G418 and individual clones were isolated. The mpl-IgG expression of isolated clones was determined using an ELISA specific for the human Fc. The best expression clone had an expression level of 1-2 mg / ml of mpl-IgG.

Ba / F3 cells expressing Mpl P

A cDNA that corresponded to the complete coding region of hu-man mpl P was cloned into pRK5-tkneo, which was then linearized with Notl and transfected into the IL-3-dependent cell line Ba / F3 by electroporation ( 1 x 107 lines, 9605 F, 250 volts). The selection was started three days later in the presence of 2 mg / ml G418. The rows were selected as pools or individual clones were obtained by limited dilution in 96-well plates. Selected lines were kept in RPMI containing 15% FBS, 1 mg / ml G418, 20 mM glutamine, 10 mM HEPES and 100 pg / ml pen strep. The expression of mpl P in the selected clones was determined by FACS analysis using a polyclonal anti-mpl P rabbit antibody.

Ba / F3 mpl ligand assay

The mpl ligand assay was performed as shown in Figure 2. In order to determine the presence of mpl ligands in various starting materials, the mpl P Ba / F3 cells were starved on IL-3 for 24 hours at a cell density of 5 X 105 cells / ml in a moist incubator at 37 ° C. in 5% CO2 and air. Following the IL-3 starvation, the rows were plated in 96-well culture plates at a density of 50,000 rows in 200 μl medium with or without a diluted analysis sample and cultured for 24 hours in a cell culture incubator. 20 μΐ serum-free RPMI media containing 1 / iCi 3H-thymidine was added to each well for at least 6 to 8 hours. The lines were then harvested on 96-well GF / C filter plates and washed five times with water. The filters were counted in the presence of 40 μΐ of a scintillation fluid (Microscint 20) in a Packard Top Count counter. EXAMPLE 2

Highly purified pig mpl ligand

Gel elution protocol

Equal amounts of the affinity-purified mpl ligand (fraction 6 eluted from the mpl-IgG column) and 2X Laemmli sample buffer were mixed at room temperature without a reducing agent and loaded onto a Novex 4-20% polyacrylamide gel as soon as possible. The sample was not heated. As a control, sample buffer without ligand was run in an adjacent lane. The gel was run for approximately 2 1/4 hours at 4-6 ° C at 135 volts. The running buffer was initially at room temperature. The gel was then removed from the gel box and the plate on one side of the gel.

A replica of the gel was prepared as follows on nitrocellulose: A piece of the nitrocellulose was moistened with distilled water and carefully placed on the surface of the exposed gel in such a way that air bubbles were excluded. Comparative markings were placed on the nitrocellulose and the gel plate so that the replica filter could be brought back into exactly the same position after staining. After approximately 2 minutes, the nitrocellulose was carefully removed and the gel was wrapped in a plastic wrap and placed in the refrigerator. The nitrocellulose was stained with Biorad total gold protein stain by first in 3 x 10 ml 0.1% Tween 20 + 0.5 M NaCl + 0.1 M Tris-HCl pH 7.5 for approximately 45 min and then · agitated in 3 x 10 ml of purified water for 5 min. The gold stain was then added and left to develop until the bands in the standard markings became visible. The replica filter was then rinsed with water, placed on the gel over the plastic cover, and carefully aligned with the reference marks. The positions of the Novex standard markers were marked on the gel plate and lines were drawn to indicate the cutting positions. The nitrocellulose and plastic wrap were then removed and the gel cut along the lines shown with a sharp razor blade. The sections were extended beyond the sample tracks so that they could be used to determine the positions of the pieces after the gel was stained. After the pieces were removed, the rest of the gel was stained silver and the positions of the standard markers and the cut marks were measured. The molecular weights corresponding to the cutting positions were determined on the basis of the Novex standard markers.

The twelve pieces of gel were placed in the rows of two Biorad Model 422 electroelution devices. 12-14 K molecular weight exclusion membrane covers were used in the rows. 50 mM ammonium bicarbonate + 0.05% SDS (approximately pH 7.8) was the elution buffer. 1 liter of the buffer was chilled in a 4-6 ° C refrigerator for approximately 1 hour before use. The gel pieces were eluted at 10 mA / cell (initially 40 V) in a 4-6 ° C cold room. The elution took approximately 4 hours. The lines were then carefully removed and the liquid over the frit removed with a pipette. The elution chamber was removed and any fluids above the membrane lid were removed with a pipette. The liquid in the membrane cover was removed with a pipetting aid (Pipetman) and stored. 50 μΙ aliquots of purified water were then placed in the lid, agitated and removed until all SDS crystals were dissolved. These washing steps were combined with the above-mentioned stored liquid. The total elution sample volume was 300-500 jul per gel slice. The samples were placed in 10 mm Spectrapor 4 12-14 K exclusion dialysis tubes which had been wetted for several hours in purified water. They were dialyzed overnight at 4-6 ° C against 600 ml of phosphate buffered saline (PBS is not about 4 mM in terms of potassium) per 6 samples. The buffer was replaced the next morning and dialysis continued for 2.5 hours. The samples were then removed from the dialysis bags and filled into microcentrifuge tubes. The tubes were placed on ice for 1 h, microcentrifuged at 14K rpm for 3 min, and the supernatants carefully removed from the precipitated SDS. The supernatants were then placed on ice for about another hour and microcentrifuged again for 4 minutes. The supernatants were diluted in phosphate-buffered saline and prepared for the activity assay. The remaining samples were frozen at -70 ° C. EXAMPLE 3

Microsequencing of the porcine mpl ligand

Fraction 6 (2.5 ml) from the mpl-IgG affinity column was concentrated in a Microcon-10 (Amicon). In order to prevent the mpl ligand from absorbing on the microcon, the membrane was rinsed with 1% SDS and 5 μΐ 10% SDS were added to fraction 6. Sample buffer (20 µΐ) of 2X was added to fraction # 6 after Microcon concentration (20 µΐ) and the entire volume (40 µΐ) was loaded onto a single trace of 4-20% gradient acrylamide gel (Novex) . The gel was run according to the Novex protocol. The gel was then equilibrated for 5 min before electroblotting in 10 mM 3- (cyclohexylamino) -l-propanesulfonic acid (CAPS) buffer, pH 11.0, containing 10% methanol. Electroblotting on Immobilon PSQ membranes (Millipore) was carried out for 45 min at a constant current strength of 250 mA in a BioRad trans-blot transfer cell (32). The PVDF-Merabran was stained with 0.1% Coomassie Blau R-250 in 40% methanol, 0.1% acetic acid for 1 min and decolorized for 2-3 min with 10% acetic acid in 50% methanol. The only proteins that were visible in the 18000-35000 molecular weight region of the blot had MWs of 30,000, 28,000 and 22,000.

The 30, 28 and 22 kDa bands were subjected to protein sequencing. Automated protein sequencing was performed using a Model 470A Applied Biosystem Sequencer, which was scanned with an on-line PTH analyzer. The sequencer was modified so that 80-90% of the sample could be injected (Rodriguez, J. Chromatogr., 350: 217-225 [1985]). Acetone (approximately 12 μΐ / ΐ) was added to solvent A to compensate for uv absorption. The electro-blotted proteins were sequenced in the blott cartridge. D. The peaks were integrated with Justice Innovation software using Nelson Analytical 970 interfaces. The sequence interpretation was carried out on a VAX 5900 (Henzel et al., J. Chromatogr., 404: 41-52 [1987]). N-terminal sequences (using the one-letter code with unsafe residues in

Brackets) and the amount of material obtained (in brackets) is shown in Table 2 '. TABLE 2 '

Sequences of the amino terminus from the mpl ligand

EXAMPLE 4

Liquid suspension megakaryocytopoiesis assay

Human peripheral stem cells (PSC) (obtained with approval from patients) were diluted 5-fold with IMDM medium (Gibco) and centrifuged for 15 min at room temperature at 800 x g. The cell pellets were resuspended in IMDM, layered on 60% Percoll (density 1.077 g / ml) (Pharmacia) and centrifuged at 800 x g for 30 min. The low-density mononuclear cells were aspirated at the intermediate phase, washed twice with IMDM and seeded at 1-2 x 106 cells / ml in IMDM containing 30% FBS (1 ml final volume) in tissue culture dishes with 24 wells (Costar) . APP or mpl ligand-free APP was added at 10% and the kuituren were grown for 12 ^ -14 days in a humidified incubator at 37 ° C in 5% CO 2 and air. The kuituren were also grown in the presence of 10% APP with 0.5 pq mpl-IqG, added on days 0, 2 and 4. APP was freed from pre mpl ligands by passing APP through an mpl IgG affinity column.

In order to quantify the megakaryocytopoiesis in these liquid suspension cultures, a modification by Solberg et al. which uses a radiolabeled mouse IgG monoclonal antibody (HP1-1D) against GPIIbllla (provided by Dr. Nichols, Mayo Clinic). 100 µg HP1-1D (see Grant, B. et al., Blood 69: 1334-1339 [1987]) was radioactively labeled with 1 mCi Na125I using Enzymo-beads (Biorad, Richmond, Ca) as by the Instructions of the manufacturer described. Radioactively labeled HP1-1D was stored at -70 ° C in PBS containing 0.01% octyl glucoside. Typical specific activities were 1-2 x 105 cpm / Mg (> 95% precipitated by 12.5% trichloroacetic acid).

The liquid suspension cultures were prepared in triplicate for each experimental point. After 12-14 days in culture, 1 ml cultures were transferred to 1.5 ml Eppendor tubes and centrifuged at 800 xg for 10 min at room temperature, and the resulting cell pellets were placed in 100 μΐ PBS containing 0.02% EDTA and 20% Calf serum, resuspended. 10 ng 125J-HP1-1D in 50 μΐ assay buffer were added to the resuspended kuituren and incubated for 60 min at room temperature (RT) with occasional shaking. The lines were then collected by centrifugation at 800 x g for 10 min at room temperature and washed 2 × with assay buffer. The pellets were counted for 1 min in a gamma counter (Packard). Non-specific binding was determined by adding 1 µg of unlabeled HP1-1D for 60 min before adding the labeled HP1-1D. Specific binding was determined as the value resulting from total 125J-HP1-1D binding minus the value resulting from 125J-HP1-1D binding in the presence of excess unlabelled HP1-1D ' . EXAMPLE 5

Oligonucleotide PCR primer

Based on the amino-terminal amino acid sequence obtained from the 30 kDa, 28 kDa and 18-22 kDa proteins, degenerate oligonucleotides were designed for use as a polymerase chain reaction (PCR) primer (see Table 4). Two primer pools were synthesized, a sense strand 20-mer pool coding for amino acid residues 2-8 (mpl 1) and an anti-sense strand 21-mer pool, complementary to the sequences that are required for amino acids 18-24 (mpl 2) code. . TABLE 4

Degenerate oligonucleotide primer pools

Pig genomic DNA isolated from porcine peripheral blood lymphocytes were used as a template for the PCR. The 50 μΙ reaction contained: 0.8 μρ pig genomic DNA in 10 mM Tris-HCl (pH 8.3), 50 mM KC1, 3 itiM MgCl2, 100 Mg / ml'BSA, 400 μΜ dNTPs, 1 μΜ of each primer pool and 2.5 units of Taq polymerase. The initial template denaturation was carried out for 8 min at 94 ° C, followed by 35 cycles of 45 s each at 94 ° C, 1 min at 55 ° C and 1 min at 72 ° C. The last cycle was allowed to extend at 72 ° C for 10 min. The PCR products were separated by electrophoresis in a 12% polyacrylamide gel and visualized by staining with ethidium bromide. It was concluded that if the amino terminal amino acid sequence was encoded by a single exon, the correct PCR product is believed to be 69 bp in length. A DNA fragment of this size was eluted from the gel and subcloned into pGEMT (Promega). Sequences of three clones are shown in Table 5 below. TABLE 5 69 bp pig genomic DNA fragments

The position of the PCR primers is indicated by the underlined bases. These results verify the N-terminal sequence obtained for amino acids 9-17 for the 30 kDa, 28 kDa and 18-22 kDa proteins and showed that this sequence is through a single exon of porcine DNA is encoded. EXAMPLE 6

Human mpl ligand genes

Based on the results of Example 5, a 45-mer deoxyoligonucleotide called pR45 was designed and synthesized to screen a genomic library. The 45-mer had the following sequence:

(SEQ ID NO: 28). 32. 3 * 5

This oligonucleotide was P-labeled with (γ P) -ATP and T4 ~ kinase and used to screen a human genomic DNA library in Xgeml2 under low stringency hybridization and washing conditions (see Example 7). Positive clones were removed, plaque-cleaned and analyzed by restriction mapping and Southern blotting. Clone # 4 was selected for additional analysis.

A 2.8 kb BamHI-Xbal fragment that hybridized to the 45-mer was SK subcloned into pBluescript. Partial DNA sequencing of this clone was carried out using oligonucleotides specific for the porcine mpl ligand DNA sequence as primers. The sequence obtained confirmed that DNA encoding the human homologous mpl ligand of the porcine mpl ligand was isolated. An EcoRI restriction site was discovered in the sequence that allowed us to isolate a 390 bp EcoRI-Xbal fragment from the 2.8 kb BamHI-Xbal and subclone it into pBluescript SK-.

Both strands of this fragment were sequenced. The human DNA sequence and the deduced amino acid sequence are shown in Fig. 9 (SEQ ID NO: 3 & 4). The predicted positions of introns in the genomic sequence are through

Arrows indicated and define a putative exon ("Exon 3").

Study of the predicted amino acid sequence confirms that a serine residue is the first amino acid of the mature mpl ligand, as determined by direct amino acid analysis. Immediately upstream of this codon, the predicted amino acid sequence strongly suggests a signal sequence that plays a role in the secretion of the mature mpl ligand. This region coding for a signal sequence is probably interrupted at nucleotide position 68 by an intron.

In the 3 'direction, the exon appears to end at nucleotide 196. This exon therefore codes for a sequence of 42 amino acids, 16 of which are probably part of a signal sequence and 26 part of the mature human mpl ligand. EXAMPLE 7

Human full length mpl ligand cDNA

Based on the human "exon 3" sequence (Example 6), two non-degenerate oligonucleotides corresponding to the 3 'and 5' ends of the "exon 3" sequence were synthesized (Table 6). TABLE 6

Non-degenerate PCR oligonucleotide primers for human cDNA

These two primers were used in PCR reactions using DNA from different human cDNA libraries or 1 ng Quick Clone cDNA (Clonetech) from different tissues as template using the conditions described in Example 5. The expected size of the correct PCR product was 140 bp. After analysis of the PCR products in a 12% polyacrylamide gel, a DNA fragment of the expected size was discovered in cDNA libraries which had been produced from adult kidney and 293-fetal kidney cells and cDNA which had been prepared from human fetal life (Clonetech Catalog No. 7171-1).

A fetal liver cDNA library in λ DR2 (Clonetech Catalog No. HL1151X) was screened with the same 45-mer oligonucleotide that was used to screen the human genomic library. The oligonucleotide was labeled with (γ32Ρ) -ΑΤΡ using T4 polynucleotide kinase. The bank was screened under low stringency hybridization conditions. The filters were prehybridized for 2 hours, then with the sample overnight at 42 ° C. in 20% formamide, 5xSSC, 10xDenhardt's, 0.05M sodium phosphate (pH 6.5), 0.1% sodium pyrophosphate, 50 mg / ml sonicated salmon sperm DNA hybridized for 16 hours. The filters were then rinsed in 2xSSC and then washed once in 0.5xSSC, 0.1% SDS at 42 ° C. The filters were exposed to Kodak X-ray film overnight. Positive clones were removed, plaque cleaned, and insert size was determined by PCR using oligonucleotides flanking the BamHI-Xbal cloning site in λ DR2 (Clonetech Catalog No. 6475-1). 5 μΐ phage stock solution was used as the template source. The initial denaturation was carried out at 94 ° C for 7 min, followed by 30 cycles of amplification (1 min at 94 ° C, 1 min at 52 ° C and 1.5 min at 72 ° C). The last extension was carried out at 72 ° C for 15 min. Clone No. FL2b had a 1.8 kb insert and was selected for further analysis.

The plasmid pDR2 (Clonetech, XDR2 & pDR2 Cloning and Expression System Library Protocol Handbook, p. 42), contained within the kDR2 phage arms, was obtained (rescued) as described by the manufacturer's instructions (Clonetech, XDR2 & pDR2 Cloning and Expression System Library Protocol Handbook, pp. 29-30). Restriction analysis of plasmid pDR2-FL2b with BamHI and Xbal indicated the presence of an internal BamHI restriction site approximately at position 650 in the insert. Digestion of the plasmid with BamHI-Xbal cut the insert into two fragments, one of 0.65 kb and one of 1.15 kb. The DNA sequence was determined using three different classes of templates derived from the plasmid pDR2-FL2b. DNA sequencing of the double-stranded plasmid DNA was carried out using the ABI373 (Applied Biosystems, Foster City, California) automated fluorescence DNA sequencer using the standard protocol for dye-labeled di-deoxynucleoside triphosphate terminators (dye Terminators) and conventional synthesized walking primers (Sanger et al., Proc. Natl. Acad. Sci. USA, 74: 5463-5467 [1977]; Smith et al., Nature, 321: 674-679 [1986 ]). Direct sequencing of the fragments of the polymerase chain reaction amplified from the plasmid was carried out by means of the ABI373 sequencer using conventional primers and dye terminator reactions. The single-stranded template was constructed with the M13 Janus vector (DNASTAR, Inc., Madison, Wisconsin) (Burland et al., Nucl. Acids Res., 21: 3385-3390 [1993]). BamHI-Xbal (1.15 kb) and BamHI (0.65 kb) fragments were isolated starting from the plasmid pDR2-FL2b, the ends were filled in with T4-DNA polymerase in the presence of deoxynucleotides and then into the Smal site of the M13 janus subcloned. Sequencing was carried out using standard protocols for dye-labeled M13 universal and reverse primers or walking primers and dye terminators. Manual sequencing reactions were performed on single strand M13 DNA using walking primer and standard dideoxy termination chemistry (Sanger et al., Proc. Natl. Acad. Sci. USA, 74: 5463-5467 (1977)) , 33P-labeled a-dATP and Sequenase (United States Biochemical Corp., Cleveland, Ohio). DNA sequence assembly was performed using Sequencher V2.Ibl2 (Gene Codes Corporation, Ann Arbor, Michigan). The nucleotide and derived sequences of the hML are provided in Fig. 1 (SEQ ID NO: 1). EXAMPLE 8

Isolation of the human mpl ligand (TPO) gene

Human genomic DNA clones of the TPO gene were obtained by screening a human genomic library in X-Geml2 with pR45, an oligonucleotide sample described above, under conditions of low stringency (see Example 7) or under conditions of high stringency with a fragment, which isolates the 3 'half of the human cDNA coding for the mpl ligand (from the BamHI site to the 3' end). Two overlapping λ clones spanning 35 kb were isolated. Two overlapping fragments (BamHÏ and EcoRI) containing the entire TPO gene were subcloned and sequenced. The structure of the human gene is composed of 6 exons within 7 kb of genomic DNA (FIGS. 14A, B and C). The surroundings of all exon / intron linkages are compatible with the consensus motif established for mammalian genes (Shapiro, M.B., et al., Nucl. Acids Res. 15: 7155 (1987)).

Exon 1 and exon 2 contain a 5 'untranslated sequence and the first four amino acids of the signal peptide. The rest of the secretion signal and the first 26 amino acids of the mature protein are encoded within exon 3. The entire carboxyl domain and both the 3'-untranslated and approximately 50 amino acids of the erythro-poietin-like domain are encoded within exon 6. The four amino acids involved in the deletion observed within hML-2 (hTPO-2) are encoded at the 5 'end of exon 6. EXAMPLE 9

Transient Expression of the Human mpl Ligand (hML)

To subclone the full length insert contained in pDR2-FL2b, the plasmid was digested entirely with Xbal, then partially with BamHI. A DNA fragment corresponding to the 1.8 kb insert was gel purified and in pRK5 (pRK5-hmpl I) (see US Pat. No. 5,258,287 for the construction of pRK5) under the control of the immediate-ear ly promoter of the cytomegalovirus. DNA of the construct pRK5-hmpl I was prepared using the PEG method and in human embryonic kidney 293 cells, which were supplemented with F-12 nutrient mixture, 20 mM Hepes (pH 7 in Dulbecco's modified Eagle1s medium (DMEM) , 4) and 10-i fetal bovine serum, -transf ected. The lines were transfected by the calcium phosphate method as described (Gorman, C. (1985) in DNA Cloning: A Practical Approach (Glover, DM, ed.) Vol. II, pp. 143-190, IRL Press , Washington, DC). 36 hours after the transfection, the supernatant of the transfected lines was checked for activity in the proliferation assay (see Example 1). The supernatant of the 293 cells transfected only with the pRK vector showed no stimulation of the Ba / F3 or Ba / F3 mpl lines (FIG. 12A). The supernatant from the lines transfected with the pRK5-hmpl I had no effect on the Ba / F3 cells, but dramatically stimulated the proliferation of the Ba / F3-mpl cells (Fig. 12A), showing that this cDNA is functional for one encoded active human mpl ligand. EXAMPLE 10

Human mpl ligand isoforms hML2, hML3 and hHL4

To alternatively identify spliced forms of hML, primers were synthesized that correspond to each end of the coding sequence of hML. These primers were used in RT-PCR to amplify human RNA from adult liver. In addition, internal primers that flank selected areas of interest (see below) were constructed and used similarly. The direct sequencing of the ends of the PCR product revealed a single sequence which corresponded exactly to the sequence of the cDNA which had been isolated from the human bank from fetal liver (see FIG. 1 [SEQ ID NO: 1]) . However, a region near the C-terminus of the EPO domain (in the middle of the PCR product) showed a complex sequence pattern, which indicates the existence of possible splice variants in this region. In order to isolate these splice variants, the primers, which are provided in Table 7 and flank the area of interest, were used in a PCR as templates for human cDNA from adult liver. TABLE 7 PCR primer for human ML isoform

The PCR products were subcloned with blunt ends in M13. The sequencing of individual subclones opened the existence of at least three ML isoforms. One of these, hML (also referred to as Î1ML332), is the longest form and corresponds exactly to the sequence that has been isolated from the bank of the f-liver. The sequences of the four human mpl ligand isoforms, listed from the longest (hML) to the shortest (hML-4) isoform, are provided in Fig. 11 (Fig. 11 [SEQ ID NO: 6, 8, 9 & 10]). EXAMPLE 11

Construction and transient expression of the human mpl ligand isoforms and substitution varxants hML2, hML3 and hML (R153A, R154A)

The isoforms hML2 and hML3 and the substitution variant hML (R153A, R154A) were based on hML using the methods of Russell Higuchi, in PCR Protocols, A guide to Methods and Applications, Acad. Press, M.A. Innis, D.H. Gelfand, J.J. Sninsky & T.J. White ed. Described reconstituted PCR technique reconstituted.

In all constructs, the "outside" primers used are shown in Table 8 and the "overlapping" primers in Table 9. TABLE 8 Outside primers

TABLE 9 Overlapping primers

All PCR amplifications were performed with cloned Pfu DNA polymerase (Stratagene) using the following conditions: Initial template denaturation was carried out at 94 ° C for 7 min, followed by 30 cycles of 1 min at 94 ° C , 1 min at 55 ° C and 1.5 min at 72 ° C. The last cycle was allowed to extend at 72 ° C for 10 min. The final PCR product was digested with Clal-Xbal, gel purified and cloned into pRK5tkneo. 293 cells were transfected with the various constructs as described above and the supernatant was examined using the Ba / F3-mpI proliferation assay. hML-2 and hML-3 showed no detectable activity in this assay, but the activity of hML (R153A, R-154A) was similar to hML, which shows that processing at this di-basic site does not require activity (see Fig. 13). EXAMPLE 12

Mouse mpl ligand cDNA mML, mML-2 and mML-3

Isolation of the mML cDNA.

A DNA fragment corresponding to the entire coding region of the human mpl ligand was obtained by PCR, gel-purified and labeled by "random priming" in the presence of 32P-dATP and 32P-dCTP. This sample was used to screen 106 clones of a mouse liver cDNA library in XGTIO (Clontech Catalog No. ML3001a). Filter duplicates were in 35% formamide, 5xSSC, lOxDenhardt's, 0.1% SDS, 0.05M sodium phosphate (pH 6.5), 0.1% sodium pyrophosphate, 100 / xg / ml sonicated salmon sperm DNA overnight in the presence hybridizes the sample. The filters were rinsed in 2xSSC and then washed once in 0.5xSSC, 0.1% SDS at 42 ° C. The hybridizing phage was plaque purified and the cDNA inserts were subcloned into the EcoRI site of the Bluescript SK plasmid. The clone "LD" with a 1.5 kb insert was chosen for further analysis and both strands were sequenced as described above for the human ML cDNA. The nucleotide and deduced amino acid sequence of clone LD are provided in Figure 14 (SEQ ID NO: 1 & 11). The derived mature ML sequence from this clone was 331 amino acid residues long and was designated 111ML331 (or mML-2 for the reasons described below). Considerable identity in terms of both nucleotide and deduced amino acid sequences was observed in the EPO-like domains of these ML's. However, after the deduced amino acid sequences of the human and mouse ML's were aligned, the mouse sequence appeared to have a tetrapeptide deletion between the human residues 111-114 that corresponds to the 12 nucleotide long deletion that corresponds to nucleotide position 618 follows and has been seen in both human (see above) and porcine (see below) cDNA's. Accordingly, additional clones were examined to discover possible mouse ML isoforms. One clone, "L7", had a 1.4 kb ^ insert with a derived sequence of 335 amino acids containing the "missing" tetrapeptide LPLQ. This shape is probably the full-length mouse ML and is referred to as mML or Ï11ML335. The nucleotide and deduced amino acid sequence for mML are provided in Figure 16 (SEQ ID NO: 12 & 13). Ultimately, clone "L2" was isolated and sequenced. This clone had the 116 nucleotide long deletion corresponding to hML3 and is therefore called mML-3. A comparison of the deduced amino acid sequences of these two isoforms is shown in FIG.

Expression of the recombinant mML.

Expression vectors for mouse ML were prepared essentially as described in Example 8. Clones encoding mML and mML-2 were subcloned into pRK5tkneo, a mammalian expression vector which enables expression under the control of the CMV promoter and an SV40 polyadenylation signal. The resulting expression vectors, mMLpRKtkneo and mML2pRKtkneo, were transiently transfected into 293 cells using the calcium phosphate method. After the transient transfection, the medium was conditioned for 5 days. The lines were kept in high glucose DMEM media supplemented with 10% fetal calf serum.

Expression of mouse mpl (mmpl) in Ba / F3 lines.

Stable cell lines expressing c-mpl were obtained by transfection of vanpl pRKtkneo, essentially as described for human mpl in Example 1. An expression vector (20 pg; linearized) containing the entire coding sequence of the mouse mpI (Skoda, RC, et al., EMBO J. 12: 2645-2653 (1993)) was generated by electroporation (5 x 10 lines, 250 Volt, 960 pF) in Ba / F3 cells, followed by a selection for neomycin resistance with 2 mg / ml G418. Expression of mpl was determined by flow cytometry analysis using rabbit anti-mouse mpl IgG antisera. The Ba / F3 cells were kept in RPMI 1640 medium from WEHI-3B lines as the source for IL-3. The supernatants from 293 lines transfected with both mML and mML-2 were examined in BaF3 cells transfected with both mmpl and hmpl, as described in Example 1. EXAMPLE 13 Pig pML and pML-2 mpl ligand cDNA

Porcine ML (pML) cDNA was isolated by RACE PCR. An oligo-dT primer and two specific primers were designed based on the sequence of the exon of the porcine ML gene encoding the amino terminus of the porcine serum-isolated ML gene. CDNA produced from various aplastic pig tissues was obtained and amplified. A 1342 bp PCR cDNA product was found in kidney and subcloned. Multiple clones were sequenced and found to encode the mature porcine mpl ligand (not comprising a complete secretion signal). The cDNA for a mature protein of 332 amino acids (PML332) was found to encode with the sequence shown in Figure 18 (SEQ ID NO: 9 & 16).

Method:

Isolation of the pML gene and cDNA.

Genomic clones of the porcine ML gene were isolated by screening a porcine genomic bank in EMBL3 (Clontech Inc.) with pR 45. The bank was screened essentially as described in Example 7. Several clones were isolated and the exon coding for the amino acid sequence identical to that obtained from purified ML was sequenced. Porcine ML cDNA was obtained using a modification of the RACE PCR protocol. Two specific ML primers were designed based on the sequence of the porcine ML gene. Polyadenylated mRNA was isolated from the kidneys of aplastic pigs essentially as described above. The cDNA was prepared by reverse transcription with the BamdT primer (BamdT: 5 'GACTCGAGG ATCCATCGAI il I I I I I I I I I I' ITT 3 ') (SEQ ID NO: 55), which is directed against the polyadenosine tail of the mRNA. An initial round of PCR amplification (28 cycles of 95 ° C for 60 seconds, 58 ° C for 60 seconds and 72 ° C for 90 seconds) was performed using the ML-specific h-forward-1 printer (h -forward-1: 5 'GCTAGCTCTAGAAATTGCTCCTCGTGGTCATGCTTCT 3) (SEQ ID NO: 43) and the BAMAD primer (BAMAD: 5' GACTCGAGGATCCATCG 3 ') (SEQ ID NO: 56) in a 100 ml reaction (50 mM KC1, 1 , 5 mM MgCl, 10 mM Tris pH 8.0, 0.2 mM dNTPs with 0.05 U / ml araplitaq polymerase [Perkin Elmer Inc.]). The PCR product was then digested with Clal, extracted with phenol-chloroform (1: 1), ethanol-precipitated and to 0.1 mg Bluescript SK vector (Stratagene Inc.), which had been cut with Clal and Kpnl was ligated. After two hours of incubation at room temperature, a quarter of the ligation mixture was directly transferred to a second round of PCR (22 cycles as described above) using a second ML-specific forward-1 primer (forward-1: 5 'GCTAGCTCTAGAAGCCCGGCTCCTCCTGCCTG 3 ') (SEQ ID NO: 57) and T3-21 (an oligonucleotide that binds to a sequence which is adjacent to the multiple cloning region within the Bluescript SK vector) (5' CGAAATTAACCCTCACTAAAG 3 ') (SEQ ID NO: 58 ). added. The resulting PCR product was digested with Xbal and Clal and subcloned into Bluescript SK-. Several kions from independent PCR reactions were sequenced.

Again, a second form, designated pML-2, was identified which encodes a protein with a 4 amino acid residue long deletion (328 amino acid residues) (see Figure 21 [SEQ ID NO: 21]). A comparison of the pML and pML-2 amino acid sequences shows that the latter form is identical, with the exception that the tetrapeptide QLPP, which the

Residues 111-114 inclusive, has been deleted (see Fig. 22 [SEQ ID NO: 18 & 21]). The four amino acid long deletion observed in mouse, human and porcine ML cDNA occurs in exactly the same position within the predicted proteins.

Example 14 CMK Assay for Thrombopoietin (TPO) Induction of Platelet Antigen GPIIblIIa Expression

The CMK cells are kept in RMPI 1640 medium (Sigma) supplemented with 10% fetal bovine serum and 10 mM glutamine.

To prepare the assay, the lines are harvested, washed and resuspended to 5 × 10 5 cells / ml in serum-free GIF medium, supplemented with 5 mg / 1 bovine insulin, 10 mg / 1 apo-transferrin, 1 X trace elements. In a flat-bottomed tissue culture plate with 96 wells, the TPO standard or experiment samples are added to each well in suitable dilutions in 100 μΙ volumes. 100 μΐ of the CMK cell suspension is added to each well, and the tissue culture plates are incubated at 37 ° C. in a 5% CO 2 incubator for 48 hours. After incubation, the plates are centrifuged at 1000 rpm at 4 ° C for 5 minutes.

The supernatants are discarded and 100 μl of the FITC-conjugated GPIIblIIa monoclonal 2D2 antibody are added to each well. After incubation at 4 ° C for 1 hour, the plates are again centrifuged at 1000 rpm for 5 minutes. The supernatants containing unbound antibody are discarded and 200 µl 0.1% BSA-PBS wash solution is added to each well. The washing step with 0.1% BSA-PBS is repeated three times. The lines are then analyzed in a FASCAN using a standard one parameter analysis, measuring the relative fluorescence intensity.

Example 15 DAMI Assay for Thrombopoietin (TPO) by Hesse's Endo-Mitotic Activity of DAMI Cells in 96-well Microtiter Plates DAMI Cells are Replenished with 10 in IMDM + 10% Horse Serum (Gibco) mM glutamine, 100 ng / ml penicillin G and 50 μg / ml streptomycin. To prepare the assay, the lines are harvested, washed and resuspended at 1x10 cells / ml in IMDM + 1% horse serum. In a plate with 9 6 wells with round bottoms, 100 μ D of the TPO standard or experiment samples are added to the DAMI cell suspension. The lines are then incubated for 48 hours at 37 ° C in a 5% CO 2 incubator. After incubation, the plates are centrifuged in a Sorvall 6000B centrifuge at 1000 rpm for 5 minutes at 4 ° C. The supernatants are discarded and the washing step with 200 μl PBS-0.1% BSA is repeated. The lines are fixed by the addition of 200 μΐ ice-cold 70% ethanol-PBS and resuspended by suction. After incubation at 4 ° C for 15 minutes, the plates are centrifuged at 2000 rpm for 5 minutes and 150 μl of 1 mg / ml RNAse, containing 0.1 mg / ml propidium iodide and 0.05% Tween-20, are added to each well . After an hour's incubation at 37 ° C, the changes in DNA content are measured by flow cytometry. Polyploïdie is measured and quantified as follows:

Normalized polyploid rate (NPR) = (% cells in> G2 -> - M /% cells in <G2 + M) with TPO (% cells in> G2 + M /% cells in <G2 + M) in control

Example 16

Thrombopoietin (TPO) in vivo assay (mouse platelet rebound assay), jjj vivo assay for determining platelet production. C57BL6 mice (obtained from Charles River) are intraperitoneally (IP) with 1 ml goat anti-mouse platelet Serum (6 amps) injected on day 1 to produce thrombocytopenia.

On days 5 and 6, the mice were given two IP injections of the factor or PBS as a control. On day 7, 30 βCi Na235S04 in 0.1 ml saline is injected intravenously and the percentage of 35S incorporation of the injected dose into circulating platelets is measured in the blood samples obtained from the treated and control mice. The platelet counts and white blood cell counts are performed at the same time in blood obtained from the retro-orbital sinus.

Example 17 KIRA ELISA for thrombopoietin (TPO) by measuring the phosphorylation of the chimeric iupl-Rse.gD receptor

The human mpl receptor has been described by Vigon et al., PNAS, USA 89: 5640-5644 (1992). A chimeric receptor that encompasses the extracellular domain (ECD) of the mpl receptor and the transmembrane (TM) and intracellular domain (ICD) of the Rse (Mark et al, J. of Biol. Chem. 269 (14): 10720 -10728 [1994]) with a carboxyl-terminal polypeptide tag (ie Rse.gD) was prepared for use in the KIRA ELISA described herein. See Figures 30 and 31 for a diagrammatic description of the assay. (a) Preparation of the trapping agent

Monoclonal anti-gD (clone 5B6) was produced against a peptide of the glycoprotein D of herpes simplex-virus (Paborsky et al., Protein Engineering 3 (6): 547-553 [1990]). The purified stock preparation was adjusted to 3.0 mg / ml in phosphate buffered saline (PBS), pH 7.4, and 1.0 ml aliquots were stored at -20 ° C. (b) Preparation of Anti-Phosphotyrosine Antibody Monoclonal anti-phosphotyrosine antibody, clone 4G10, was obtained from UBI (Lake Placid, NY) and using long-arm biotin-N-hydroxysuccinamide (Biotin-X-NHS, Research Organics, Cleveland, OH) biotinylated. (c) Ligand

The mpl ligand was made by the recombinant techniques described herein. The purified mpl ligand was stored as a stock solution at 4 ° C. (d) Preparation of the Rse.gD nucleic acid

Synthetic double-stranded oligonucleotides were used to reconstitute the coding sequence for the C-terminal 10 amino acids (880-890) of human Rse and to add an additional 21 araino acids containing an anti-body 5B6 epitope and a stop codon. Table 10 represents the final sequence of the synthetic part of the fusion gene.

Table 10

Synthetic double-stranded part of human Rse fusion

Gene

The synthetic DNA was ligated with the cDNA coding for the amino acids 1-880 of human Rse at the PstI site, starting at nucleotide 2644 of the published cDNA sequence of human Rse (Mark et al, Journal of Biological Chemistry 269 (14 ): 10720-10728 [1994]), and the HindIII positions in the polylinker of the expression vector pSV17.ID.LL (see FIG. 32A-L; SEQ ID NO: 22) to the expression plasmid pSV.ID.Rse to generate .gD. The expression plasmid comprises a dicistronic primary transcript containing the sequence coding for the DHFR, which is surrounded by 5 'splice donor and 31 splice acceptor intron splice sites, followed by the sequence coding for the Rse.gD. The full-length messenger RNA (not spliced) contains DHFR as the first open reading frame and therefore generates a DHFR protein, by means of which stable transformants can be selected. (e) Preparation of the mpl-Rse.gD nucleic acid

The expression plasmid pSV.ID.Rse.gD, prepared as described above, was modified to produce the plasmid pSV.ID.M.tmRd6, which fuses the coding sequence of the ECO of human mpl (amino acids 1-491) to the

Transmembrane domain and the intracellular domain of Rse.gD (amino acids 429-911). Synthetic oligonucleotides were used to link the coding sequence of a portion of the extracellular domain of human mpl to a portion of the Rse coding sequence in a two-step PCR cloning reaction as described by Mark et.al., J.Biol. Chem. 267: 26166-26171 (1992). The primers used for the first PCR reaction were MI (5'-TCTCGCTACCGTTTACAG-3 ') (SEQ ID NO: 61) and M2 (5'-CAGGTACCCACCAGGCGGTCTCGGT-3') (SEQ ID NO: 62)

with an mpl cDNA template and RI (5'-GGGCCATGACACTGTCAA-3 ') (SEQ ID NO: 63) and R2 (5'-GACCGCCACCGAGACCGCCTGGTGGGTACCTGTGGTCCTT-3') (SEQ ID NO: 64) with a Rse-cDNA . The PvuII-Smal part of this fusion connection was used for the construction of the chimeric receptor voiler Lange. (f) Cell transformation DP12.CHO cells (EP 307247, published March 15, 1989) were transformed by electroporation with pSV.ID.M.tmRd6, which had been linearized at a single NotI site in the plasmid part. The DNA was ethanol-precipitated after a phenol / chloroform extract and resuspended in 20 μΐ 1/10 Tris EDTA. Then 10 pg DNA with 107 CHO DP12 cells in 1 ml PBS before electroporation at 400 volts and 330 μί for 10 minutes on ice. The lines were brought back to ice for 10 minutes before being seeded in non-selective medium. After 24 hours, the lines were incubated in nucleoside-free medium in order to select for stable DHFR + clones.

(g) Selection of the transformed lines for use in the KIRA ELISA

Clones expressing Mpl / Rse.gD were identified by Western blotting of the total cell lysates after fractionation by SDS-PAGE using the antibody 5B6, which recognizes the gD epitope tag. (h) media

The lines were cultivated in F12 / DMEM 50:50 (Gibco / BRL, Life Technologies, Grand Island, NY). The media were supplemented with 10% diafiltered FBS (HyClone, Logan, Utah), 25 mM HEPES and 2 mM L-glutamine.

(i) KIRA ELISA

DP12CHO cells transformed with mpl-Rse.gD were sown in the wells of a culture plate with flat bottom and 96 wells in 100 μl medium (3 × 10 4 per well) and cultivated overnight at 37 ° C. in 5% CO 2. The following morning, the supernatants in the wells were decanted and the plates were gently tapped on a paper towel. 50 μl medium, containing either the experiment samples or 200, 50, 12.5, 3.12r 0.78, 0.19, 0.048 or 0 ng / ml mpl ligand, were then added to each well. The lines were stimulated at 37 ° C for 30 minutes, the supernatants in the wells were decanted and the plates were gently tapped again on a paper towel. In order to lyse the lines and to bring the chimeric receptors into solution, 100 μl lysis buffer was added to each well. The lysis buffer consisted of 150 mM NaCl containing 50 mM HEPES (Gibco), 0.5% Triton-X 100 (Gibco), 0.01% thimerosal, 30 KIU / ml aprotinin (ICN Biochemicals, Aurora, OH), 1 mM 4- (2-Aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF; ICN Biochemicals), 50 μL leupeptin (ICN

Biochemicals) and 2 mM sodium orthovanadate (Na3V04; Sigma Chemical Co, St. Louis, MO), pH 7.5. The plate was then gently agitated on a plate shaker (Bellco Instruments, Vineland, NJ) for 60 minutes at room temperature. While the lines were dissolving, an ELISA micro-titer plate (Nunc Maxisorp, Inter Med, Denmark) was run overnight at 4 ° C with the anti-gD monoclonal antibody 5B6 (5.0 jLig / ml in 50 mM Carbonate buffer, pH 9.6, .100 μΙ / well) had been coated, decanted, tapped on a paper towel and blocked with 150 μΙ / well blocking buffer [PBS, containing 0.5% BSA (Intergen Company, Purchase, NY) and 0.01% thimerosal] saturated (blocked) for 60 minutes at room temperature with gentle movement.

After 60 minutes, the plate coated with anti-gD 5B6 was washed 6 times with wash buffer (PBS containing 0.05% Tween-20 and 0.01% thimerosal) using an automated plate washer (ScanWasher 300, Skatron Instruments,

Inc., Sterling, VA).

The lysate, which contained solubilized MPL / Rse.gD from the cell culture microtiter wells, was transferred to anti-gD 5B6 coated and saturated ELISA wells (85 μΙ / well) and for 2 hours at room temperature with gentle agitation xnkubxert. Unbound mpl Rse.gD was removed by washing with wash buffer, and 100 μl biotinylated 4G10 (anti-phosphotyrosine), 1: 18000 in dilution buffer (PBS, containing 0.5% BSA, 0.05-s Tween-20, 5 mM EDTA and 0.01% thimerosal), ie 56 ng / ml was added to each well. After 2 hours incubation at room temperature, the plate was washed, and 100 µΐ horseradish peroxidase (HRPO) conjugated streptavidin (Zymed Laboratories, S.San Francisco, CA), diluted 1: 60,000 in dilution buffer, were added to each well. The plate was incubated for 30 minutes at room temperature with gentle agitation. The free avidin conjugate was washed away and 100 μl of freshly prepared substrate solution (tetramethylbenzidine [TMB]; 2-component substrate kit, Kirkegaard and Perry, Gaithersburg, MD) were added to each well. The reaction was allowed to proceed for 10 minutes, after which color development was stopped by the addition of 100 μl / well of 1.0 M H3P04. Absorbance at 450 nm was measured at a reference wavelength of 650 nm (ABS450 / 650) using a vmax plate reader (Molecular Devices, Palo Alto, CA) by a Macintosh Centris 650 (Apple Computers, Cupertino, CA) and DeltaSoft Software (BioMetallics, Inc. Princeton, NJ).

The standard curve was generated by stimulating dpl2.trkA, B or C.gD cells with 200, 50, 12.5, 3.12, 0.78, 0.19, 0.048 or 0 ng / ml mpl ligand as ng / ml TPO versus

Average ABS450 / 650 - presented using the DeltaSoft program. The sample concentrations were obtained by interpolating their absorption on the standard curve and are expressed as ng / ml TPO activity.

It has been found that the mpl ligand can activate the mpl-Rse.gD chimeric receptor in a concentration-dependent and ligand-specific manner. Furthermore, it was found that the mpl-Rse.gD KIRA ELISA tolerates up to 100% human serum (shown) or 100% plasma (not shown), which allows the use of the assay to quickly screen patient and pK samples . 293-produced TPO332 hour curve

TPO conc. (Ng / ml)

Example 18

Receptor-mediated ELISA for thrombopoietin (TPO) ELISA plates were coated with rabbit F (ab ') 2-anti-human IgG (Fc) in carbonate buffer, pH 9.6., At 4 ° C. overnight. The plates were saturated with 0.5% bovine serum albumin in PBS for one hour at room temperature. The fermentation crop containing the chimeric receptor, mpl-IgG, was added to the plates and incubated for 2 hours. Two-fold serial dilutions (0.39-25 ng / ml) of the standard (TPO332, produced in 293 cells with the concentration determined by means of quantitative amino acid analysis) and serially diluted samples in 0.5% bovine serum albumin, 05% Twen-20 was added to the plates and incubated for 2 hours. Bound TPO was purified with Protein A, bio-tinylated rabbit antibodies to TP0155, which was prepared in E. coli (1 hour incubation), followed by streptavidin peroxidase (30 minutes incubation) and 3.3 ', 5.5' -Tetramethylbenzidin detected as a substrate. The absorbance was read at 450 nm. The plates were washed between steps. For data analysis, the

Standard curve fitted using a Kaleidagraph four-parameter curve fitting program. The sample concentrations were calculated based on the standard curve.

Example 19

Expression and purification of the TPO of 293 cells 1. Preparation of the expression vectors for 293 cells.

A cDNA corresponding to the entire open reading frame for TPO was obtained by PCR using the following oligonucleotides as primers:

Table 11 293 PCR primers

PRK5-hmpl I (described in Example 9) was used as a template for the reaction in the presence of the pfu DNA polymerase (Stratagene). The initial denaturation was for 7 minutes at 94 ° C, followed by 25 amplification cycles (1 minute at 94 °, 1 minute at 55 ° C and 1 minute at 72 ° C). The last extension was carried out for 15 minutes at 12 ° C. The PCR product was purified and cloned between the Clal and Xbal restriction sites of plasmid pRK5tkneo, a vector derived from pRK5 and modified to express a neomycin resistance gene under the control of the thymidine kinase promoter, to the vector pRK5tkneo. Get ORF. A second construct, corresponding to the epo-homologous domain, was generated in the same way, but using the Cla.FL.F as a forward primer and the following reverse primer:

Arg.STOP.Xba: 5 'TCT AGA TCT AGA TCA CCT GAG GCA GAG GGT GGA CC 3' (SEQ ID NO: 66)

The final construct is called pRK5-tkneoEP0-D.

The sequence of both constructs was verified as described in Example 7. 2. Transfection of human embryonic kidney cells

These two constructs were transfected into human embryonic kidney cells using the CaP04 method as described in Example 9. 24 hours after the transfection, the selection of neomycin-resistant clones was started in the presence of 0.4 mg / ml G418. 10 to 15 days later, single colonies were transferred to 96-well plates and grown to confluence. Expression of ML153 or ML332 in the conditioned media from these clones was determined using the Ba / F3-mpI proliferation assay (described in Example 1). 3. Purification of rhML332

Medium conditioned by 293-rb.ML.332 was applied to a Blue-Sepharose (Pharmacia) column, which was equilibrated in 10 mM sodium phosphate pH 7.4 (buffer A). The column was then washed with 10 column volumes of buffer A and buffer A containing 2M urea, respectively. The column was then eluted with Buffer A containing 2M urea and 1 M NaCl. The Blue Sepharose elution pool was then applied directly to a WGA-Sepharose column equilibrated in buffer A. The WGA-Sepharose column was then washed with 10 column volumes of buffer A containing 2M urea and 1M NaCl and eluted with the same buffer containing 0.5 M N-acetyl-D-glucosamine. The WGA-Sepharose eluate was applied to a C4-HPLC column (Synchrom, Inc.) equilibrated in 0.1% TFA. The C4-HPLC column was eluted with a discontinuous propanol gradient (0-25%, 25-35%, 35-70%). It was found that rhML332 elutes in the 28-30% propanol range of the gradient. In the SDS-PAGE the purified rhML332 migrates as a broad band in the 68-80 kDa range of the gel (see FIG. 15). 4. Cleaning the rhML153

Medium conditioned by 293-rhML153 was separated on Blue-Sepha-rose as described for rhML332. The Blue Sepharose eluate was applied directly to an mpl caffinity column as described above. RhML153 eluted from the mpI affinity column was purified for homogeneity using a C4 HPLC column run under the same conditions described for rhML332. In the SDS-PAGE, the purified rhML153 separates into two larger and two smaller bands with MG of approximately 18000-21000 (see FIG. 15).

Example 20

Expression and Purification of the TPO from CHO 1. Description of the CHO Expression Vectors

The expression vectors used in the electroporation protocols described below have been designated as follows: pSVI5. ID.LL.MLORF (full length or ΜΡ0332) and pSVI5.ID.LL.MLEPO-D (shortened or hTP0153).

The relevant features of these plasmids are shown in FIGS. 23 and 24. 2. Preparation of the CHO expression vectors

A cDNA corresponding to the entire open reading frame of hTPO was obtained by PCR using the oligonucleotide primers of Table 12.

PRK5-hmpl I (described in Examples 7 and 9) was used as a template for the reaction in the presence of the pfu DNA polymerase (Stratagene). The initial denaturation was carried out for 7 minutes at 94 ° C, followed by 25 amplification cycles (1 minute at 94 ° C, 1 minute at 55 ° C and 1 minute at 72 ° C). The last extension was carried out for 15 minutes at 72 ° C. The PCR product was purified and cloned between the restriction sites Clal and Sali of the plasmid pSVI5.ID.LL in order to obtain the vector pSVI5.ID.LL.MLORF. A second construct corresponding to the EPO homologous domain was generated in the same way but using Cla.FL.2 as the forward primer and the following reverse primer: EPOD.SaJ 5 'AGT CG A CGT CGA CTC ACC TGA CGC AG A GGG T GG ACC 3 '(SEQ ID NO: 68)

The final construct is called pSVI5.ID.LL.MLEPO-D. The sequence of both constructs was verified as described in Examples 7 and 9.

Essentially, the coding sequences for the full-length ligand and the truncated ligand were introduced into the multiple cloning site of the CHO expression vector pSVI5.ID.LL. This vector contains the SV40 early promoter / enhancer region, a modified splice unit containing the mouse DHFR cDNA, a multiple cloning site for the introduction of the gene of interest (in this case the TPO sequences described), a poly-adenylation signal of SV40 and an origin of replication and the beta-lactamase gene for plasmid selection and amplification in bacteria. 3. Methodology for Establishing Stable CHO Cell Lines Expressing Recombined Human TPO332 and TPO153 a. Description of the CHO parent cell line

The CHO (Chinese hamster ovary) host cell line used for expression of the TPO molecules described herein is known as CHO-DP12 (see EP 307 247, published March 15, 1989). This mammalian cell line was derived from a transfection of the parent line (CHO-K1 DUX-B11 (DHFR -) - obtained from Dr. Frank Lee from Standford University with the permission of Dr. L. Chasin) with a preproinsulin expressing vector clonally selected in order to obtain clones with reduced insulin requirements. These lines are also DHFR minus, and clones can be selected for the presence of DHFR cDNA vector sequences by growth on medium lacking nucleoside additives (glycine, hypoxanthine and thymidine). This S'e lesson system for stably expressing CHO cell lines is widely used. b. Transfection method (electroporation) TPO332 and TPO153 expressing cell lines were obtained by transfection of DP12 cells by means of electroporation (see for example Andreason, GLJTiss. Cult. Meth., 15.56 [1993]) with line-arized pSVI5.ID.LL.MLORF- or pSVI5. ID.LL.MLEPO-D-Plas-miden generated. Three (3) restriction enzyme reaction mixtures were made to cut the plasmids, respectively; 10 μg, 25 μg and 50 μg of the vector with the enzyme NotX by standard methods of molecular biology. This restriction site is found only once in the vector in the linearization region 3 'and outside the TPO ligand transcription units (see FIG. 23). The 100 μΙ reactions were prepared for an overnight incubation at 37 ° C. The next day the mixtures were extracted once with phenol-chloroform-isoamyl alcohol (50: 49: 1) and ethanol-precipitated on dry ice for about one hour. The precipitate was then collected by microcentrifugation for 15 minutes and dried. The linearized DNA was resuspended in 50 μΐ Ham's DMEM-F12 1: 1 medium, supplemented with standard antibiotics and 2 mM glutamine.

DP12 cells growing in suspension culture were seeded, washed once in the medium described for resuspending the DNA and finally resuspended in the same medium to a concentration of 107 lines per 750 μl. Aliquots of the lines (750 μΐ) and each mixture of the linearized DNA were incubated together for one hour at room temperature and then transferred to a BRL electroporation chamber. Each reaction mixture was then electroporated in a standard BRL electroporation apparatus at 350 volts, set at 330 ßF, and low capacitance. After electroporation, the lines were allowed to settle on the apparatus for 5 minutes and then for an additional 10 minute incubation period on ice. The electroporated lines were placed in 60 mm cell culture dishes containing 5 ml of complete standard growth medium for CHO cells (DMEM-F12 50:50 with high glucose concentration without glycine, supplemented with IX GHT, 2 mM glutamine and 5 % fetal calf serum) and grown overnight in a 5% C02 cell culture incubator. c. Selection and screening method The next day, the lines were detached from the plates with trypsin by standard methods and placed in 150 mm tissue culture dishes containing selective DHFR medium (Ham's DMEM-F12 described above, 1: 1 medium, supplemented with entwe - the 2% or 5% dialyzed fetal calf serum but free from glycine, hypoxanthine and thymidine; this is the DHFR standard selection medium we use). The cell from each 60 mm bowl was subsequently sown again in 5/150 mm bowls. The lines were then incubated for 10-15 days (with medium change) at 37 ° C / 5% CO 2 until clones started to appear and reached sizes suitable for transfer to 96-well dishes. Over a period of 4-5 days, the cell lines were used in sterile yellow tips on a pipetting aid set to 50 ml in 96-well dishes

Convictions transferred. The lines were allowed to grow to confluence (usually 3-5 days), after which the dishes were trypsinized and two copies of the original dish were reproduced. Two of these copies were temporarily stored in the freezer with rows in each well, diluted in 50 μΐ 10% FCS in DMSO. Conditioned, serum-free medium samples for 5 days were examined from confluently grown wells in the third dish for TPO expression using the activity assay based on Ba / F cells. The highest expressing clones based on this assay were reactivated from storage and pulled into two confluent 250 mm T-bottles for transfer to the cell culture group for suspension adjustment, control and preservation. d. Amplification action log

Several cell lines with the highest titers from the selection described above were subsequently passed through a standard methotrexate amplification treatment in order to generate clones with high titers. CHO cell clones are expanded and seeded in 10 cm dishes with 4 methotrexate concentrations (i.e. 50 nM, 100 nM, 200 nM and 400 nM) to two or three cell numbers (105, 5x105 and 106 lines per dish). These kuituren are then incubated at 37 ° C / 5% CO2 until clones are established and suitable for transfer in 96-well dishes for further assay. Several high-clone clones from this selection were exposed to even greater concentrations of methotrexate (ie, 600 nM, 800 nM, 1000 nM and 1200 nM) and the resistant clones were allowed to establish as described above, after which they were transferred to 96-well dishes and were examined. 4. Cultivate stable CHO cell lines that express recombinant human TP0332 and TP0153.

Stored lines are thawed and the cell population is expanded by standard cell growth techniques in either serum-free or serum-containing media. After expansion to a sufficient cell density, the rows are washed to remove used cell culture medium. The rows are then cultured by any standard method, including batches, fed batches, or continuous culture at 25-40 ° C, neutral pH, with a minimum of 5 dissolved O2 % until the constitutively secreted TPO has accumulated. The cell culture fluid is then separated from the cells by mechanical means such as centrifugation. 5. Purification of the recombinant human TPO from CHO culture liquids

Harvested cell culture fluid (HCCF) is applied directly to a

Blue-Sepharose 6 Fast Flow column;

Pharmacia) which applied in 0.01 M Na phosphate pH 7.4, 0.15 M NaCl at a rate of approximately 1001 HCCF per liter of resin and 2 a linear flow rate of approximately 300 ml / h / cm. The column is then washed with 3 to 5 column volumes of the calibration buffer, followed by 3 to 5 column volumes of 0.01M Na phosphate pH 7.4, 2, OM urea. The TPO is then eluted with 3 to 5 column volumes of 0.01M Na phosphate pH 7.4, 2, OM urea, 1.0M NaCl.

The Blue Sepharose pool containing TPO is then changed to a

Wheat germ lectin Sepharose 6MB column (Pharmacia) applied,

those in 0.01M Na phosphate pH 7.4, 2, OM urea and 1.0M

NaCl applied at a rate of 8 to 16 ml of the Blue Sepharose Pool 2 per ml of resin at a flow rate of approximately 50 ml / h / cm. The column is then washed with 2 to 3 column volumes of the equilibration buffer. The TPO is then eluted to 2 to 5 column volumes of 0.01M Na phosphate pH 7.4, 2, OM urea, 0.5M N-acetyl-D-glucosamine.

The wheat germ lectin pool is then adjusted to a final concentration of 0.04% C12E8 and 0.1% trifluoroacetic acid (TFA). The resulting pool is applied to a C4 reverse phase column (Vydac 214TP1022), which is loaded in 0.1% TFA, 0.04% C] _2E8 to a loading of approximately 0.2 to 0.5 mg. . 2 protein per ml of resin was equilibrated at a flow rate of 157 ml / h / cm.

The protein is eluted in a two-phase linear acetonitrile gradient containing 0.1% TFA, 0.04% C ^ Eg. The first phase consists of a linear gradient from 0 to 30% acetonitrile in 15 minutes. The second phase consists of a linear gradient of 30 to 60% acetonitrile in 60 minutes. The TPO elutes at approximately 50% acetonitrile. A pool is created based on SDS-PAGE.

The C4 pool is then diluted with 2 volumes of 0.01M Na phosphate pH 7.4, 0.15M NaCl and against approximately 6 volumes of 0.01M NA phosphate pH 7.4, 0.15M NaCl on an Amicon YM or a similar ultrafiltration membrane with a molecular weight cutoff size of 10,000 to 30,000 Daltons. The resulting diafiltrate can then be directly processed further or further concentrated by ultrafiltration. The diafiltrate / concentrate is adjusted to a final concentration of 0.01% Tween-80.

All or part of the diafiltrate / concentrate, which is equivalent to 2 to 5% of the calculated column volume, is then applied to a Sephacryl S-300 HR column (Pharmacia), which is in 0.01M Na phosphate pH 7.4 , 0.15M NaCl, 0.1% Tween-80, and chromatographed at a flow rate of approximately 17 ml / h / cm2. The TPO-containing fractions that are free of aggregates and proteolytic degradation products are collected on the basis of SDS-PAGE. The resulting pool is filtered through a 0.22 μ filter, Millex-GV or the like, and stored at 2-8 ° C.

Example 21

Transformation and induction of TPO protein synthesis in E.coli 1. Construction of E.coli TPO expression vectors

The plasmids pMP21, pMP151, pMP41, pMP57 and pMP202 are designed for the expression of the first 155 amino acids of the TPO downstream from a small leader which varies among the different constructs. The leaders enable a strong translation initiation and fast cleaning. The plasmids pMP210-l, -T8, -21, -22, -24, -25 are designed for the expression of the first 153 amino acids of the TPO downstream from an initiation methionine and differ only in the codon usage for the first amino acids of the TPO , while the plasmid pMP251 is a derivative of pMP210-l, in which the carboxy-terminal end of the TPO is extended by two amino acids.

All of the above plasmids will show high rates of intracellular TPO expression in E. coli after induction of the. Produce tryptophan promoters (Yansura, D.G. et al. Methods in Enzymology (Goeddel, D.V., Hog.) 185: 54-60 Academic Press, San Diego [1990]). The plasmids pMPl and pMP172 are intermediate products · in the construction of the above plasmids for intracellular expression of the TPO. (a) Plasmid pMPl

The plasmid pMPl is a secretion vector for the first 155 amino acids of the TPO and was constructed by ligating together 5 DNA fragments as shown in FIG. 33. The first of these was the vector pPho21, in which the small MluI-BamHI fragment had been removed. pPho21 is a derivative of phGHl (Chang, CN et. al., Gene 55: 189-196 [1987]), in which the gene for human growth hormone has been replaced by the E. coli phoA gene and a Mlul Restriction site has been incorporated into the coding sequence for the STII signal sequence at amino acids 20-21.

The next two fragments, a 258 base pair Hinfl-Pstl DNA piece of the pRK5-hmplI coding for the TPO amino acids 19-103 (Example 9), and the following synthetic DNA coding for the amino acids 1-18.

S'-CQCGTATGCCAGCCCGGCTCCTCCTGCTTGTGACCTCCGAGTCCTCAGTAAACTGCTTCG

TG

ATACGGTCGGGCCGAGGAGGACGAACACTGGAGGCTCAGGAGTCATTTGACGAAGC ACTGA-5 '(SEQ ID NO: 69) (SEQ ID NO: 70) were pre-ligated using T4 DNA ligase and then cut with Pstl. The fourth was a 152 base pair Pstl-Haelll fragment from pRK5hmpII encoding amino acids 104-155 of the TPO. The last was a 412 base pair StuI-BamHI fragment from pdhl08, which contains the lambda transcription terminator as previously described (Scholtissek, S. et.al., NAR 15: 3185 [1987]). (b) Plasmid pMP21

The plasmid pMP21 is designed for the expression of the first 155 amino acids of the TPO with the aid of a 13 amino acid leader which comprises part of the SII signal sequence. It was constructed by ligating together three (3) DNA fragments as shown in Figure 34, the first of which is the vector pVEG31 in which the small Xbal-SphI fragment was removed. The vector pVEG31 is a derivative of pHGH207-l (de Boer, HA et. Al., In Promoter Structure and Function (Rodriguez, RL and Chamberlain, MJ, ed.) 462, Praeger, New York [1982]), in which the gene for human growth hormone has been replaced by the gene for vascular endothelial growth factor (this identical vector fragment can be obtained from the latter plasmid).

The second part in the ligation was a synthetic DNA double strand with the following sequence:

S'-CTAGAATTATGAAAAAGAATATCGCATTTCTTCTTAA TTAATACTTTTTCTTATAGCGTAAAGAAGAATTGCGC-5 '(SEQ ID NO: 71) (SEQ ID NO: 72)

The last piece was a 1072 base pair (MluI-3phI fragment of the pMPl coding for 155 amino acids of the TPO. (C) Plasmid pMP151

The plasmid pMP151 is designed for the expression of the first 155 amino acids of the TPO downstream of a leader which comprises 7 amino acids of the STII signal sequence, 8 histidines and a factor Xa cleavage site. As shown in Figure 35, pMP151 was constructed by ligating three DNA fragments together, the first of which is the vector pVEG31 described above, from which the small Xbal-SphI fragment has been removed. The second was a synthetic DNA double strand with the following sequence:

5'-CTAGAATTATGAAAAAGAATATCGCATTTCATCACCATCACCATCACCATCACATCGAAG

GTCGTAGCC

TTAATACTTTTTCTTATAGCGTAAAGTAGTGGTAGTGGTAGTGGTAGTGTAGCTTC CAGCAT-5 '(SEQ ID NO: 73) (SEQ ID NO: 74)

The last was a 1064 base pair BglI-SphI fragment of pMPll coding for 154 amino acids of the TPO. The plasmid pMPll is identical to pMPl with the exception of a few codon changes in the STII signal sequence (this fragment can be obtained from the pMPl). (d) Plasmid pMP202

The plasmid pMP202 is very similar to the expression vector pMP151, except that the factor Xa site in the leader has been replaced by a thrombin site. As shown in Fig. 36, the pMP202 was constructed by ligating three DNA fragments together. The first of these was pVEG31, described above, in which the small Xbal-SphI fragment had been removed. The second was a synthetic DNA duplex with the following sequence:

5'-CTAG AATTAT G AAA AAG AAT AT CG C ATTT CATC ACC AT C ACC AT C ACC ATC ACATCG AA CCACGTAGCC ttaatactttttcttatagcgtaaagtagtggtagtggtagtggtagtgtagctt GGTGCAT-5 '(SEQ ID NO: 75) (SEQ ID NO: 76)

The last piece was a 1064 base pair BglI-SphI fragment from the plasmid pMPll described above. (e) Plasmid pMP172

The plasmid pMP172 is a secretion vector for the first 153 amino acids of the TPO and an intermediate for the construction of the pMP210. As shown in Figure 37, the pMP172 was prepared by ligating three DNA fragments together, the first of which was the vector pLS321amB in which the small EcoRI-HindIIII segment had been removed. The second was a 946 base pair EcoRI-Hgal fragment from the plasmid pMPll described above. The last piece was a synthetic DNA double strand with the following frequency: 5'-TCCACCCTCTGCGTCAGGT (SEQ ID NO: 77) GGAGACGCAGTCCATCGA-5 '(SEQ ID NO: 78) (f) plasmid pMP210

The plasmid pMP210 is designed for expression of the first 153 amino acids of the TPO after a translation initiation methionine. This plasmid was actually prepared as a plasmid bank in which the first 6 codons of the TPO were randomly changed in the third position of each codon and, as shown in FIG. 38, was obtained by ligation of three DNA Fragments constructed. The first of these was the vector pVEG31 described above, in which the small Xbal-SphI fragment had been removed. The second was a synthetic DNA double strand, shown below, which was first treated with DNA polymerase (Klenow), followed by digestion with Xbal and Hinfl, and that for initiation methionine and the randomly modified first 6 Codons of the TPO coded.

5'-GCAGCAGTTCTAGAATTATGTCNCCNGCNCCNCCNGCNTGTGACCTCCGA

ACACTGGAGGCT GTTCTCAGTAAA (SEQ ID NO: 79) CAAGAGTCATTTGACGAAGCACTGAGGGTACAGGAAG-5 (SEQ ID NO: 80)

The third was an 890 base pair HinfI-Sphl fragment from for. amino acid 19-153 of the pMP172 encoding TPO.

The pMP210 plasmid library from approximately 3700 clones was re-transformed onto LB plates with high tetracycline concentration (50 mg / ml) in order to select clones with a high translation initiation rate (Yansura, D.G. et. Al.,

Methods: A Companion to Methods in Enzymology 4: 151-158 [1992]). Of the 8 colonies that emerged on the high tetracycline concentration plates, 5 of the best clones in terms of TPO expression were subjected to DNA sequencing, and the results are shown in Fig. 39 (SEQ ID NO: 23, 24, 25, 26, 27 and 28). (g) Plasmid pMP41

The plasmid pMP41 is designed for expression of the first 155 amino acids of the TPO fused to a leader consisting of 7 amino acids of the STII signal sequence followed by a factor Xa site. The plasmid was constructed, as shown in Figure 40, by ligating three pieces of DNA together, the first of which was the vector pVEG31 described above in which the small XBal-SphI fragment was removed. The second was the following synthetic DNA double strand: 5'-CT AG A ATT AT G AA AA AG A AT AT CG C ATTTATCGAAGGTCGTAGCC (SEQ iD NO: 81) TTA ATACI TT TT CTT ATAG CGT AA AT AG CTT CC AG C AT -5 '(SEQ ID NO: 82)

The last piece of ligation was the 1064 base pair BglI-Sphl fragment from the plasmid described above. pMPll. (h) Plasmid pMP57

The plasmid pMP57 expresses the first 155 amino acids of the TPO downstream from a leader consisting of 9 amino acids of the STII sinal sequence and the dibasic site Lys-Arg. This dibasic site offers the possibility to remove the leader with the Potease ArgC. This plasmid was constructed as shown in Figure 41 by ligating three pieces of DNA together. The first of these was the vector pVEG31 described above in which the small Xbal-SphI fragment had been removed. The second was the following synthetic DNA duplex: 5'-CTAG AATTAT G AAAAAG AATATCGC ATTT CTTCTTAAACGTAGCC (SEQ ID NO: 83) TTAATACTTTTTCTTATAGCGTAAAGAAGAATTTGCAT-5 '(SEQ ID NO: 84)

The last part of the ligation was the 1064 base pair BglI-Sphl fragment from the plasmid pMPll described above. (i) Plasmid pMP251

Plasmid 251 is a derivative of pMP210-1, which includes two additional amino acids of the TPO at the carboxy-terminal end. As shown in Figure 42, the plasmid was constructed by combining two pieces of DNA, the first of which is the above-described pMP21, in which the small Xbal-Apal fragment had been removed . The second part of the ligation was a 316 base pair Xbal-Apal fragment from pMP210-l. 2. Transformation and induction of E. coli with TPO expression vectors

The above TPO expression plasmids were used to make E.coli strain 44C6 (w3110 tonAA rpoHts IonA clpPA galE). using the CaCl2 heat shock method (Mandei, M. et al., J. Mol.Biol., 53: 159-162, [1970]).

The transformed lines were first cultivated at 37 ° C. in LB medium containing 50 μg / ml carbenicillin until the optical density (600 nm) of the culture reached approximately 2-3. The LB culture was then diluted 20 times in M9 medium containing 0.49% casamino acids (w / v) and 50 mg / ml carbenicillin. After culturing at 30 ° C for one hour with aeration, indole-3-acrylic acid was added to a final concentration of 50 ßq / ml. The culture was then allowed to continue growing at 30 ° C with aeration for a further 15 hours, after which the rows were harvested by centrifugation.

Example 22

Production of biologically active TPO (Met-1 1-153) in E.coli

The process described below for the production of biologically active, refolded TPO (met-1 1-153) can be used in an analogous manner for the production of further TPO variants, including N and C-terminally extended forms (see Example 23) . A. Obtaining non-soluble TPO (Met-1 1-153) E.c.oli cells which are encoded by the plasmid pMP210-1 TPO. (Met-1 1-153) were fermented as described above. Typically, approximately 100 g lines in 11 (10 volumes) cell disruption buffer (10 mM Tris, 5 mM EDTA, pH8) are resuspended with a Polytron homogenizer and the lines centrifuged at 5000 x g for 30 minutes. The washed cell pellet is resuspended again in 1 1 cell disruption buffer using the Polytron homogenizer and the cell suspension is passed through an LH cell disrupter (LH Inceltech, Ine.) Or a microfluidizer (Microfluidics International) according to the instructions, sent by the manufacturer. The suspension is centrifuged at 5000 g for 30 minutes, resuspended and centrifuged a second time to produce a washed pellet of broken bodies. The washed pellet is used immediately or stored frozen at -70 ° C. B. Solubilization and Purification of Monomeric TPO (Met-1 1-153)

The above-mentioned pellet is resuspended in 5 weight volumes of 20 mM Tris, pH 8, with 6-8 M guanidine and 25 mM DTT (dithiothreitol) and stirred for 1 to 3 hours or overnight at 4 ° C to solubilize the TPO -Proteins. High concentrations of urea (6-8M) are also suitable, but usually give lower yields compared to guanidine. After solubilization, the solution is centrifuged at 30,000 x g for 30 minutes to produce a clear supernatant containing the denatured monomeric TPO protein. The supernatant is then chromatographed on a Superdex 200 gel filtration column (Pharmacia, 2.6 x 60 cm) at a flow rate of 2 ml / min and that

Protein ir.it 2 0 mM Na phosphate, pH 6.0, eluted with 10 mM DTT. The monomeric fractions containing denatured TPO protein eluting between 160 and 200 ml are pooled. The TPO protein is further purified on a semi-preparative C4 reverse phase column (2 x 20 cm VYDAC).

The sample is applied at 5 ml / min to a column which has been equilibrated in 0.1% HTFA (trifluoroacetic acid) with 30% acetonitrile. The protein is eluted with a linear acetonitrile gradient (30-60% in 60 min). The purified, reduced protein elutes at approximately 50% acetonitrile. This material is used for the refolding in order to obtain a biologically active TPO variant. C. Generation of Biologically Active TPO (Met-1 1-153)

Approximately 20 mg of the monomeric, reduced and denatured TPO protein in 40 ml 0.1% TFA / 50% acetonitrile is diluted in 360 ml of the refolding buffer, which optimally contains the following reagents:

50 mM Tris 0.3-M NaCl 5 mM EDTA 2% CHAPS detergent 25% glycerin 5 mM oxidized glutathione 1 mM reduced glutathione pH adjusted to 8.3

After mixing, the refolding buffer is gently stirred at 4 ° C for 12-48 h to effect maximum refolding yields of the TPO form with correct disulfide bonds (see below). The solution is then acidified with TFA to a final concentration of 0.2%, filtered through a 0.45 or 0.22 μ filter and 1/10 volume of acetonitrile added. The solution is then pumped directly onto a C4 reverse phase column and the purified, refolded TPO (Met-1-153) is eluted with the same gradient program as described above. Refolded, biologically active TPO is eluted at approximately 45% acetonitrile under these conditions. Versions of TPQ with incorrect disulfide bonds are eluted earlier. The final purified TPO ((Met-1 1-153) is more than 95% pure, as determined by SDS gels and analytical C4 reverse phase chromatography. For animal studies, the C4 purified material was tested in physiological ver - Portable buffers dialysed Isotonic buffers (10 mM Na acetate, pH 5.5, 10 mM Na succinate, pH 5.5 or 10 mM Na phosphate, pH 7.4), containing 150 mM NaCl and 0.01 % Tween-80 was used.

Due to the high effectiveness of the TPO in the Ba / F3 assay (half maximum stimulation is achieved at approximately 3 pg / ml), it is possible to obtain biologically active material using many different buffers, detergents and redox conditions. However, only a small amount of precisely folded material (<10%) is obtained under most conditions. For commercial manufacturing processes, it is desirable to achieve refolding rates of at least 10%, preferably 30-50%, and most preferably> 50%. Many different detergents (Triton X-100, Dode-cyl-beta-maltoside, CHAPS, CHAPSO, SDS, Salkosyl, Tween-20 and Tween-80, Zwittergent 3-14 and others) have been tested for their efficiency in ensuring high refolding yields -Afford. Among these detergents, it has been found that only the CHAPS family (CHAPS and CHAPSO) are generally suitable for the refolding reaction to limit protein aggregation and false disulfide formation. Amounts of CHAPS greater than 1% were most suitable. Sodium chloride was needed for the best yields, with optimal amounts between 0.1 M and 0.5 M. The presence of EDTA (1-5 mM) limited the amount of metal-catalyzed oxidation (and aggregation) that some Preparations were observed. Glyceine concentrations greater than 15% gave the optimal refolding conditions. For maximum yields, it was important to use both oxidized and reduced glutathione or oxidized and reduced cysteine as the redox reagent pair. Higher yields are usually observed when the molar ratio of the oxidized reagent is equal to or in excess over the reduced reagent member of the redox couple. pH values between 7.5 and approximately 9 were optimal for the refolding of these TPO variants. Organic solvents (e.g. ethanol, acetonitrile, methanol) were tolerated in concentrations of 10-15% or lower. High levels of organic solvents increased the amount of misfolded shapes. Tris and phosphate buffers were usually suitable. Incubation at 4 ° C also resulted in higher amounts of properly folded TPO. Refolding yields of 40-60% (based on the amount of reduced and denatured TPO used in the refolding reaction) are typical of TPO preparations that have been purified by the first C4 stage. Active material can be obtained if less pure preparations (e.g. immediately after the Superdex 200 column or after the initial extraction of refractive bodies) are used, although the yields are due to extensive precipitation and interference from the non-TPO -Proteins are lower during the TPO refolding process.

Since TPO (Met-1 1-153) contains 4 cysteine residues, it is possible to produce three different disulfide versions of this protein:

Version 1: Disulfide bonds between cysteine residues 1-4 and 2-3

Version 2: disulfide bonds between cysteine residues 1-2 and 3-4

Version 3: Disulfide bonds between cysteine residues 1-3 and 2-4 During the initial investigation to determine the refolding conditions, several different peaks containing TPO protein were separated by C4 reverse phase chromatography. Only one of these peaks had significant biological activity as determined using the Ba / F3 assay. The refolding conditions were then optimized in order to preferably obtain this version. Under these conditions, the incorrectly folded versions make up less than 10-20% of the total monomeric TPO obtained.

The disulfide pattern for the biologically active TPO has been determined by mass spectrometry and protein sequencing to be 1-4 and 2-3 (i.e. version 1). Aliquots of the various C4-resolved peaks (5-10 nmol) were digested with trypsin - (1:25 mol ratio of trypsin to protein). The digestion mixture was analyzed by matrix-assisted laser desorption mass spectrometry before and after the reduction with DTT. After reduction, masses corresponding to most of the larger tryptic peptides of the TPO were detected. Some of these masses were missing in the non-reduced samples and new masses were observed. The mass of the new peaks basically corresponded to the sum of the individual tryptic peptides which play a role in the disulfide binding. Thus, it was possible to unambiguously determine the disulfide pattern of the refolded, recombinant, biologically active TPO as 1-4 and 2-3. This is compatible with the known disulfide pattern of the related molecule erythropoietin. D. Biological Activity of Recombinant Refolded TPO (Met 1-153) Refolded and purified TPO (Met-1 1-153) had both in vitro and in vivo assay activity. Half-maximum stimulation of thymidine incorporation in the Ba / F3 cells was achieved in the Ba / F3 assay at 3.3 pg / ml (0.3 pM). Half-maximal activity at 1.9 ng / ml (120 pM) was found in the mpl receptor-mediated ELISA. In normal and myelo-suppressed animals produced by almost lethal X-rays, TPO (Met-1 1-153) was highly effective (activity was observed at doses as low as 30 ng / mouse), the production of to stimulate new platelets.

Example 23

Production of further biologically active TPO variants in E.coli

Three other TPO variants that have been manufactured, cleaned and refolded into biologically active forms in E.coli are provided below. (1) MLF - 13 residues from the bacterially derived signal sequence STII are fused to the N-terminal domain of the TPO (residues 1-155). The resulting sequence is MKKNIAFLLNAYASPAPPAC ..... CVRRA (SEQ ID NO: 85) _ _ 7 in which the leader sequence is small and represents C ...... C cys to Cys151. This variant was designed to provide a tyrosine for the radio-iodination of the TPO for receptor and biological studies. (2) H8MLF - 7 residues from the STII sequence, 8 histide residues and the enzyme cut sequence IEGR of factor Xa are fused to the N-terminal domain (residues 1-155) of the TPO. The sequence is MKKNIAFHHHHHHHHIEGRSPAPPAC CVRRA (SEQ ID NO: 86) 7 in which the leader sequence is underlined and represents C ..... c cys to Cys151. When purified and refolded, this variant can be treated with the enzyme factor Xa, which will cut behind the arginine residue of the IEGR sequence, resulting in a 155 residue long TPO variant with a natural serine as the N-terminal amino acid . (3) T-H8MLF - is prepared as described for variant (2) above, except that a thrombin-sensitive sequence IEPR is fused to the N-terminal domain of the TPO. The resulting sequence is MKKN1AFHHHHHHHH1EPRSPAPPAC ..... CVRRÀ (SEQ ID NO: 87) 7 where the leader sequence is underlined and C ..... C Cys 151. until Cys represents. After cleaning and refolding, this variant can be treated with the enzyme thrombin to produce a 155-residue variant of the TPO with a natural N-

Generate term. A. Extraction, solubilization and purification of the monomeric, biologically active TPO variants (1), (2) and (3).

All variants were expressed in E. coli. The majority of the variants were found in refractive bodies, as was observed in Example 22 for TPO (Met-1 1-153). Identical procedures for the recovery, solubilization and purification of the monomeric TPO variants, as described in Example 22, were used. Refolding conditions identical to those used for TPO (Met 1 1-153) were used in overall yields of 30-50%. After refolding, the TPO variants were purified by C4 reverse phase chromatography in 0.1% TFA using an acetonitrile gradient as described above. All TPO variants (in their non-proteolyzed forms) had biological activity as determined by the Ba / F3 assay with half maximum activities of 2-5 pM. B. Proteolytic processing of variants (2) and (3) to produce authentic, N-terminal TPO (1-155). The TPO variants (2) and (3) described above were designed with an enzymatically cutable leader peptide in front of the normal N-terminal amino acid residue of the TPO. After refolding and purification of variants (2) and (3) as described above, each was subjected to digestion with the appropriate enzyme. For each variant, the acetonitrile was removed from the C4 reverse phase step by blowing a gentle stream of nitrogen onto the solution. The two variants were then treated with either factor Xa or thrombin as described below. For the TPO variant (2), 1 M Tris buffer, pH 8, was added to the acetonitrile-free solution to a final concentration of 50 mM and the pH was adjusted to 8 if necessary. NaCl and CaCl2 were added to 0.1 M and 2 mM, respectively. Factor Xa (New England Biolabs) was added so that approximately a molar ratio of enzyme to variant from 1:25 to 1: 100 was achieved. The sample was incubated at room temperature for 1-2 h to determine maximum cleavage by determining a change in migration in SDS gels, indicating loss of the leader sequence. The reaction mixture was then purified by C4 reverse phase chromatography using the same gradient and conditions as described above for the purification of the properly folded variants. Non-cut variant B was separated from cut variant (2) by these conditions. The N-terminal amino acids were shown to be SPAPP, indicating that removal of the N-terminal leader sequence was successful. The factor Xa also generated variable amounts of internal cleavage within the TPO domain; the cleavage was observed behind the agenin residue at position number 118, whereby an additional N-terminal sequence of TTAHKDP (SEQ ID NO: 88) was generated. In non-reducing SDS gels, a single band was observed at approximately 17,000 daltons for the factor Xa cut variant; two bands with a molecular weight of approximately 12000 and 5000 daltons were seen in reducing gels, which is compatible with cleavage at arginine 118. This observation also confirmed that the two parts of the molecule were held together by a disulfide bond between the first and fourth cysteine residues are concluded from the tryptic digest experiments described above. In the biological Ba / F3 assay, the purified TPO (1-155) variant had a half-maximum activity of 0.2 to 0.3 picomolar after removal of the N-terminal leader sequence and with internal cleavage. The intact variant with the leader sequence had a half-maximum activity of 2-4 picomolar. For variant (3) the cleavage buffer was composed of 50 mM Tris, pH 8.2% CHAPS, 0.3 M NaCl, 5 mM EDTA and human or bovine thrombin (Calbiochem) at 1:25 to 1:50 from enzyme to TPO variant protein based on weight. The cleavage was carried out at room temperature for 2-6 hours. The progress of the cleavage was determined by SDS gels as described above for the factor Xa cleavage reaction. Usually more than 90% cleavage of the leader sequence was achieved. The resulting TPO was purified on C4 reverse phase columns as described above and was shown to have the desired N-terminus by amino acid sequencing. Only a small (<5%) fraction of internal cleavage on the same arginine-threonine bond as observed with factor Xa above was obtained.

The resulting TPO protein had high biological activity with half maximal responses in the Ba / F3 assay at 0.2-0.4 picomolar protein. In the mpl receptor-mediated ELISA, this protein produced a half-maximal response at 2-4 ng / ml of purified protein (120-240 picomolar), while the intact variant containing the leader sequence was 5-10 times less effective in both assays. For the animal experiments, the HPLC-purified, cut protein was in physiologically acceptable buffers, with 150 mM NaCl, 0.01% Tween-80 and 10 mM sodium succinate, pH 5.5 or 10 mM sodium acetate, pH 5 , 5 or 10 mM sodium phosphate, pH 7.4 dialyzed. The purified protein became stable for several weeks by HPLC and SDS gels when stored at 4 ° C.

In normal and myelo-suppressed mice, this purified TPO with the authentic N-terminal sequence was highly effective, stimulating the production of platelets at doses as low as 30 ng / mouse.

Example 24

Synthetic mpl ligand

Although the human mpl ligand (hML) is usually made using recombinant methods, it can also be synthesized by enzymatically ligating synthetic peptide fragments using the methods described below. Synthetic production of hML allows the incorporation of unnatural amino acids or synthetic functional groups such as polyethylene glycol. A mutant of the. Serine protease BPN, subtiligase (S221C / P225A) designed and manufactured to efficiently ligate peptide esters in aqueous solution (Abrahmsen et al., Bioch., 30: 4151-4159 [1991]). It has now been shown that synthetic peptides can be ligated enzymatically in a sequential manner to produce enzymatically active, long peptides and proteins such as Ribonucléase A (Jackson et al., Science [1994]). This technology, which is described in detail below, has enabled us to chemically synthesize long proteins that previously could only be produced using recombinant DNA technology.

The general strategy for hMLi53 synthesis using subtiligase is shown (Scheme 1). Starting with a peptide whose protective groups have been completely removed and which corresponds to the C-terminal fragment of the protein, an N-terminally protected, C-terminally activated peptide ester is added together with subtiligase. After the reaction is complete, the product is isolated by reverse phase HPLC and the protective. group removed from the N-terminus. The next peptide fragment is ligated, the protecting groups are removed and the process is repeated using the successive peptides until the full length protein is obtained. The procedure is similar to the solid phase methodology in which an N-terminally protected, C-terminally activated peptide is ligated to the N-terminus of a previous peptide and the protein is synthesized in a C-> N direction. However, since each linkage also results in up to 50 residues and the products are isolated after each ligation, much longer, highly pure proteins can be synthesized in reasonable yields.

Scheme 1. Strategy for the synthesis of hML using subtiligase

To our knowledge of both the sequence specificity of the subtiligase and the amino acid sequence of the biologically active "epo-domain" of the hML, we divided hMLi53 into seven fragments of 18-25 residues. Test ligation tetrapeptides were synthesized to determine appropriate ligation linkages for the 18-25 mers. Table 13 shows the results of these test ligations.

Table 13 hML test ligations. Donor and nucleophilic peptides were dissolved at 10 mM in 100 mM tricine (pH 7.8) at 22 ° C. The ligase was added to a final concentration of 10μΜ starting from a 1.6 mg / ml stock solution (approximately 70μΜ) and the ligation was allowed to proceed overnight. The yields are expressed as% ligation vs. Hydrolysis of the donor peptides.

Based on these experiments, the ligation peptides shown in Table 14 should be ligated efficiently by the subtilidase. A suitable protecting group for the N-terminus of each donor peptide ester was needed to prevent self-ligation. We chose an isonicotinyl (iNOC) protecting group (Veber et al., J. Org. Chem., 42: 3286-3289 [1977]) because it is water-soluble, incorporated in the last stage of solid-phase peptide synthesis and against anhydrous HF stable, which is used to remove the protecting groups and cleave the peptides from the solid phase resin. In addition, it can be removed from the peptide after each ligation under mild reducing conditions (Zn / CH3CO2H) to provide a free N-terminus for subsequent ligations. A glycolate-lysyl-amide (glc-K-NH2) ester was used for the C-terminal activation based on previous experiments, which showed that it is efficiently acylated by subtiligase. (Abrahmsen et al., Biochem., 30: 4151-4159 [1991]). The iNOC-protected glc-K amide-activated peptides can be synthesized using standard solid-phase methods as outlined (Scheme 2). The peptides are then sequentially ligated until the complete protein is made and the final product is refolded in vitro. Based on homology with EPO, it is believed that the disulfide pairs are formed between cysteine residues 7 and 151 and between 28 and 85. The oxidation of the disulfides can be achieved by simply stirring the reduced material under an oxygen atmosphere for several hours. The refolded material can then be purified by HPLC, and the fractions containing the active protein can be combined and lyophilized. As an alternative, the disulfides can be protected differentially to control sequential oxidation between specific disulfide pairs. Protecting cysteine 7 and 151 with acetamidomethyl (acm) groups would ensure the oxidation of 28 and 85. The acm groups could then be removed and residues 7 and 151 oxidized. Conversely, residues 28 and 85 could be aciu-protected and oxidized if sequential oxidation was required for the correct moth. Optionally, cysteine 28 and 85 may be substituted with a natural or unnatural residue other than Cys to ensure correct oxidation of cysteine 7 and 151.

Table 14

Peptide fragments used for the overall synthesis of h-MI »using the subtiligase

fragment

Sequence 1 (SEQ ID NO: 110) iNOC-HN-SPAPPACDLRVLSKLLRDSHVL-glc-K-NH2 (1-22) 2 (SEQ ID NO: 111) iNOC-HN-HSRLSQCPEVHPLPTPVLLPAVDF-glc-K-NH2 (23-46) 3 (SEQ ID NO: 112) iNOC-HN-SLGEWKTQMEETKAQDILGAVTL-g! CK-NH2 (47-69) 4 (SEQ ID NO: 113) iNOC-HN-LLEGVMAARGQLGPTCLSSLL-glc-K-NH2 (70-90) 5 (SEQ ID NO: 114) iNOC-HN-GQLSGQVRLLLGALQS-glc-K-NH2 (90-106) 6 (SEQ ID NO: 115) iNOC-HN-LLGTQLPPQGRTTAHKDPNAIF-glc-K-NH2 (107-128) 7 (SEQ ID NO : 116) H2N-LSFQHLLRGKVRFLMtVGGSTLCVR-CQ2 (129-153)

The peptide ligations are carried out at 25 ° C. in 100 mM tricine, pH 8 (freshly prepared and degassed by vacuum filtration through a 5 μm filter). Typically, the C-terminal fragment is dissolved in buffer (2-5 mM peptide) and a 10% stock solution of the subtiligase (1 mg / ml in 100 mM tricine, pH 8) is added to bring the final enzyme concentration to approximately 5 μm. A 3-5 molar excess of the glc-K-NH2 -activated fourth donor peptide is then added as a solid, dissolved, and the mixture is left to stand at 25.degree. The ligations were monitored by analytical reverse phase C18 HPLC (CH3CN / H20 gradient with 0.1% TFA). The ligation products are purified by preparative HPLC and lyophilized. The cleavage of the isonicotinyl (iNOC) groups was carried out by stirring HCl-activated zinc dust with the protected peptide in acetic acid. The zinc dust is removed by filtration and the acetic acid is removed under vacuum. The resulting peptide can be used directly in the next ligation and the procedure is repeated. Synthetic hML153 can be ligated to the synthetic or recombinant hML154_332) by methods analogous to those described above to produce full-length synthetic or semi-synthetic hML.

Synthetic hML has many advantages over recombinant. Unnatural side chains can be introduced to improve effectiveness or specificity. Functional polymer groups, such as polyethylene glycol, can be incorporated to improve the duration of action. For example, polyethylene glycol can be attached to lysine residues of the individual fragments (Table 14) before or after one or more ligation steps have been carried out. Protease sensitive peptide bonds can be removed or altered to improve stability in vivo. In addition, heavy atom derivatives can be synthesized to aid in structure determination.

Scheme 2. Solid phase synthesis of peptide fragments for segment ligation

a) Lysyl-paramethylbenzhydrylamine (MBHA) resin 1 (0.63 meq./g. Advanced ChemTech) is for one hour at 25 ° C with bromoacetic acid (5 eq.) and diisopropylcarbodiimide (5 eq.) in dimethylacetamide (DMA) rowed to produce the bromoacetyl derivative 2. b) The resin is washed thoroughly with DMA, and the individual Boc-protected amino acids (3 eq.,

Bachem) are esterified by stirring with sodium bicarbonate (6 eq.) In dimethylformamide (DMF) for 24 h at 50 ° C. to produce the corresponding glycolate-phenylalanyl amide resin 3. The amino-acetylated resin 3 is washed with DMF (3x) and dichloromethane (CH2CI2) (3x) and can be stored at room temperature for several months. The resin 3 can then be introduced into an automated peptide synthesizer (Applied Biosystems 430A) and the peptides elongated using standard solid-phase methods (5), c) The Ν-α-Boc group is mixed with a solution of 45 % Tri-fluoroacetic acid removed in CH2CI2. d) The following Boc-protected amino acids (5 eq.) are obtained using benzotriazol-l-yl-oxy-tris (dimethylamino) phosphonium hexafluorophosphate (BOP, 4 eq.) and N-methylmorpholine (NMM, 10 eq .) pre-activated in DMA and linked for 1-2 h. e) The last Ν-α-Boc group is removed (TFA / CH2CI2) to produce 4 and the isonicotinyl (iNOC) protecting group is stirred as described above (4) with 4-isonicotinyl2-4-dinitrophenyl carbonate (3 eq .) and NMM (6 eq.) in DMA at 25 ° C for 24 h. f) The cleavage and removal of the protective groups of the peptide by treatment with anhydrous HF (5% anisole / 5% ethylmethylsulfide) at 0 ° C. for one hour results in the iNOC-protected, glycolate-lys-amide-activated peptide 5, which is reversed -18 HPLC (CH3CN / H20-Gra-serves, 0.1% TFA) is cleaned. The identity of all subtrates is confirmed by mass spectrometry.

The invention as claimed is practicable in accordance with the foregoing description and the readily available references and starting materials.

Nevertheless, the applicants have deposited the cell line listed below with the American Type Culture Collection, Rockville, Md., USA (ATCC):

Escherichia coli, DHlOB-pBSK-hmpl I 1.8, ATCC Accession No. CRL 69575, deposited on February 24, 1994.

Plasmid, pSVI5.ID.LL.MLORF, ATCC accession no. CRL 75958, filed December 2, 1994 and CHO DP-12 cells, ML 1/50 MCB (marked # 1594), ATCC accession no. CRL 11770, filed December 6, 1994.

This deposit was made in accordance with the Budapest Treaty on International Recognition, the deposit of micro-organisms for the purposes of patent proceedings and the relevant regulations (Budapest Treaty). This ensures the accessibility of a living culture for 30 years after the deposit date. The organism is made accessible by the ATCC in accordance with the provisions of the Budapest Treaty.

Claims (39)

1. Isolated, substantially homogeneous mpI ligand poly-peptide. 2. The mpl ligand polypeptide according to claim 1 selected from the group consisting of (a) a polypeptide fragment; (b) a polypeptide variant; and (c) a chimeric polypeptide. 3. The mpl ligand polypeptide according to claim 1 selected from the group consisting of (a) the polypeptide isolated from a mammal; (b) the polypeptide made by recombinant means; and (c) the polypeptide made by synthetic means. 4. The mpl ligand polypeptide of claim 1 selected from the group consisting of (a) the polypeptide that is human; and (b) the polypeptide that is not immunogenic in a human. 5. An isolated, essentially homogeneous mpl agonist, characterized in that: (a) the agonist incorporates labeled nucleotides (3H-thymidine) into the DNA of IL-3-dependent Ba / F3 cells which are infected with human mpl P are transfected, stimulated; or (b) the agonist stimulates 35S incorporation into circulating platelets in a platelet rebound assay. 6. A polypeptide fragment according to claim 2, represented by

wherein hTPO (7-151) represents human TPO (hML) amino acid 7 ,. ^ 151. sequence from including Cys to Cys, X represents an amino group from Cys7 or an amino-terminal amino acid residue (s) selected from the group

Y represents the carboxy-terminal group of Cys or carboxy-terminal or carboxy-terminal amino acid residue (s) selected from the group

and amino terminal amino acid residue (s) extensions comprising one or more residues 176-332 of human ML, as shown in Figure 1 (SEQ ID NO: 1). 7. A polypeptide fragment according to claim 6, selected from the group TPO (I-153) and TPO (1-245). A polypeptide fragment according to claim 2, wherein the amino acid sequence comprises the polypeptide fragment

and combinations thereof. 9. A polypeptide according to claim 6 which is not glycosylated. 10. Isolated polypeptide encoded by a nucleic acid. with a sequence that hybridizes under moderately stringent conditions to the nucleic acid molecule with a nucleic acid sequence shown in FIG. 1 (SEQ ID NO: 2). 11. A polypeptide according to claim 11 which is biologically active. 12. Polypeptide according to claim 1, selected from the group hML, hML153, hML (R153A, R154A), hML2, hML3, hML4, mMl, ïïlML2, ïïiML3, pML and pML2. 13. The polypeptide according to claim 2, wherein the amino acid sequence of the polypeptide comprises amino acid residues 1 to X from FIG. 1 (SEQ ID NO: 1), wherein X is selected from the group 153, 155, 164, 174, 191, 205, 207, 217, 229, 245 and 332. 14. An isolated, substantially homogeneous mpl ligand poly-peptide that shares at least 80% sequence identity with the polypeptide of claim 13. 15. The polypeptide of claim 13, wherein X is 153. 16. The chimera comprising the mpl ligand from claim 13 fused to a heterologous polypeptide. 17. The chimera of claim 16, wherein the heterologous polypeptide is an immunoglobulin polypeptide. 18. The chimera of claim 16, wherein the heterologous polypeptide is an interleukin polypeptide. 19th A chimera comprising the N-terminal residues 1 to about 153 to 157 of hML substituted with one or more, but not all, of the human EPO residues appended or substituted in the N-terminal residues of hML at positions that are in alignment (alignment) as shown in Fig. 10 correspond. 20. An antibody which can bind the mpl ligand polypeptide according to claim 13. 21. A hybridoma cell line producing the antibody of claim 20. 22. An isolated nucleic acid molecule encoding the mpl ligand polypeptide of claim 1. 23. An isolated nucleic acid molecule encoding the mpl ligand polypeptide according to claim 13. 24. An isolated nucleic acid molecule comprising the nucleic acid sequence with the open reading frame as shown in Fig. 1 (SEQ ID NO: 2). 25. The isolated nucleic acid molecule according to claim 24, which encodes an mpl ligand polypeptide selected from the group hML, hMLi_53, hML (R153A, R154A), hML2, hML3, hML4, mMl, mML2, mML3, pML and pML2. 26. An isolated nucleic acid molecule selected from the group consisting of (a) a cDNA clone comprising the nucleotide sequence of the coding region of the mpl ligand gene; (b) a DNA sequence that is more stringent. Can hybridize conditions to a clone from (a); and (c) a genetic variant of one of the DNA sequences from (a) and (b) which codes for a polypeptide which has a biological property of its naturally occurring mpI ligand polypeptide. 27. An isolated DNA molecule having a sequence that can hybridize to a DNA sequence shown in Figure 1 (SEQ ID NO: 2) under moderately stringent conditions, wherein the DNA molecule encodes a biologically active mpl ligand polypeptide . 28. The nucleic acid molecule of claim 25, further comprising a promoter operably linked to the nucleic acid molecule. 29. Expression vector comprising the nucleic acid sequence from claim 25, operably linked to control sequences which are recognized by a host line which is transformed with the vector. 30. host line transformed with the vector of claim 29. 31. A method using a nucleic acid molecule coolant encoding the mpl ligand polypeptide to effect the production of the mpl ligand polypeptide, comprising culturing the host line according to claim 30. 32. The method of claim 31, wherein the mpl ligand polypeptide is recovered from the host line. • 33. The method of claim 31, wherein the mpl ligand polypeptide is recovered from the host cell culture medium. 34. A method for determining the presence of mpl ligand polypeptide, comprising hybridizing DNA encoding the mpl ligand polypeptide with a nucleic acid analysis sample and determining the presence of DNA for the mpl ligand polypeptide. 35. A method of amplifying a nucleic acid analysis sample comprising priming a nucleic acid polymerase reaction with nucleic acid encoding an mpl ligand polypeptide. 36. A composition comprising the mpl ligand polypeptide according to claim 1 and a pharmaceutically acceptable carrier. 37. Use of an mpl ligand polypeptide according to claim 1 for treating a mammal with or at risk for thrombocytopenia. 38. The composition of claim 36, further comprising a therapeutically effective amount of an agent selected from the group consisting of cytokine, colony stimulating factor and interleukin. 39. The composition of claim 38, wherein the agent is selected from KL, LIF, G-CSF, GM-CSF, M-CSF, EPO, IL-1, IL-2, IL-3, IL-5, IL -6, IL-7, IL-8, IL-9 and IL-11. *