JP4828121B2 - Targeted therapeutic protein - Google Patents

Description

(Reference to related applications)
No. 60 / 384,452 (filed on May 29, 2002); 60 / 386,019 (filed on June 5, 2002); 60 / 408,816 ( Claims the benefit of No. 10 / 272,531 (filed Oct. 16, 2002); and No. 60 / 445,734 (filed Feb. 6, 2003). Each of which is incorporated herein by reference in its entirety. No. 60 / 287,531 (filed Apr. 30, 2001); 60 / 304,609 (filed Jul. 10, 2001); No. 60 / 329,461 (Filed on October 15, 2001), 60 / 351,276 (filed on January 23, 2002); 10 / 136,841 (filed on April 30, 2002); No./272,483 (filed Oct. 16, 2002), the contents of each of which are hereby incorporated by reference in their entirety.

  The present invention provides a means of specifically delivering a protein to a targeted subcellular compartment of a mammalian cell. The ability to target proteins to subcellular compartments is very useful in the treatment of metabolic diseases, such as lysosomal storage diseases, particularly over 40 genetic disease classes that lack or lack lysosomal enzymes. .

(background)
Enzyme deficiency in cellular compartments (eg, Golgi, endoplasmic reticulum and lysosomes) results in a wide range of human diseases. For example, lysyl hydroxylase (which is usually an enzyme in the lumen of the endoplasmic reticulum) is required for proper processing of collagen; lack of this enzyme results in Erlers-Dunlo syndrome type VI (severe Connective tissue disorder). GnT II (usually found in the Golgi) is required for normal glycosylation of proteins and lack of GnT II induces defects in brain development. Over 40 lysosomal storage diseases (LSD) are caused either directly or indirectly by the lack of one or more proteins in the lysosome.

  Mammalian lysosomal enzymes are synthesized in the cytoplasm, where the enzyme is glycosylated with an N-linked high mannose carbohydrate, and penetrate the endoplasmic reticulum (ER). In the Golgi, high mannose carbohydrates are modified with this lysosomal protein by the addition of mannose-6-phosphate (M6P), which targets the lysosomal protein to the lysosome. M6P modified proteins are delivered to lysosomes through interaction with either of the two M6P receptors. The most convenient form of modification is when two M6Ps are added to the high mannose carbohydrate.

  Enzyme replacement therapy for lysosomal storage diseases (LSD) is being actively explored. Treatments other than those in Gaucher disease generally require that the LSD protein be taken up and delivered to lysosomes of various cell types in an M6P-dependent manner. One possible approach involves purifying the LSD protein and modifying it to incorporate M6P into the carbohydrate moiety. This modifier can be taken up by cells more efficiently than unmodified LSD protein due to interaction with the M6P receptor on the cell surface. However, new and simpler to target therapeutic agents to cellular compartments due to the time and expense required to prepare, purify and modify proteins for use in subcellular targeting There remains a need for more efficient and more cost effective methods.

(Summary of the Invention)
The present invention facilitates the treatment of metabolic diseases by providing targeted protein therapeutics that localize to subcellular compartments of cells where the therapeutic is needed. The present invention simplifies the preparation of targeted protein therapeutics by reducing the need for post-translational or post-synthesis processing of this protein. For example, the targeted therapeutic agent of the invention can be synthesized as a fusion protein comprising a therapeutic domain and a domain that targets the fusion protein to the correct subcellular compartment ("fusion protein" as used herein). Refers to a single polypeptide having at least two domains that are not normally present in the same polypeptide, so a naturally occurring protein is not a “fusion protein” as used herein) . Synthesis as a fusion protein allows, for example, targeting a therapeutic domain to a desired subcellular compartment without the complexity associated with separate therapeutic and chemical cross-linking of the target domain.

  The present invention also allows targeting therapeutics to lysosomes in an M6P-independent manner. Thus, targeted therapeutics do not need to be synthesized in mammalian cells, but can be synthesized chemically or in bacteria, yeast, protists or other organisms, regardless of glycosylation pattern. Which can be synthesized, facilitating the production of targeted therapeutic agents in high yield and relatively low cost. This targeted therapeutic can be synthesized as a fusion protein and can be further produced by simplifying production or by combining the targeting moiety with an independently synthesized therapeutic agent.

  The present invention allows lysosomes to be targeted for therapeutics without the need for addition of M6P to high mannose carbohydrates. This is based in part on the finding that one of these two M6P receptors also binds other ligands with high affinity. For example, the cation-independent mannose-6-phosphate receptor is also known as insulin-like growth factor 2 (IGF-II) receptor because it binds to IGF-II with high affinity. This low molecular weight polypeptide interacts with three receptors (insulin receptor, IGF-I receptor and M6P / IGF-II receptor). The biological effect is thought to be exerted mainly through the interaction with the former two receptors, while the interaction with the cation-independent M6P receptor is caused by lysosomes (where IGF-II is It is thought to produce predominantly IGF-II which is transported to (degraded).

  Accordingly, the present invention relates to one aspect of a targeted therapeutic comprising a targeting moiety that is therapeutically active in a mammalian lysosome and a therapeutic agent. As used herein, “therapeutically active” includes at least a polypeptide or other molecule that provides enzymatic activity to a cell that lacks enzymatic activity, or a compartment thereof. “Therapeutically active” also includes other polypeptides or other molecules intended to ameliorate or compensate for biochemical deficiencies in the cell, but are primarily cytotoxic or cytostatic. Does not contain certain molecules (eg chemotherapeutics).

In one embodiment, the targeting moiety is a means for binding to the extracellular domain of the human cation-independent M6P receptor in an M6P-independent manner when the receptor is present in the plasma membrane of the target cell ( For example, molecule). In another embodiment, the targeting moiety is a non-glucosylated lysosomal targeting domain that binds to the extracellular domain of the human cation-independent M6P receptor. In any embodiment, the targeting moiety includes, for example, IGF-II; retinoic acid or a derivative thereof; a protein having an amino acid sequence at least 70% identical to the domain of a urokinase-type plasminogen activator receptor; Antibody variable domains; or variants thereof. In some embodiments, the targeting moiety has a dissociation constant of less than micromolar at pH 7.4 or about pH 7.4 (eg, less than 10 ⁻⁸ M, less than 10 ⁻⁹ M, less than 10 ⁻¹⁰ M, or Between 10 ⁻⁷ M and 10 ⁻¹¹ M), and at a pH 5.5 or about pH 5.5, a dissociation constant of at least 10 ⁻⁶ M and a dissociation constant of at least 10 times at pH 7.4 or about pH 7.4 Binds to the receptor. In certain embodiments, the means for binding is at least 10 times less (ie, at least 10 times higher) at pH 5.5 or about pH 5.5 than at pH 7.4 or about pH 7.4. Binds to the extracellular domain (with a dissociation constant); in one embodiment, the dissociation constant at pH 5.5 or about pH 5.5 is at least 10 ⁻⁶ M. In further embodiments, the binding of the targeted therapeutic to the binding means is destabilized by a pH change from pH 7.4 or about pH 7.4 to pH 5.5 or about pH 5.5.

In another embodiment, the targeting moiety binds to the extracellular domain of the human cation-independent M6P receptor but does not bind to the receptor mutein in which amino acid 1572 is changed from isoleucine to threonine, or at least 10-fold. A lysosomal targeting domain that binds to this mutein with weak affinity (ie, with a dissociation constant at least 10 times higher). In another embodiment, the targeting moiety is a lysosomal targeting domain that can bind to a receptor domain consisting essentially of repeats 10-15 of the human cation-independent M6P receptor; the lysosomal targeting domain is a protein Can bind to proteins containing repeats 10-15, even if it does not contain other moieties that bind to this lysosomal targeting domain. Preferably, the lysosomal targeting domain is capable of binding to a receptor domain consisting essentially of repeats 10-13 of the human cation-independent mannose-6-phosphate receptor. More preferably, the lysosomal targeting domain can bind to a receptor domain consisting essentially of repeats 11-12, repeat 11 or amino acids 1508-1566 of the human cation-independent M6P receptor. In each of these embodiments, the lysosomal targeting domain preferably binds to the receptor or receptor domain with a dissociation constant of less than micromolar at pH 7.4 or about pH 7.4. In one preferred embodiment, the lysosomal targeting domain binds with a dissociation constant of about 10 ⁻⁷ M. In another preferred embodiment, the dissociation constant is less than 10 ⁻⁷ M.

  In another embodiment, the targeting moiety is a sufficiently overlapping binding moiety of human IGF-II so that the binding moiety binds to a human cation-independent M6P receptor. This binding moiety may sufficiently overlap by including an amino acid sequence that is sufficiently homologous to at least a portion of IGF-II, or by including a molecular structure that corresponds sufficiently to at least a portion of IGF-II, As a result, this binding moiety binds to the cation-independent M6P receptor. This binding moiety is selected from the group consisting of at least a portion of IGF-II (eg, amino acids 48-55 of human IGF-II, or amino acids 8, 48, 49, 50, 54 and 55 of human IGF-II). It can be an organic molecule having a three-dimensional shape corresponding to at least three amino acids. Preferred organic molecules have a hydrophobic moiety at the portion corresponding to amino acid 48 of human IGF-II, and at a portion corresponding to amino acid 49 of human IGF-II at pH 7.4 or about pH 7.4. Have a positive charge. In one embodiment, the binding moiety is a polypeptide comprising a polypeptide having an antiparallel alpha helix separated by 5 or fewer amino acids. In another embodiment, the binding moiety is a polypeptide having the amino acid sequence of an IGF-I or IGF-I mutein in which amino acids 55-56 are altered and / or amino acids 1-4 are deleted or changed. including. In further embodiments, the binding moiety comprises a polypeptide having an amino acid sequence that is at least 60% identical to human IGF-II; amino acids at positions corresponding to positions 54 and 55 of human IGF-II are preferably , PH 7.4 or at about pH 7.4, uncharged or negatively charged.

  In one embodiment, the targeting moiety is a polypeptide comprising the amino acid sequence phenylalanine-arginine-serine. In another embodiment, the targeting moiety is a polypeptide comprising an amino acid sequence that is at least 75% homologous to amino acids 48-55 of human IGF-II. In another embodiment, the targeting moiety comprises amino acids 8-28 and 41-61 of human IGF-II on a single polypeptide or a separate polypeptide. In another embodiment, this targeting moiety is amino acids 41-61 of human IGF-II and amino acids 8-28 of human IGF-II that differs from the human sequence at amino acids 9, 19, 26 and / or 27. Including muteins.

  In some embodiments, the binding between the therapeutic agent and the targeting moiety is unstable at pH 5.5 or about pH 5.5. In preferred embodiments, the binding of the targeting moiety to the therapeutic agent is mediated by a protein acceptor having a pKa between 5.5 and 7.4 (eg, a derivative thereof such as imidazole or histidine). Preferably, one of the therapeutic agent or targeting moiety is linked to a metal and the other is linked to a pH dependent metal binding moiety.

  In another aspect, the present invention relates to a therapeutic fusion protein comprising a therapeutic domain and a subcellular targeting domain. This subcellular targeting domain binds to the extracellular domain of the receptor on the outer surface of the cell. Upon internalization of this receptor, the subcellular targeting domain allows localization of the therapeutic domain to the subcellular compartment (eg, lysosome, endosome, endoplasmic reticulum (ER) or Golgi complex) Here, this therapeutic domain is therapeutically active. In one embodiment, the receptor undergoes constitutive endocytosis. In another embodiment, the therapeutic domain has therapeutic enzyme activity. This enzymatic activity is preferably an activity whose deficiency (in a cell or a specific compartment of a cell) is associated with a human disease (eg lysosomal storage disease).

  In a further aspect, the present invention relates to nucleic acids encoding therapeutic proteins and cells (eg, mammalian cells, insect cells, yeast cells, protists, or bacteria) containing these nucleic acids. The invention also provides these cells with conditions that allow expression of the protein (eg, in the context of in vitro culture or by maintaining the cells in the mammal). Provide a method to generate. The protein can then be recovered (eg when produced in vitro) or used without intervention of a recovery step (eg when produced in vivo in a patient). Thus, the present invention also allows for in vivo protein synthesis by administering a therapeutic protein (by injection, in situ synthesis, etc.) or by administering a nucleic acid encoding this protein. Or a method of treating a patient by administering a cell comprising a nucleic acid encoding the protein. In one embodiment, the method includes synthesizing a targeted therapeutic and administering the targeted therapeutic to a patient, the targeted therapeutic comprising mammalian lysosomes, and mannose. A therapeutic agent that is therapeutically active at the targeting moiety that binds the human cation-independent mannose-6-phosphate receptor in a -6-phosphate independent manner is included. This method also includes the step of identifying the targeting moiety (eg, by recombinant display technology (eg, phage display, bacterial display or yeast two hybrids) or by screening the library for the required binding properties). Include. In another embodiment, the method provides (eg, on a computer) a molecular model that defines a three-dimensional shape corresponding to at least a portion of human IGF-II; at least a portion of IGF-II (eg, Identifying a candidate IGF-II analog having a three-dimensional shape corresponding to amino acids 48-55) and active in mammalian lysosomes, directly or indirectly to this candidate IGF-II analog Producing a therapeutic agent to be bound. The method can also include determining whether the candidate IGF-II analog binds to a human cation-independent M6P receptor.

  The present invention also provides a method of producing therapeutic proteins that are targeted to lysosomes and / or cross the blood brain barrier and possess a prolonged half-life in circulation in mammals. The method includes producing an underglycosylated therapeutic protein. As used herein, “incomplete glycosylation” refers to the omission of one or more carbohydrate structures normally present when a protein is produced in mammalian cells (eg, CHO cells). Refers to a protein that is removed, modified, or masked, thereby extending the half-life of the protein in a mammal. Thus, proteins can actually be incompletely glycosylated by the absence of one or more carbohydrate structures that promote clearance from the circulation, or are functional by modifying or masking one or more carbohydrate structures. Can be incompletely glycosylated. For example, the structure is (ii) bound by the addition of one or more additional moieties (eg, carbohydrate groups, phosphate groups, alkyl groups, etc.) that interfere with the recognition of the structure by the mannose receptor or asialoglycoprotein receptor. By covalent or non-covalent binding of glycoproteins to moieties (eg, lectins or extracellular portions of mannose receptors or asialoglycoprotein receptors), or (iii) carbohydrates It can be masked by any other modification to the polypeptide or carbohydrate portion of the glycoprotein that reduces its clearance from the blood by masking the presence of all or part of the structure.

  In one embodiment, the therapeutic protein comprises a peptide targeting moiety (eg, IGF-I, IGF-II, or that moiety effective to bind to a target receptor), wherein the targeting moiety is a conventional Produced in a host (eg, a bacterium or yeast) that does not glycosylate the protein like mammalian cells (eg, Chinese hamster ovary (CHO) cells) do. For example, a protein produced by a host cell may lack a terminal mannose residue, fucose residue and / or N-acetylglucosamine residue (these are identified by the mannose receptor) or may be completely May be non-glycosylated. In another embodiment, a therapeutic protein that can be produced in a mammalian cell or other host is chemically or enzymatically treated to produce one or more carbohydrate residues (eg, one or more mannose residues, fucose). Residues and / or N-acetylglucosamine residues) or one or more carbohydrate residues are modified or masked. Such modification or masking can reduce the binding of therapeutic proteins to the hepatic mannose receptor and / or hepatic asialoglycoprotein receptor. In another embodiment, one or more potential glycosylation sites are removed by mutation of a nucleic acid encoding the targeted therapeutic protein, thereby being synthesized in a mammalian cell or other cell that glycosylates the protein. May reduce protein glycosylation.

(Detailed description of the invention)
As used herein and unless otherwise specified, “glycosylation-independent lysosomal targeting” and “GILT” refer to lysosomal targeting that is mannose-6-phosphate independent.

  As used herein, a “GILT construct” refers to a construct comprising a mannose-6-phosphate independent lysosomal targeting and therapeutic moiety that is effective in mammalian lysosomes.

  As used herein, “GUS” refers to β-glucuronidase (an exemplary therapeutic moiety).

  As used herein, “GUSΔC18” refers to GUS with the C-terminal 18 amino acids deleted and potential proteolytic sites removed.

  As used herein, “GUS-GILT” refers to a GILT construct having a GUS linked to an IGF-II targeting moiety.

  All references to amino acid positions in IGF-II refer to positions in adult human IGF-II. Thus, for example, positions 1, 2 and 3 are occupied by alanine, tyrosine and arginine, respectively.

  As used herein, GILTΔ1-7 refers to an IGF-II targeting moiety lacking the N-terminal 7 amino acids.

  As used herein, GUSΔC18-GILTΔ1-7 refers to a fusion protein in which GUSΔC18 is fused to the N-terminus of GILTΔ1-7.

  The present invention provides a targeted therapeutic that, when provided outside of a cell, penetrates into the cell and localizes to the subcellular compartment in which the targeted therapeutic is active. Facilitates treatment. This targeted therapeutic binds to at least a therapeutic agent and a targeting moiety (eg, a subcellular targeting domain of a protein) or human cation-independent mannose-6-phosphate receptor for lysosomal targeting. Means (eg, proteins, peptides, peptide analogs or organic chemicals).

(Binding between therapeutic agent and targeting moiety)
The therapeutic agent and targeting moiety are necessarily bound directly or indirectly. In one embodiment, the therapeutic agent and targeting moiety are non-covalently linked. This binding is preferably stable at pH 7.4 or about pH 7.4. For example, the targeting moiety can be biotinylated and bind to avidin conjugated to a therapeutic agent. Alternatively, the targeting moiety and therapeutic agent can each be bound to different subunits of the multimeric protein (eg, as a fusion protein). In another embodiment, the targeting moiety and therapeutic agent are crosslinked to each other (eg, using a chemical crosslinking agent).

  In preferred embodiments, the therapeutic agent is fused to the targeting moiety as a fusion protein. The targeting moiety can be the amino terminus, carboxy terminus of the fusion protein, or inserted within the sequence of the therapeutic agent at a position where the presence of the targeting moiety does not unduly interfere with the therapeutic activity of the therapeutic agent. Can be done.

  When the therapeutic agent is a heteromeric protein, one or more of the subunits can be bound to a targeting moiety. Hexosaminidase A (eg, a lysosomal protein that acts in Tay-Sachs disease) contains an α subunit and a β subunit. The α subunit, β subunit, or both can be bound to a targeting moiety in accordance with the present invention. For example, if the α subunit is bound to a targeting moiety and is co-expressed with the β subunit, the active complex is appropriately formed and targeted (eg, to lysosomes).

  For the targeting of therapeutics to lysosomes, the therapeutic agent is via an interaction that is disrupted by lowering the pH to pH 7.4 or pH about 7.4 to about 5.5 or about 5.5. Can be bound to a targeting moiety. This targeting moiety binds to a receptor external to the cell; the selected receptor is the receptor that undergoes endocytosis and passes through the late endosome, which has a pH of about 5.5. Thus, in late endosomes, the therapeutic agent dissociates from the targeting moiety and heads to the lysosome, where the therapeutic agent acts. For example, the targeting moiety can be chemically modified to incorporate a chelating agent (eg, EDTA, EGTA, or trinitrilotriacetic acid) that binds tightly to a metal ion (eg, nickel). This targeting moiety (eg, GUS) can be expressed as a fusion protein with six histidine tags (eg, at the amino terminus, the carboxy terminus, or in a flexible loop accessible to the surface). At pH 7.4 or about pH 7.4, the six histidine tags are substantially deprotonated and bind with high affinity to metal ions (eg, nickel). At pH 5.5 or about pH 5.5, the six histidine tags are substantially protonated leading to the release of nickel and consequently releasing the therapeutic agent from the targeting moiety.

(Therapeutic agent)
While the methods and compositions of the present invention are useful for generating and delivering any therapeutic agent to the subcellular compartment, the present invention provides for the delivery of gene products for treating metabolic diseases. It is particularly useful.

  Preferred LSD genes are shown in Table 1, and preferred genes associated with Golgi or endoplasmic reticulum defects are shown in Table 2. In a preferred embodiment, the wild type LSD gene product is delivered to a patient suffering from the same LSD gene deficiency. In an alternative embodiment, a functional sequence or species variant of the LSD gene is used. In a further embodiment, genes encoding different enzymes that can rescue LSD gene defects are used according to the methods of the invention.

１つの特に好ましい治療剤は、ゴーシェ病のための効果的な酵素置換治療剤としてＧｅｎｚｙｍｅによって現在製造される、グルコセレブロシダーゼである。現在、その酵素は、曝露されたマンノース残基（これは、マクロファージ系列の細胞に、タンパク質を特異的に標的化する）によって調製される。１型ゴーシェ病患者の主な病理学は、グルコセレブロシドを蓄積するマクロファージに起因するが、他の細胞型に対してグルコセレブロシダーゼを送達することには治療効果が存在し得る。本発明を使用するリソソームへのグルコセレブロシダーゼの標的化は、複数の細胞型に対して薬剤を標的化し、そして他の調製物と比較して治療効果を有し得る。

One particularly preferred therapeutic agent is glucocerebrosidase, currently manufactured by Genzyme as an effective enzyme replacement therapeutic for Gaucher disease. Currently, the enzyme is prepared by exposed mannose residues, which specifically target proteins to cells of the macrophage lineage. The main pathology of patients with type 1 Gaucher disease is due to macrophages that accumulate glucocerebrosides, but there may be a therapeutic effect in delivering glucocerebrosidase to other cell types. Targeting glucocerebrosidase to lysosomes using the present invention targets drugs to multiple cell types and may have a therapeutic effect compared to other preparations.

  Another preferred therapeutic agent is acid α-glucosidase (GAA), a lysosomal enzyme that is deficient in Pomp disease (see discussion in Example 13B). Pomp disease, also known as acid maltase deficiency (AMD), type II glycogen storage disease (GSDII), type II glycogen storage disease, or GAA deficiency, is a lysosomal storage disease caused by insufficient activity of GAA It is. GAA is an enzyme that hydrolyzes α1-α4 linkages in maltose and other linear oligosaccharides (including the outer branch of glycogen), thereby destroying excess glycogen in the lysosome (Hirschhorn et al. (2001) The Metabolic and Molecular Basis of Inherited Disease, Scriver et al. (2001) McGraw-Hill: New York, p. 3389-3420). The loss of enzyme activity is caused by various missense or nonsense mutations in the gene encoding GAA. As a result, glycogen accumulates in the lysosomes of all cells of patients with Pomp disease. In particular, glycogen accumulation is most prominent in lysosomes of myocardium and skeletal muscle, liver, and other tissues. Accumulated glycogen ultimately impairs muscle function. In the most severe form of Pomp disease, death by cardiac respiratory failure occurs before the age of two.

  Currently, there are no approved treatments that can cure or delay the progression of Pomp. Currently, enzyme replacement therapy in clinical trials requires that administered recombinant GAA is taken up by cells and transferred to lysosomes in these cells in muscle and liver tissues. However, recombinant GAA produced in engineered CHO cells and in the milk of transgenic rabbits currently used for enzyme replacement therapy contains very little M6P (Van Hove et al. (1996) Proc Natl Acad Sci U S A, 93 (1): 65-70; and US Pat. No. 6,537,785). Thus, M6P-dependent delivery of recombinant GAA to lysosomes is not efficient and requires high doses and frequent injections. In contrast, the present invention allows M6P-independent targeting of GAA patients to lysosomes as described in more detail in Example 13B.

(Subcellular targeting domain)
The present invention allows the targeting of therapeutic agents to lysosomes using a protein, or an analog of a protein that specifically binds to the cellular receptor of this protein. The exterior of the cell surface is topologically equivalent to the endosome, lysosome, Golgi, and endoplasmic reticulum compartment. Thus, endocytosis of the molecule through interaction with the appropriate receptor allows the transport of the molecule to any of these compartments without traversing the membrane. Genetic defects result in a loss of specific enzyme activity in any of these compartments, and delivery of therapeutic proteins can be achieved by tagging this protein with a ligand for the appropriate receptor.

  Multiple pathways have been characterized that direct receptor binding proteins from the plasma membrane to the Golgi and / or endoplasmic reticulum. Thus, for example, by using a targeting moiety derived from SV40, cholera toxin, or plant ricin toxin, each of which selects one or more of these subcellular transport pathways, Can be targeted to a desired location within. In each case, uptake is initiated by the binding of the substance to the outside of the cell. For example, SV40 binds to MHC class I receptors, cholera toxin binds to GM1 ganglioside molecules, and ricin binds to glycolipids and glycoproteins with terminal galactose on the surface of the cell. Following this initial step, the molecule reaches the ER by various pathways. For example, SV40 causes vesicle endocytosis and reaches the ER in a two-step process that bypasses the Golgi, whereas cholera toxin causes vesicle endocytosis but before the ER is reached Cross.

  When targeting moieties associated with cholera toxin or ricin are used, it is important that the toxicity of cholera toxin or ricin is avoided. Both cholera toxin and ricin are heteromeric proteins, and their cell surface binding domain and catalytic activity responsible for toxicity reside on separate polypeptides. Thus, targeting moieties can be constructed that include a receptor binding polypeptide, but no polypeptide responsible for toxicity. For example, in the case of ricin, the B subunit has galactose binding activity responsible for protein internalization and can be fused to a therapeutic protein. If further presence of the A subunit improves subcellular localization, a mutated version of the A chain that is properly folded but catalytically inactive (mutein) can be provided with the B subunit-therapeutic agent fusion protein. .

  The protein delivered to the Golgi can be transported to the endoplasmic reticulum (ER) via the KDEL receptor, which recovers the ER targeted protein escaping to the Golgi. Thus, including a KDEL motif at the end of the targeting domain that orients the therapeutic protein to the Golgi allows for subsequent localization to the ER. For example, targeting moieties that bind to the cation-independent M6P receptor at both pH 7.4 or about pH 7.4, and pH 5.5 or about pH 5.5 (eg, antibodies, or phage display, yeast two hybrids, chips Peptides identified by high-throughput screening, such as chip-based and solution-based assays) enable the targeting of therapeutic agents to the Golgi; and the addition of the KDEL motif allows targeting to the ER To do.

(Lysosome targeting moiety)
The present invention allows targeting of therapeutic agents to lysosomes. Targeting can occur, for example, through binding of plasma membrane receptors that later pass through lysosomes. Alternatively, targeting can occur through the binding of protoplasmic receptors that later pass through late endosomes; then the therapeutic agent can migrate from late endosomes to lysosomes. A preferred lysosomal targeting mechanism involves binding to the cation-independent M6P receptor.

(Cation-independent M6P receptor)
The cation-independent M6P receptor is a 275 kDa single chain transmembrane glycoprotein that is ubiquitously expressed in mammalian tissues. It is one of two mammalian receptors that bind to M6P; the second is called the cation-dependent M6P receptor. This cation-dependent M6P receptor requires a divalent cation for M6P binding; a cation-independent M6P receptor is not required. These receptors play an important role in lysosomal enzyme transport by recognition of the M6P moiety in high mannose carbohydrates on lysosomal enzymes. The extracellular domain of the cation-independent M6P receptor contains 15 homologous domains (“repeats”) that bind various groups of ligands at distinct locations on the receptor.

  The cation-independent M6P receptor contains two binding sites for M6P: one is located in repeats 1-3 and the other is located in repeats 7-9. This receptor binds to a monovalent M6P ligand with a dissociation constant in the μM range, while remaining bound to a divalent M6P ligand with a dissociation constant in the nM range, probably due to receptor oligomerization. The uptake of IGF-II by the receptor is enhanced by the concomitant binding of multivalent M6P ligands such as lysosomal enzymes to the receptor.

  The cation-independent M6P receptor also contains binding sites for at least three separate ligands, which can be used as targeting moieties. The cation-independent M6P receptor binds to IGF-II primarily through interaction with repeat 11 at pH 7.4 or about pH 7.4 with a dissociation constant of about 14 nM. Consistent with its function in targeting IGF-II to lysosomes, the dissociation constant increases about 100-fold at pH 5.5 or about pH 5.5, promoting dissociation of IGF-II in acidic late endosomes. This receptor can bind to the high molecular weight O-glycosylated IGF-II form.

  A further useful ligand for the cation-independent M6P receptor is retinoic acid. Retinoic acid binds to the receptor with a dissociation constant of 2.5 nM. Affinity photolabeling of the cation-independent M6P receptor with retinoic acid does not interfere with IGF-II or M6P binding to the receptor, indicating that retinoic acid binds to separate sites on the receptor. Show. Retinoic acid binding to the receptor alters the subcellular distribution of receptors with greater accumulation of the receptor in cytoplasmic vesicles and also enhances the uptake of M6P-modified β-glucuronidase. Retinoic acid has a photoactivatable moiety that can be used to bind to a therapeutic agent without interfering with its ability to bind to the cation-independent M6P receptor.

  This cation-independent M6P receptor also binds to the urokinase-type plasminogen receptor (uPAR) with a dissociation constant of 9 μM. uPAR is a GPI-anchor receptor on the surface of most cell types, where it functions as an adhesion molecule and in the proteolytic activation of plasminogen and TGF-β. Binding of uPAR to the CI-M6P receptor targets the uPAR receptor to the lysosome, thereby altering its activity. Thus, the fusion of uPAR's extracellular domain, or a portion thereof, capable of binding to the cation-independent M6P receptor to a therapeutic agent allows targeting of the drug to lysosomes.

(IGF-II)
In preferred embodiments, the lysosomal targeting moiety is a protein, peptide, or other moiety that binds to a cation-independent M6P / IGF-II receptor in a mannose-6-phosphate-independent manner. Advantageously, this embodiment mimics the normal biological mechanism for LSD protein uptake but does this in a manner that is independent of mannose-6-phosphate.

  For example, by fusing DNA encoding a mature IGF-II polypeptide to the 3 'end of an LSD gene cassette, a fusion protein is made that can be taken up by various cell types and transported to lysosomes. Alternatively, DNA encoding the precursor IGF-II polypeptide can be fused to the 3 ′ end of the LSD gene cassette; the precursor has a carboxy-terminal portion that is cleaved in mammalian cells to yield the mature IGF-II polypeptide. Although included, the IGF-II signal peptide is preferably omitted (or moved to the 5 ′ end of the LSD gene cassette). This method has many advantages over methods involving glycosylation, including simplicity and cost efficiency, because once the protein is isolated no further modifications need to be made.

  IGF-II is preferably specifically targeted to the M6P receptor. Particularly useful are mutations in IGF-II polypeptides that result in proteins that bind to the M6P receptor with high affinity but no longer bind to the other two receptors with appreciable affinity. IGF-II can also be modified to minimize binding to serum IGF-binding protein to avoid sequestration of the IGF-II / GILT construct (Baxter (2000) Am. J. Physiol Endocrinol Metab. 278). (6): 967-76). Many studies have identified residues in IGF-I and IGF-II that are required to bind to IGF-binding proteins. Constructs with mutations at these residues can be screened for retention of high affinity binding to the M6P / IGF-II receptor and for reduced affinity for IGF-binding proteins. For example, replacement of IGF-II PHE26 with SER reduces the affinity of IGF-II for IGFBP-1 and IGFBP-6 without affecting binding to the M6P / IGF-II receptor. Has been reported (Bach et al. (1993) J. Biol. Chem. 268 (13): 9246-54). Other substitutions such as PHE19 to SER and GLU9 to LYS may also be advantageous. Similar mutations produce large reductions in IGF-BP binding, either separately or in combination, in regions in IGF-I that are highly conserved with IGF-II (Magee et al. (1999) Biochemistry 38 (48): 15863. -70).

  An alternative approach is to identify the minimal region of IGF-II that can bind with high affinity to the M6P / IGF-II receptor. Residues involved in IGF-II that bind to the M6P / IGF-II receptor are mostly clustered on one face of IGF-II (Tarazawa et al. (1994) EMBO J. 13 (23): 5590-7). . The tertiary structure of IGF-II is usually maintained by three intramolecular disulfide bonds, but peptides incorporating the amino acid sequence on the M6P / IGF-II receptor that bind to the surface of IGF-II are properly folded, And can be designed to have binding activity. Such minimal binding peptides are highly preferred targeting moieties. Peptides designed based on the region around amino acids 48-55 can be tested for binding to the M6P / IGF-II receptor. Alternatively, a random library of peptides can be screened for the ability to bind to the M6P / IGF-II receptor, either through a yeast two-hybrid assay or through a phage display-type assay.

(Blood brain barrier)
One challenge in the treatment for lysosomal storage diseases is that many of these diseases have significant neurological involvement. Therapeutic enzymes administered to the bloodstream generally do not cross the blood brain barrier and therefore cannot alleviate the neurological symptoms associated with this disease. However, IGF-II has been reported to promote transport across the blood brain barrier via transcytosis (Bickel et al. (2001) Adv. Drug Deliv. Rev. 46 (1-3): 247- 79). Thus, a properly designed GILT construct should be able to cross the blood brain barrier and provides for the first time a means of treating neurological symptoms associated with lysosomal storage diseases. This construct can be tested using GUS minus mice as described in Example 12. For further details regarding the design, construction and testing of targeted therapeutic agents that can reach the nerve tissue from the blood, see US Pat. No. 60 / 329,650 (filed Oct. 16, 2001) and US Pat. No. 10/136, No. 639 (filed Apr. 30, 2002).

(Structure of IGF-II)
The NMR structure of IGF-II has been solved by two groups (Terasawa et al. (1994) EMBO J. 13 (23): 5590-7; Torres et al. (1995) J. Mol. Biol. 248 (2): 385-401) (See, eg, Protein Data Bank Record 1 IGL). The general features of the IGF-II structure are similar to IGF-I and insulin. The A and B domains of IGF-II correspond to the A and B chains of insulin. Secondary structural features include an alpha helix from residues 11-21 of the B region connected by a retroturn turn of residues 22-25 to a short beta chain in residues 26-28. Residues 25-27 appear to form a small antiparallel β sheet; residues 59-61 and residues 26-28 may also be involved in intermolecular β sheet formation. In the A domain of IGF-II, the α helix spans residues 42-49 and 53-59 which are arranged in an antiparallel configuration perpendicular to the B-domain helix. Hydrophobic clusters formed by two of the three disulfide bridges and conserved hydrophobic residues stabilize these secondary structural features. The N-terminus and C-terminus remain poorly defined as the region between residues 31-40.

  IGF-II binds to the IGF-II / M6P and IGF-I receptors with relatively high affinity, and binds to the insulin receptor with lower affinity. IGF-II also interacts with many serum IGFBPs.

(Binding to IGF-II / M6P receptor)
Substitution of residues 48-50 (Phe Arg Ser) of IGF-II with the corresponding residue from insulin (Thr Ser Ile), or correspondence of residues 54-55 (Ala Leu) from IGF-I Substitution with a residue (Arg Arg) results in decreased binding to the IGF-II / M6P receptor but retains binding to IGF-I and the insulin receptor (Sakano et al. (1991) J. Biol. Chem). 266 (31): 20626-35).

  IGF-I and IGF-II share the same sequence and structure in the region of residues 48-50, but have a 1000-fold difference in affinity for the IGF-II receptor. This NMR structure represents the structural difference between IGF-I and IGF-II in the region of residues 53-58 of IGF-II (residues 54-59 of IGF-I): Better defined in IGF-II than in IGF-I, and unlike IGF-I, there is no bending in the skeleton around residues 53 and 54 (Torres et al. (1995) J. Mol. Biol. 248 (2): 385-401). This structural difference correlates with the substitution of Ala54 and Leu55 in IGF-II with Arg55 and Arg56 in IGF-I. Binding to the IGF-II receptor is directly broken by the presence of charged residues in this region, or changes in structure caused by charged residues cause changes in binding to the IGF-II receptor. Any of what happens is possible. In either case, substitution of two Arg residues in IGF-I with uncharged residues resulted in higher affinity for the IGF-II receptor (Cacciali et al. (1987) Pediatrician 14 (3). : 146-53). Thus, the presence of positively charged residues at these positions is associated with a loss of binding to the IGF-II receptor.

  IGF-II binds to repeat 11 of the cation-independent M6P receptor. Indeed, a mini-receptor, in which only repeat 11 is fused to the transmembrane and cytoplasmic domains of the cation-independent M6P receptor, can bind to IGF-II (with approximately one-tenth the affinity of the full-length receptor) And mediate the internalization of IGF-II and its delivery to lysosomes (Grimme et al. (2000) J. Biol. Chem. 275 (43): 33697-33703). The structure of domain 11 of the M6P receptor is known (protein database accession numbers 1GP0 and 1GP3; Brown et al. (2002) EMBO J. 21 (5): 1054-1062). This putative IGF-II binding site is a hydrophobic pocket that is thought to interact with the hydrophobic amino acids of IGF-II; candidate amino acids for IGF-II include leucine 8, phenylalanine 48, alanine 54, and leucine 55. Can be mentioned. While repeat 11 is sufficient for IGF-II binding, the construct contains a larger portion of the cation-independent M6P receptor (eg, repeats 10-13, or 1-15) and generally has a greater affinity. And binds to IGF-II in an increased pH-dependent manner (see, for example, Linnell et al. (2001) J. Biol. Chem. 276 (26): 23986-23991).

(Binding to IGF-I receptor)
Substitution of IGF-II residue Tyr27 with Leu, substitution of Leu43 with Val or substitution of Ser26 with Phe increases the affinity of IGF-II for the IGF-I receptor by 94, 56 and 4 respectively. (Tores et al. (1995) J. Mol. Biol. 248 (2): 385-401). Deletion of residues 1-7 of human IGF-II results in a 30-fold decrease in affinity for human IGF-I receptor and a concomitant 12-fold increase in affinity for rat IGF-II receptor. (Hashimoto et al. (1995) J. Biol. Chem. 270 (30): 18013-8). The NMR structure of IGF-II shows that Thr7 is located near residues 48Phe and 50Ser and near the 9Cys-47Cys disulfide bridge. It is believed that Thr7 interaction with these residues can stabilize the flexible N-terminal hexapeptide required for IGF-I receptor binding (Terazawa et al. (1994) EMBO J. 13 (23). 5590-7). At the same time, this interaction can regulate binding to the IGF-II receptor. Shortening of the C-terminus of IGF-II (residues 62-67) also appears to reduce IGF-II's affinity for the IGF-I receptor by a factor of 5 (Roth et al. (1991) Biochem. Biophys. Res. Commun. 181 (2): 907-14).

(Deletion mutation of IGF-II)
The binding surface for IGF-I and the cation-independent M6P receptor is on a separate face of IGF-II. Based on structural and mutational data, a functional cation-independent M6P binding domain can be constructed that is substantially smaller than human IGF-II. For example, amino terminal amino acids 1-7 and / or carboxy terminal residues 62-67 can be deleted or substituted. In addition, amino acids 29-40 may appear to be removed or replaced without altering the folding of the remainder of the polypeptide or binding to the cation-independent M6P receptor. Thus, a targeting moiety comprising amino acids 8-28 and 41-61 can be constructed. These stretches of amino acids can probably be joined directly or separated by a linker. Alternatively, amino acids 8-28 and 41-61 can be provided on separate polypeptide chains. A comparable domain of insulin that is homologous to IGF-II and has a tertiary structure close to that of IGF-II, even when present in separate polypeptide chains, is suitable for re-appropriation to the appropriate tertiary structure. Has sufficient structural information to allow folding (Wang et al. (1991) Trends Biochem. Sci. 279-281). Thus, for example, amino acids 8-28, or conservative substitution variants thereof can be fused to a therapeutic agent; the resulting fusion protein can be mixed with amino acids 41-61 or a conservative substitution variant thereof and administered to a patient Can be done.

(Binding to IGF binding protein)
IGF-II and related constructs can be modified to reduce affinity for IGFBP, thereby increasing the bioavailability of the tagged protein.

  Substitution of IGF-II residue phenylalanine 26 with serine reduces binding to IGFBP 1-5 by 1/5 to 1/75 (Bach et al. (1993) J. Biol. Chem. 268 (13): 9246-54). Substitution of residues 48-50 of IGF-II with threonine-serine-isoleucine reduces IGFBP binding to the majority (Bach et al. (1993) J. Biol. Chem. 268 (13 ): 9246-54); however, these residues are also important for binding to the cation-independent mannose-6-phosphate receptor. A Y27L substitution that disrupts binding to the IGF-I receptor prevents the formation of tertiary complexes with IGFBP3 and acid labile subunits (Hashimoto et al. (1997) J. Biol. Chem. 272 (44): 27936. -42); this tertiary complex accounts for the majority of IGF-II in the circulation (Yu et al. (1999) J. Clin. Lab Anal. 13 (4): 166-72). Deletion of the first 6 residues of IGF-II also prevents IGFBP binding (Luthi et al. (1992) Eur. J. Biochem. 205 (2): 483-90).

  Studies of the interaction of IGF-I with IGFBP further indicate that substitution of phenylalanine 16 with serine reduces IGFBP binding by a factor of 1/40 to 1/300 while not affecting secondary structure. (Magee et al. (1999) Biochemistry 38 (48): 15863-70). The change of glutamate 9 to lysine also resulted in a significant decrease in IGFBP binding. Furthermore, the lysine 9 / serine 16 double mutation showed the lowest affinity for IGFBP. Although these mutants have not been previously tested in IGF-II, the conservation of sequence between this region of IGF-I and IGF-II indicates that a similar mutation (glutamate 12 lysine / phenylalanine 19 serine) is IGF- It suggests that a similar effect is observed when made during II.

(IGF-II homolog)
The amino acid sequence of human IGF-II or a portion thereof that affects binding to the cation-independent M6P receptor determines whether the candidate sequence has sufficient amino acid similarity to have a reasonable chance of success in the methods of the invention. It can be used as a reference sequence for determining. Preferably, the variant sequence is at least 70% similar or 60% identical to human IGF-II, more preferably at least 75% similar or 65% identical, and most preferably 80% similar Or 70% identical.

  In order to determine whether a candidate peptide region has the necessary percentage of similarity or identity to human IGF-II, this candidate amino acid sequence and human IGF-II are firstly derived from Henikoff and Henikoff (1992). PNAS 89: 10915-10919 in combination with the BLOSUM62 substitution matrix described in FIG. Mol. Biol. 147: 195-197. For the present invention, a suitable value for the gap insertion penalty is -12 and a suitable value for the gap extension penalty is -4. Computer programs that implement alignment using the Smith-Waterman algorithm and the BLOSUM62 matrix (eg, the GCG program suite (Oxford Molecular Group, Oxford, England)) are commercially available and widely used by those skilled in the art. .

  Once the alignment between the candidate sequence and the reference sequence is created, a percent similarity score can be calculated. The individual amino acids of each sequence are sequentially compared according to their similarity to each other. If the value in the BLOSUM62 matrix corresponding to two aligned amino acids is zero or negative, the pairwise similarity score is zero; otherwise, the pairwise similarity score is 1.0 is there. The raw similarity score is the sum of the aligned pairwise similarity scores of the amino acids. The raw score is then normalized by dividing this by the number of amino acids in the smaller of the candidate or reference sequences. This normalized raw score is a percent similarity. Alternatively, the aligned amino acids of each sequence are again compared sequentially to calculate percent identity. If the amino acids are not identical, the pairwise identity score is zero; otherwise, the pairwise identity score is 1.0. The raw identity score is the sum of identically aligned amino acids. The raw score is then normalized by dividing this by the number of amino acids in the smaller of the candidate or reference sequences. This normalized raw score is the percent identity. Insertions and deletions are ignored for purposes of calculating percent similarity and percent identity. Thus, gap penalties are not used in this calculation, but they are used in the initial alignment.

(IGF-II structure analog)
Known structures of human IGF-II and cation-independent M6P receptors are described in IGF using computer aided design principles as discussed in US Pat. Nos. 6,226,603 and 6,273,598. -Allows the design of II analogs and other cation-independent M6P receptor binding proteins. For example, known atomic coordinates of IGF-II can be obtained from conventional computer modeling programs (eg, INSIGHT II, DISCOVER, or DELPHI, commercially available from Biosym, Technologies Inc., or QUANTA or CHARMM, commercially available from Molecular Simulations, Inc.). Can be provided to a computer equipped with These and other software programs allow molecular structure analysis and simulation to estimate the effects of molecular changes on structure and intermolecular interactions. For example, the software can be used to identify modified analogs that have the ability to form additional intermolecular hydrogen or ionic bonds that improve the affinity of the analog for the target receptor.

  The software also allows the design of peptides and organic molecules with structural and chemical features that mimic the same features shown on at least a portion of the surface of the cation-independent M6P receptor binding surface of IGF-II . Since the main contribution to the receptor binding surface is the spatial arrangement of chemically interacting moieties present in the side chains of amino acids that together define the receptor binding surface, preferred embodiments of the present invention are IGFs. A framework that possesses chemically interacting moieties in a spatial relationship that mimics the spatial relationship of the chemical moieties located to the amino acid side chains that make up the cation-independent M6P receptor binding surface of II Designing and making synthetic organic molecules having Preferred chemical moieties include, but are not limited to, chemical moieties defined by the amino acid side chains of the amino acids that make up the cation-independent M6P receptor binding surface of IGF-II. Thus, it will be appreciated that the receptor binding surface of an IGF-II analog need not include an amino acid residue disposed therein, but must include a chemical moiety.

  For example, one of skill in the art using conventional computer programs in identifying related chemical groups can design small molecules with receptor-interacting chemical moieties placed on a suitable carrier framework. Useful computer programs are described, for example, in Dixon (1992) Tibtech 10: 357-363; Tschinke et al. Med. Chem. 36: 3863-3870; and Eisen et al. (1994) Proteins: Structure, Function, and Genetics 19: 199-221, the disclosures of which are incorporated herein by reference.

  One particular computer program named “CAVEAT” searches a database (eg, Cambridge Structural Database) for structures having the desired spatial orientation of the chemical moiety (Bartlett et al. (1989) “Molecular Recognition: Chemical and Biological”. “Problems” (Roberts, S. M.) 182-196). This CAVEAT program is used to design tendamistat analogs based on the orientation of selected amino acid side chains in the three-dimensional structure of tendamistat (a 74-residue inhibitor of α-amylase). (Bartlett et al. (1989) supra).

  Alternatively, in identifying a series of analogs that mimic the cation-independent M6P receptor binding activity of IGF-II, one of skill in the art will assist one of skill in developing a quantitative structure-activity relationship (QSAR) and further A variety of computer programs can be used to further assist in the de novo design of morphogen analogs. Other useful computer programs are described in, for example, Connolly-Martin (1991) Methods in Enzymology 203: 587-613; Dixon (1992) supra; and Waszkowyz et al. Med. Chemm. 37: 3994-4002.

(Affinity of the targeting moiety)
Preferred targeting moieties bind to their target receptors with a dissociation constant of less than micromolar. In general, lower dissociation constants (eg, less than 10 ⁻⁷ M, less than 10 ⁻⁸ M, or less than 10 ⁻⁹ M) are increasingly preferred. The determination of the dissociation constant is preferably described in Linnell et al. Biol. Chem. 276 (26): 23986-23991 determined by surface plasmon resonance. Soluble forms of the extracellular domain of the target receptor (eg, cation-independent M6P receptor repeats 1-15) are made and immobilized to the chip via avidin-biotin interactions. By passing this targeting moiety over the chip and measuring the change in mass with respect to the chip surface, the rate and equilibrium constants are detected and calculated.

(Nucleic acids and expression systems)
Chimeric fusion proteins can be expressed in a variety of expression systems including in vitro translation systems and intact cells. Since modification of M6P is not essential for targeting, various expression systems that do not glycosylate proteins (including yeast systems, baculovirus systems and even prokaryotic systems such as E. coli) Suitable for expression of chemotherapeutic proteins. Indeed, non-glycosylated proteins generally have improved bioavailability because glycosylated proteins are rapidly cleared from circulation via binding to mannose receptors in the liver sinusoidal capillary endothelium. .

  Alternatively, the production of chimeric targeted lysosomal enzymes in mammalian cell expression systems produces proteins with multiple binding determinants for the cation-independent M6P receptor. Synergy between two or more cation-independent M6P receptor ligands (eg, M6P and IGF-II, or M6P and retinoic acid) can be utilized: multivalent ligands enhance binding to the receptor by receptor cross-linking It has been shown.

  In general, gene cassettes encoding chimeric therapeutic proteins contain sequences necessary for optimal expression, including changes in promoters, ribosome binding sites, introns, or coding sequences to optimize codon usage. It can be made for a specific expression system to incorporate. Since this protein is preferably secreted from the producer cell, the DNA encoding a signal peptide compatible with the expression system can be substituted for the endogenous signal peptide. For example, for the expression of β-glucuronidase and α-galactosidase A tagged with IGF-II in Leishmania, the DNA cassette encoding the Leishmania signal peptide (GP63 or SAP) is an endogenous signal to achieve optimal expression. It is inserted in place of the DNA encoding the peptide. In mammalian expression systems, an endogenous signal peptide can be utilized, but it may be desirable to use an IGF-II signal peptide when the IGF-II tag is fused at the 5 'end of the coding sequence.

  CHO cells are the preferred mammalian host for the production of therapeutic proteins. Traditional methods for achieving high yield expression from CHO cells are CHO cell lines deficient in dihydrofolate reductase (DHFR) (see, eg, CHO strain DUKX (O'Dell et al. (1998) Int. Biochem.Cell Biol.30 (7): 767-71)). This strain of CHO cells requires hypoxanthine and thymidine for growth. Co-transfection of genes overexpressed with the DHFR gene cassette, either on separate plasmids or on a single plasmid, allows selection for the DHFR gene, and generally also the selected recombinant protein It also allows the isolation of expressing clones. For example, plasmid pcDNA3 uses a cytomegalovirus (CMV) early region regulatory region promoter to drive expression of the gene of interest and pSV2DHFR to promote expression of DHFR. Subsequent exposure of cells with the recombinant gene cassette to increasing concentrations of the folate analog methotrexate results in gene copy number amplification of both the DHFR gene and the co-transfected gene.

  A preferred plasmid for eukaryotic expression in this system contains the gene of interest placed downstream of a strong promoter such as CMV. Introns can be placed 3 'to the gene cassette. The DHFR cassette can be driven by a second promoter from the same plasmid or from a separate plasmid. In addition, it can be useful to incorporate additional selectable markers (eg, neomycin phosphotransferase that confers resistance to G418) into the plasmid.

  Another CHO expression system (Ulmasov et al. (2000) PNAS 97 (26): 14212-14217) relies on amplification of the gene of interest using G418 instead of the DHFR / methotrexate system described above. A slightly neomycin phosphotransferase deficient pCXN vector driven by a weak promoter (see, eg, Niwa et al. (1991) Gene 108: 193-200) is a high copy number (> 300) transfectant in one step. Allows selection about.

  Alternatively, recombinant proteins can be produced in the human HEK293 cell line using an expression system based on the Epstein-Barr virus (EBV) replication system. It is composed of the EBV origin of replication oriP and the EBV ori binding protein EBNA-1. The binding of EBNA-1 to oriP causes replication and subsequent amplification of the extrachromosomal plasmid. This amplification then results in a high level of expression of the gene cassette housed in the plasmid. The plasmid containing oriP can be transfected into EBNA-1 transformed HEK293 cells (commercially available from Invitrogen) or alternatively it drives expression of EBNA-1 and contains an EBV oriP (eg, pCEP4 (Commercially available from Invitrogen)).

  E. In E. coli, the therapeutic protein is preferably secreted into the periplasmic space. This can be accomplished by replacing the DNA encoding the endogenous signal peptide of the LSD protein with a nucleic acid cassette encoding a bacterial signal peptide (eg, the ompA signal sequence). Expression can be driven by any of a number of strong inducible promoters (eg, lac, trp, or tac promoters). One suitable vector is pBAD / gIII (commercially available from Invitrogen), which uses the Gene III signal peptide and the araBAD promoter.

(Refolding in vitro)
One useful IGF-II targeting moiety has three intramolecular disulfide bonds. E. In E. coli, a GILT fusion protein (eg, GUS-GILT) can be constructed, which directs the protein to the periplasmic space. When fused to the C-terminus of another protein, IGF-II is E. coli. can be secreted in an active form in the periplasm of E. coli (Wadensten et al. (1991) Biotechnol. Appl. Biochem. 13 (3): 412-21). In order to facilitate optimal folding of the IGF-II moiety, appropriate concentrations of reduced and oxidized glutathione are preferably added to promote disulfide bond formation in the cellular environment. If the fusion protein having a disulfide bond is incompletely soluble, any insoluble material is preferably treated with a chaotropic agent (eg, urea) to solubilize the denatured protein and the appropriate disulfide bond To facilitate formation, refolded in a buffer with appropriate concentrations of reduced and oxidized glutathione, or other oxidizing and reducing agents (Smith et al. (1989) J. Biol. Chem. 264 ( 16): 9314-21). For example, IGF-I has been refolded using 6M guanidine-HCl and 0.1M tris (2-carboxyethyl) phosphine reducing agent for denaturation and reduction of IGF-II (Yang et al. (1999) J Biol.Chem.274 (53): 37598-604). Protein refolding was achieved in 0.1 M Tris-HCl buffer (pH 8.7) containing 1 mM oxidized glutathione, 10 mM reduced glutathione, 0.2 M KCl and 1 mM EDTA.

(Underglycosylation)
The targeted therapeutic protein is preferably incompletely glycosylated: one or more carbohydrate structures that are normally present when the protein is produced in mammalian cells are preferably omitted, removed, modified, or masked. Increase the half-life of proteins in mammals. Incomplete glycosylation can be achieved by a number of methods, some of which are illustrated in FIG. As shown in FIG. 1, a protein can actually be incompletely glycosylated (actually lacks one or more carbohydrate structures) or functions through modification or masking of one or more carbohydrate structures. Incompletely glycosylated. The protein, when synthesized as discussed in Example 14, can actually be incompletely glycosylated and fully glycosylated (as is the case in E. coli). Or partially non-glycosylated (as if synthesized in a mammalian system after breaking one or more glycosylation sites by site-directed mutagenesis) or non-mammalian glycosylation patterns Can have. Actual incomplete glycosylation can also be achieved by deglycosylation of the protein after synthesis. As discussed in Example 14, deglycosylation can be done via chemical or enzymatic treatment and results in complete deglycosylation or when only a portion of the carbohydrate structure is removed. Deglycosylated.

(In vivo expression)
Nucleic acid encoding a therapeutic protein (preferably a secreted therapeutic protein) can be advantageously provided directly to a patient afflicted with a disease, or, following ex vivo cells, the patient can receive this live cell. Can be provided by administration. In vivo gene therapy methods known in the art include providing purified DNA (eg, as in a plasmid), providing DNA in a viral vector, or liposomes or other vesicles (Eg, US Pat. No. 5,827,703 which discloses lipid carriers for use in gene therapy, and US Pat. No. 5,819, which provides adenoviral vectors useful for gene therapy). No. 6,281,010).

  Methods for treating diseases by transplanting cells that have been modified to express recombinant proteins are also well known. See, eg, US Pat. No. 5,399,346, which discloses a method for introducing nucleic acids into primordial human cells for introduction into humans. Although the use of human cells for ex vivo treatment is preferred in some embodiments, other cells (eg, bacterial cells) are transplanted into the patient's vasculature to continuously receive the therapeutic agent. May be released. See, for example, U.S. Pat. Nos. 4,309,776 and 5,704,910.

  The methods of the invention are particularly useful for targeting proteins directly to subcellular compartments without the need for purification steps. In one embodiment, the IGF-II fusion protein is expressed in a symbiotic or attenuated parasite administered to the host. The expressed IGF-II fusion protein is secreted by this organism, taken up by the host cell, and targeted to the lysosome of the host cell.

  In some embodiments of the invention, the GILT protein is delivered in situ via live Leishmania that secretes this protein into the lysosomes of infected macrophages. From this organelle, the protein leaves the cell and is taken up by neighboring cells that are not of the macrophage lineage. Thus, GILT tags and therapeutic agents necessarily remain intact while this protein is present in macrophage lysosomes. Thus, when GILT proteins are expressed in situ, they are preferably modified to ensure compatibility with the lysosomal environment. Human-β-glucuronidase (human “GUS”) (an exemplary therapeutic moiety) typically undergoes C-terminal peptide cleavage either in lysosomes or during transport to lysosomes (eg, residues in GUS). Between group 633 and residue 634). Thus, in embodiments where the GUS-GILT construct is expressed by Leishmania in macrophage lysosomes, human GUS is preferably modified to confer resistance to cleavage of this protein, or the residue following residue 633 is , Preferably simply removed from the GILT fusion protein. Similarly, IGF-II (an exemplary targeting moiety) is preferably modified to increase its resistance to proteolysis or identified by minimal binding peptides (eg, phage display or yeast two-hybrid) Is substituted for the wild type IGF-II moiety.

(Administration)
The targeted therapeutic produced in accordance with the present invention can be administered to a mammalian host by any route. Thus, if desired, administration can be oral or parenteral, including intravenous and intraperitoneal routes of administration. Further, administration can be by periodic injection of a bolus of therapeutic agent or can be made more continuous by intravenous or intraperitoneal administration from an external reservoir (eg, an iv bag). . In certain embodiments, the therapeutics of the invention can be pharmaceutical grade. That is, certain embodiments follow the purity standards and quality controls required for human administration. Veterinary applications are also within the intended meaning as used herein.

  Both formulations for veterinary and human medical use of the treatments according to the invention are typically associated with a pharmaceutically acceptable carrier therefor and other ingredients as required. Including such therapeutics. A carrier can be “acceptable” in the sense of being compatible with the other ingredients of the formulation and not injurious to the recipient thereof. In this regard, pharmaceutically acceptable carriers may include any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, and the like that are compatible with pharmaceutical administration. Intended. The use of such media and agents for pharmaceutically active substances is known in the art. Except insofar as any conventional media or agent is incompatible with the active compound, its use in the composition is contemplated. Supplementary active compounds (identified according to the present invention and / or known in the art) can also be incorporated into the compositions. The formulations can conveniently be provided in dosage unit form and can be prepared by any method well known in the art of pharmacy / microbiology. In general, some formulations are prepared by associating the treatment with a liquid carrier or a finely divided solid carrier or both, and then if necessary shaping the product into the desired formulation. Is done.

  A pharmaceutical composition of the invention is formulated to be compatible with its intended route of administration. Examples of routes of administration include oral or parenteral (eg, intravenous, intradermal, inhalation, transdermal (topical), transmucosal, and rectal administration). Solutions or suspensions used for parenteral, intradermal, or subcutaneous applications may contain the following components: sterile diluents (eg, water for injection, saline solution, non-volatile oil, polyethylene Glycols, glycerin, propylene glycol or other synthetic solvents); antibacterial agents (eg benzyl alcohol or methyl paraben); antioxidants (ascorbic acid or sodium bisulfite); chelating agents (eg ethylenediaminetetraacetic acid); buffers (eg , Acetate, citrate or phosphate) and agents for tonicity adjustment (eg sodium chloride or glucose). The pH can be adjusted with acids or bases, such as hydrochloric acid or sodium hydroxide.

  Useful solutions for oral or parenteral administration are described, for example, in Remington's Pharmaceutical Sciences, (Gennaro, A. Ed.), Mack Pub. , 1990, can be prepared by any method known in the pharmaceutical arts. Formulations for parenteral administration may also include glycocholate for buccal administration, methoxysalicylate for rectal administration, or citric acid for vaginal administration. The parenteral preparation can be enclosed in ampoules, disposable syringes or multiple dose vials made of glass or plastic. Suppositories for rectal administration also mix the drug with non-irritating excipients such as cocoa butter, other glycerides, or other compositions that are solid at room temperature and liquid at body temperature Can be prepared. The formulation can also include, for example, polyalkylene glycols (eg, polyethylene glycol), oils of plant origin, hydrogenated naphthalene, and the like. Formulations for direct administration can include glycerol and other compositions of high viscosity. Other potentially useful parenteral carriers for these therapeutics include ethylene-vinyl acetate copolymer particles, osmotic pumps, implantable infusion systems, and liposomes. Formulations for inhalation administration may contain, for example, lactose as an excipient, or in the form of an aqueous solution containing, for example, polyoxyethylene-9-lauryl ether, glycocholate and deoxycholate, or nasal drops As an oily solution for administration, or as a gel for application in the nasal cavity. A retention enema can also be used for rectal delivery.

  Formulations of the present invention suitable for oral administration are in discrete unit forms (eg, capsules, gelatin capsules, sachets, tablets, troches, or lozenges, each containing a predetermined amount of drug) In the form of powders or granules; in the form of a solution or suspension in an aqueous or non-aqueous liquid; or in the form of an oil-in-water emulsion or a water-in-oil emulsion. The treatment can also be administered in the form of a bolus, electuary or paste. A tablet may be made by compressing or molding the drug, optionally with one or more accessory ingredients. Compressed tablets can be placed in a free-flowing form (eg powders or granules) with a suitable machine, optionally with binders, lubricants, inert diluents, surfactants or dispersants. It can be prepared by mixing and compressing. Molded tablets may be made by molding in a suitable machine a mixture of the powdered drug moistened with an inert liquid diluent and a suitable carrier.

  Oral compositions generally include an inert diluent or an edible carrier. For the purpose of oral therapeutic administration, the active compound can be incorporated with excipients. An oral composition prepared with a fluid carrier for use as a mouthwash contains the compound in a fluid carrier and is orally applied, squeaked and vomited or swallowed. Pharmaceutically compatible binding agents, and / or adjuvant materials can be included as part of the composition. Tablets, pills, capsules, troches and the like may contain any of the following excipients or compounds of similar nature: binders (eg, microcrystalline cellulose, gum tragacanth or gelatin); excipients (eg, Starch or lactose); disintegrants (eg, alginic acid, primogel, or corn starch); lubricants (eg, magnesium stearate or Sterote); glidants (eg, colloidal silicon dioxide); sweeteners (eg, sucrose) Or saccharin); or flavoring (eg, peppermint, methyl salicylate, or orange flavor).

  Pharmaceutical compositions suitable for injectable use include sterile aqueous solutions (where water soluble) or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersion. Include. For intravenous administration, suitable carriers include physiological saline, bacteriostatic water, Cremophor ELTM (BASF, Parsippany, NJ) or phosphate buffered saline (PBS). In all cases, the composition can be sterile and can be fluid to the extent that easy syringability exists. The composition can be stable under the conditions of manufacture and storage and can be preserved against the contaminating action of microorganisms such as bacteria and fungi. The carrier can be a solvent or dispersion medium and include, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyethylene glycol, and the like), and suitable mixtures thereof. The proper fluidity can be maintained, for example, by the use of a coating, such as lecithin, by the maintenance of the required particle size in the case of dispersion and by the use of surfactants. Prevention of the action of microorganisms can be achieved by various antibacterial and antifungal agents (eg, parabens, chlorobutanol, phenol, ascorbic acid, thimerosal, etc.). In many cases, it will be preferable to include isotonic agents, for example, sugars, polyalcohols such as mannitol, sorbitol, and sodium chloride in the composition. Prolonged absorption of injectable compositions can be accomplished by including in the composition an agent that delays absorption, for example, aluminum monostearate and gelatin.

  Sterile injectable solutions can be prepared by incorporating the required amount of active compound in a suitable solvent having one or a combination of the ingredients listed above, followed by filter filtration if necessary. Do. Generally, dispersions are prepared by incorporating the active compound into a sterile vehicle, which contains the basic dispersion medium and other ingredients as required from the ingredients listed above. In the case of a sterile powder for a sterile injectable solution preparation, the method of preparation is vacuum drying to yield a powder of the active ingredient and any additional ingredient powder from the solution of this ingredient that has been previously filter sterilized. Process and lyophilization process.

  Formulations suitable for intra-articular administration can be in the form of a sterile aqueous therapeutic preparation, which can be in microcrystalline form, for example, in the form of an aqueous microcrystalline suspension. Liposomal formulations or biodegradable polymer systems can also be used to provide therapeutics for both intra-articular and intraocular administration.

  Formulations suitable for topical administration (including eye treatment) include liquid or semi-liquid preparations (eg, liniments, lotions, gels, sprays, oil-in-water or water-in-oil emulsions (eg, creams, ointments) Or a paste); or a solution or suspension (eg, drop)). Formulations for topical administration to the skin surface can be prepared by spraying a treatment with a dermatologically acceptable carrier (eg, lotion, cream, ointment or soap). In some embodiments, carriers capable of forming a film or layer on the skin are useful to localize the application and prevent removal. Where adhesion to the surface of the tissue is desired, the composition may contain a therapeutic dispersed in a fibrinogen-thrombin composition or other bioadhesive. The treatment can then be painted, sprayed or otherwise applied to the desired tissue surface. For topical administration to the internal tissue surface, the drug can be dispersed in a liquid tissue adhesive or other material known to enhance adsorption to the tissue surface. For example, hydroxypropylcellulose or fibrinogen / thrombin solutions can be used advantageously. Alternatively, a tissue coating solution (eg, a pectin-containing formulation) can be used.

  For inhalation treatments (eg, for asthma), inhalation of powders (self-propelled or spray formulations) dispensed with spray cans, nebulizers, or atomizers can be used. Such formulations may be in the form of a finely divided powder or a self-propelled powder dispensing formulation for pulmonary administration from a powder inhalation device. In the case of self-propelling solutions and spray formulations, the effect is suspended by the selection of a valve having the desired spray characteristics (i.e. the characteristics that can produce a spray having the desired particle diameter) or controlled particle size suspension. Can be achieved either by incorporating the active ingredient as a powder. For administration by inhalation, the therapeutic can also be delivered in aerosol spray form from a pressurized container or dispenser that contains a suitable propellant, such as a gas (eg, carbon dioxide), or a nebulizer. Nasal drops can also be used.

  Systemic administration can also be transmucosal or transdermal. For transmucosal or transdermal administration, penetrants appropriate to the barrier to be permeated are used in the formulation. Such penetrants are generally known in the art, and such penetrants include, for example, for transmucosal administration, surfactants, bile salts, and phyllic acid derivatives. Transdermal administration can be accomplished through the use of spray nasal sprays or suppositories. For transdermal administration, the treatment is typically formulated in an ointment, salve, gel, or cream as generally known in the art.

  In one embodiment, the treatment is prepared with a carrier (eg, a controlled release formulation, including implants and microencapsulated delivery systems) that protects against rapid elimination from the body. Biodegradable, biocompatible polymers can be used, such as ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid. Those skilled in the art will understand how to prepare such formulations. Materials are also available from Alza Corporation and Nova Pharmaceuticals, Inc. Can be obtained commercially. Liposomal suspensions can also be used as pharmaceutically acceptable carriers. These can be prepared according to methods known to those skilled in the art, for example, as described in US Pat. No. 4,522,811. Microsomes and microparticles can also be used.

  Oral or parenteral compositions can be formulated in dosage unit form for ease of administration and uniformity of dosage. Dosage unit form refers to a physically discrete unit suitable for the subject to be treated as a single dosage; each unit will produce the desired therapeutic effect with the required pharmaceutical carrier. Contains a calculated predetermined amount of active compound. The specifications for the dosage unit forms of the present invention are such as the active compound for the treatment of an individual, due to the inherent properties of the active compound and the specific therapeutic effect to be achieved, as well as the limitations inherent in the field of formulation and these Directly depends on

  In general, a therapeutic identified in accordance with the present invention provides, for example, a therapeutically effective amount (eg, an appropriate concentration of drug for a time sufficient to induce the desired effect on the target tissue. Amount) can be formulated for parenteral or oral administration to humans or other mammals. Furthermore, the therapeutics of the invention can be administered alone or in combination with other molecules known to have a beneficial effect on the particular disease or condition of interest. By way of example only, useful cofactors include symptom relief cofactors (including antiseptics, antibacterials, antivirals and antifungals), and analgesics and anesthetics.

  The effective concentration of a therapeutic identified in accordance with the present invention to be delivered in a therapeutic composition will vary depending on a number of factors, including the final desired dosage of the drug being administered and the route of administration. To do. The preferred dosages administered are also the type and extent of the disease or condition being treated, the overall health status of the particular patient, the relative biological efficacy of the delivered therapeutic, the therapeutic formulation, the formulation It will depend on such variables as the presence and type of excipients in the product and the route of administration. In some embodiments, the therapeutics of the invention are provided to individuals using representative dosage units deduced from mammalian studies using the most recently described non-human primates and rodents. obtain. As described above, a dosage unit is a physiological (including a single (ie, a single dose that can be administered to a patient) and a mixture of the treatment itself or a solid or liquid pharmaceutical diluent or carrier. And a single dose that can be easily handled and filled, remaining as a biologically stable unit dose.

  In certain embodiments, the organism is engineered to produce a therapeutic identified in accordance with the present invention. These organisms can release the therapeutic for collection or can be introduced directly into the patient. In another series of embodiments, the cells can be utilized to act as carriers for therapeutics identified according to the present invention.

  The therapeutics of the present invention also include “prodrug” derivatives. The term prodrug is a pharmacologically inactive (or partial) of a parent molecule that requires biotransformation either spontaneously or enzymatically in an organism that releases or activates the active ingredient. Inactive) derivatives. Prodrugs are variants or derivatives of the therapeutics of the invention that have groups cleavable under metabolic conditions. Prodrugs are therapeutics of the invention that are pharmaceutically active in vivo when they undergo solvodifferentiation under physiological conditions or undergo enzymatic degradation. The prodrugs of the invention can be referred to as single, double, triple, etc., depending on the number of biotransformations required to release or activate the active drug component in the organism, and This indicates the number of functional groups present in the precursor form. Prodrug forms often provide the advantages of solubility, histocompatibility, or delayed release in mammalian organisms (Bundgard, Design of Prodrugs, pages 7-9, 21-24, Elsevier, Amsterdam 1985 and (See Silverman, The Organic Chemistry of Drug Design and Drug Action, pages 352-401, Academic Press, San Diego, Calif., 1992). Furthermore, prodrug derivatives according to the present invention may be combined with other features that increase bioavailability.

( Reference Example 1. GILT construct)
The IGF-II cassette was synthesized by ligation of a series of overlapping oligos and cloned into Pir1-SAT (a standard Leishmania expression vector). Four IGF-II cassettes were created: one encoding a wild type mature polypeptide, one having a Δ1-7 deletion, one having a Y27L mutation, and one both It has a mutation. These mutations have been reported to reduce IGF-II binding to other receptors but not affect binding to the M6P receptor.

  The coding sequence for human IGF-II is shown in FIG. The protein is synthesized as a pre-pro-protein with a single peptide of 24 amino acids at the amino terminus and a carboxy terminal region of 89 amino acids, both of which are described in O'Dell et al. (1998) Int. J. et al. Biochem Cell Biol. 30 (7): 767-71, removed after translation. The mature protein is 67 amino acids. A Leishmania codon optimized version of mature IGF-II is shown in FIG. 3 (Langford et al. (1992) Exp. Parasitol 74 (3): 360-1). This cassette was constructed by annealing overlapping oligonucleotides. The sequence of this oligonucleotide is shown in Table 3. Additional cassettes containing a deletion of amino acids 1-7 of the mature polypeptide (Δ1-7), a tyrosine to leucine residue 27 mutation (Y27L), or both mutations (Δ1-7, Y27L) are described below. It was made to produce an IGF-II cassette with specificity only for the desired receptor, as described. Oligo GILT1-9 was annealed and ligated to generate a wild type IGF-II cassette. To make the Y27L cassette, oligos 1, 12, 3, 4, 5, 16, 7, 8 and 9 were annealed and ligated. After ligation, the two cassettes were column purified. The wild type cassette and the Y27L cassette were amplified by PCR using oligos GILT 20 and 10 and the appropriate template. To incorporate the Δ1-7 deletion, two templates were amplified using oligos GILT11 and 10. The resulting four IGF-II cassettes (wild type, Y27L, Δ1-7 and Y27LΔ1-7) were column purified, digested with XbaI, gel purified, and ligated to Pir1-SAT cut with XbaI.

  The gene cassette was then cloned between the XmaI site (not shown) upstream of XbaI and the AscI site in the vector in such a way as to preserve the reading frame. The overlapping DAM methylase site of the 3'XbaI site allowed the use of a 5'XbaI site instead of the XmaI site for cloning. This AscI site adds a bridge of three amino acid residues.

示した変異をＩＧＦ−ＩＩカセットに組み込む目的は、融合タンパク質を適切なレセプターに標的化することを保証するためである。ヒトＩＧＦ−ＩＩは、ＩＧＦ−Ｉ（例えば、図７を参照のこと）およびインスリンのＢ鎖およびＡ鎖に対して高い程度の配列類似性および構造類似性を有する（Ｔｅｒａｓａｗａら（１９９４）Ｅｍｂｏ Ｊ．１３（２３）：５５９０−７）。従って、これらのホルモンが、重複するレセプター結合特異性を有することは驚くべきことではない。ＩＧＦ−ＩＩは、インスリンレセプター、ＩＧＦ−Ｉレセプターおよびカチオン依存マンノース６−ホスフェート／ＩＧＦ−ＩＩレセプター（ＣＩＭ６Ｐ／ＩＧＦ−ＩＩ）に結合する。ＣＩＭ６Ｐ／ＩＧＦ−ＩＩレセプターは、ＩＧＦ−ＩＩに対するレセプターとして、およびリソソームの加水分解酵素の分類に関与するマンノース６−ホスフェートレセプターとして作用する、二重活性レセプターである。長年にわたり、これらの２つの活性は、両方の活性が単一のタンパク質中に存在することが決定されるまで、別々のタンパク質に起因すると考えられてきた（Ｍｏｒｇａｎら（１９８７）Ｎａｔｕｒｅ ３２９（６１３７）：３０１−７）；（Ｔｏｎｇら（１９８８）Ｊ．Ｂｉｏｌ．Ｃｈｅｍ．２６３（６）：２５８５−８）。

The purpose of incorporating the indicated mutations into the IGF-II cassette is to ensure that the fusion protein is targeted to the appropriate receptor. Human IGF-II has a high degree of sequence similarity and structural similarity to IGF-I (see, eg, FIG. 7) and insulin B and A chains (Terazawa et al. (1994) Embo J). .13 (23): 5590-7). It is therefore not surprising that these hormones have overlapping receptor binding specificities. IGF-II binds to insulin receptor, IGF-I receptor and cation-dependent mannose 6-phosphate / IGF-II receptor (CIM6P / IGF-II). The CIM6P / IGF-II receptor is a dual active receptor that acts as a receptor for IGF-II and as a mannose 6-phosphate receptor involved in the classification of lysosomal hydrolases. Over the years, these two activities have been attributed to separate proteins until it is determined that both activities are present in a single protein (Morgan et al. (1987) Nature 329 (6137). : 301-7); (Tong et al. (1988) J. Biol. Chem. 263 (6): 2585-8).

  Most significant biological effects of IGF-II (eg, its mitogenic effect) are mediated through the IGF-I receptor rather than the CIM6P / IGF-II receptor (Ludwig et al. (1995) Trends in Cell Biology). 5: 202-206) (see also Korner et al. (1995) J. Biol. Chem. 270 (1): 287-95). The first result of IGF-II binding to the CIM6P / IGF-II receptor is thought to be transport to the lysosome for subsequent degradation. This represents an important means of controlling IGF-II levels and why mice with null mutations in the CIM6P / IGF-II receptor show prenatal death unless IGF-II is also deleted (Lau et al. (1994) Genes Dev. 8 (24): 2953-63); (Wang et al. (1994) Nature 372 (6505); 464-7); (Ludwig et al. (1996) Dev. Biol. 177. (2): 517-35). In the method of the present invention, it is desirable to bind the IGF-II fusion protein to the CIM6P / IGF-II receptor. Y27L and Δ1-7 mutations reduce IGF-II binding to IGF-I and insulin receptors without altering affinity for CIM6P / IGF-II receptor (Sakano et al. (1991) J. Biol. Chem). 266 (31): 20626-35); (Hashimoto et al. (1995) J. Biol. Chem. 270 (30): 18013-8). Thus, according to the present invention, these mutant forms of IGF-II provide a means of targeting fusion proteins specifically to the CIM6P / IGF-II receptor.

  In one experiment, four different IGF-II cassettes (wild type, Δ1-7, Y27L and Δ1-7 / Y27L) with the appropriate sequence were made. The β-GUS cassette was fused to the IGF-II cassette and these constructs were placed in parasites. The α-galactosidase cassette is also fused to the IGF-II cassette. The GUS fusion was tested and shown to produce an enzymatically active protein.

  One preferred construct shown in FIG. It contains the signal peptide of mexicana secreted acid phosphatase (SAP-1) and is cloned into the XbaI site of the modified Pir1-SAT from which a single SalI site has been removed. What is fused in-frame is the mature β-GUS sequence, which is connected to the IGF-II tag by a three amino acid bridge.

( Reference Example 2. Preparation of GILT protein)
L. expresses and secretes β-GUS. mexicana was added to 100 ml of Standard Promastigote medium (40 mM HEPES, pH 7.5, 0.1 mM adenine, 0.0005% hemin, 0.0001% biotin, 5% fetal calf serum, 5% embryo fluid, 50 units / ml penicillin, The cells were grown at 26 ° C. in M199) containing 50 μg / ml streptomycin and 50 μg / ml narseothricin. After reaching a density of about 5 × 10 ⁶ flagellar cells / ml, these flagellar cells were collected by centrifugation at 1000 × g for 10 minutes at room temperature; using these flagellar cells, At room temperature, 1 liter of low protein medium (M199 supplemented with 0.1 mM adenine, 0.0001% biotin, 50 units / ml penicillin and 50 μg / ml streptomycin) was inoculated. The 1 liter culture was included in a 2 liter capped flask equipped with a sterile stir bar and the culture could be incubated at 26 ° C. with gentle agitation. The 1 liter cultures were aerated twice a day by moving them into a laminar flow hood, removing the cap, and swirling vigorously before replacing the cap. When the culture reached a density of 2-3 × 10 ⁷ flagellar cells / ml, the culture was centrifuged as described above. However, the flagellar cell pellet was discarded and the medium was decanted into a sterile flask. Addition of 434 g (NH ₄ ) ₂ SO ₄ per liter caused active GUS protein to precipitate from the medium; the salt folding medium was stored at 4 ° C. overnight. The precipitated protein was recovered either by centrifugation at 10,500 × g for 30 minutes or by filtration through a Gelman Super-800 membrane; the protein was resuspended in 10 mM Tris pH 8, 1 mM CaCl _2. And stored at −80 ° C. until dialysis. Dissolve the crude precipitate from several liters of media, pool, place in dialysis tubing (Spectra / Por-7, MWCO 25,000), and DMEM (Dulbecco's) containing two 1 liter volumes of bicarbonate. Dialyzed overnight against modified Eagle medium).

( Reference Example 3. GILT uptake assay)
The dermal fibroblast cell line GM4668 (human, NIGMS Human Genetic Mutant Cell Repository) is derived from a patient with type VII mucopolysaccharide deposition disease; therefore, the cells have little or no β-GUS activity. Absent. Thus, GM4668 cells are particularly useful for testing uptake of GUS-GILT constructs into human cells. GM4668 cells were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 15% (v / v) fetal calf serum in 12-well tissue culture plates at 37 ° C. in 5% CO ₂ . Fibroblasts were prepared as described in Reference Example 2, human β-glucuronidase (GUS) expressed in Leishmania, GUS-IGF-II fusion protein (GUS-GILT), or mutant GUS-IGF-II. Incubated overnight in the presence of about 150 units of preparation of the fusion protein (GUSΔ-GILT). Control wells did not contain added enzyme (DMEM medium blank). After incubation, the medium was removed from the wells and assayed for GUS activity in triplicate. Wells were washed 5 times with 1 ml of 37 ° C. phosphate buffered saline, then 0.2 ml lysis buffer (10 mM Tris pH 7.5, 100 mM NaCl, 5 mM EDTA, 2 mM 4- (2-amino) In ethyl) -benzenesulfonyl fluoride hydrochloride (AEBSF, Sigma), and 1% NP-40) at room temperature for 15 minutes. The cell lysate was transferred to a microcentrifuge tube and then centrifuged at 13,000 rpm for 5 minutes to remove cell debris. Three 10 μL aliquots of lysate were assayed for protein concentration (Pierce Micro BCA protein assay, Pierce, IL).

Three 38 μL aliquots of the lysate were adapted from a standardized fluorescent assay using a standardized fluorescent assay from Wolfe et al. (1996) Protocols for Gene Transfer in Neuroscience: Howards Gene Therapy of Neurological Disorders 263-274. Assayed. The assay is performed in a disposable fluorometric cuvette. Add 150 μl of reaction mixture to each cuvette. The 1 ml reaction mixture is 860 μl H ₂ O, 100 μl 1M sodium acetate, 40 μl 25 × β-GUS substrate mixture. The 25 × β-GUS substrate mixture is a suspension of 250 mg 4-methylumbelliferyl-β-D glucuronide in 4.55 ml ethanol stored at −20 ° C. in a desiccator. 38 μl of sample is added to the reaction mixture and the reaction is incubated at 37 ° C. The reaction is terminated by addition of 2 ml stop solution (10.6 g Na ₂ CO ₃ , 12.01 g glycine, 500 ml with H ₂ O, pH 10.5). Next, the fluorescence output is measured with a fluorometer.

  The results of the uptake experiment show that the amount of cell-associated GUS-GILT is 10 times greater than that of unmodified GUS (FIG. 5). Double mutant constructs are about 5 times more effective than unmodified GUS. These results indicate that GILT technology is an effective means of targeting lysosomal enzymes for uptake. Uptake can also be confirmed using standard immunofluorescence techniques.

( Reference example 4. Competition experiment)
To confirm that GILT-mediated uptake occurs via an IGF-II binding site on the cation-independent M6P receptor, competition experiments were performed using recombinant IGF-II. The experimental design was the same as described above, except that GM4668 fibroblasts were incubated with the designated proteins for about 18 hours in serum-free DMEM + 2% BSA. Each β-GUS derivative was added at 150 U / well. For competition, 2.85 μg of IGF-II was added to each well. This corresponds to an approximately 100-fold molar excess over GILT-GUS, a concentration sufficient to compete for binding to the M6P / IGF-II receptor.

  The result of the competition experiment is shown in FIG. Over 24 units of GILT-GUS / mg lysate was detected in the absence of IGF-II. Upon addition of IGF-II, the amount of cell associated GILT-GUS dropped to 5.4U. This level is similar to the level of unmodified GUS taken up by fibroblasts. Thus, the bulk of GILT protein uptake can be competed by IGF-II, indicating that this uptake actually occurs through specific receptor-ligand interactions.

( Reference Example 5. Expression of gene product in serum-free medium)
The expression product can also be isolated from serum free media. In general, expression strains are grown in serum-containing medium, diluted in serum-free medium, and grown for several generations (preferably 2-5 generations) before isolating the expression product. For example, production of secreted targeted therapeutic protein can be isolated from Leishmania mexicana flagellar cells that are initially cultured at 27 ° C. in 50 ml of 1 × M199 medium in a 75 cm ² flask. When the cell density reaches 1-3 × 10 ⁷ / ml, this culture is used to inoculate 1.2 L of M199 medium. When the density of this culture reaches approximately 5 × 10 ⁶ / ml, the cells are harvested by centrifugation, resuspended in 180 ml of supernatant and placed in 12 L of “Zima” medium in a 16 L spinner flask. Used for inoculation. The initial cell density of this culture is typically about 5 × 10 ⁵ / ml. The culture is grown to a cell density of about 1.0-1.7 × 10e ⁷ cells / ml. When this cell density is reached, the cells are separated from the culture medium by centrifugation, and the supernatant is filtered through a 0.2 μ filter at 4 ° C. to remove residual flagellar cells. The filtered medium was concentrated from 12.0 L to 500 ml using a tangential flow filtration device (MILLIPORE Prep / Scale-TFF cartridge).

Suitable growth media for this method are M199 growth media and “Zima” growth media. However, other serum-containing and serum-free media are also useful. The M199 growth medium is as follows: (1 L batch) = 200 ml 5 × M199 (with phenol red pH indicator) +636 ml H ₂ O, 50.0 ml fetal calf serum, 50.0 ml EF bovine embryo fluid, 1.0 ml Of 50 mg / ml norseothricin, 0.25% hemin in 2.0 ml of 50% triethanolamine, 10 mM adenine in 10 ml of 50 mM Hepes pH 7.5, 40.0 ml of 1 M Hepes pH 7.5, 95 ml of 1 ml 0.1% biotin in ethanol, 10.0 ml penicillin / streptomycin. All sera used are inactivated by heat. Final volume = 1 L and sterile filtered. The “Zima” modified M199 medium is as follows: (20.0 L batch) = 219.2 g M199 powder (−) phenol red + 7.0 g sodium bicarbonate, 10 mM adenine in 200.0 ml 50 mM Hepes pH 7.5, 800.0 ml Hepes free acid pH 7.5, 20.0 ml 0.1% biotin in 95% ethanol, 200.0 ml penicillin / streptomycin, final volume = 20.0 L, and sterile filtered.

The targeted therapeutic protein is preferably purified by Concanavalin A (Con A) chromatography. For example, if the culture reaches a density exceeding 1.0 × 10 ⁷ flagellar cells / ml, The mexicana is removed by centrifugation at 500 xg for 10 minutes. Pass the collected culture medium through a 0.2 μm filter and remove the particles before loading directly onto a ConA-agarose column (4% cross-linked beaded agarose, Sigma). The ConA-agarose column was pretreated with 1M NaCl, 20 mM Tris pH 7.4, 5 mM CaCl ₂ , MgCl ₂ and MnCl ₂ each, and then 5 volumes of column buffer (20 mM Tris pH 7.4, 1 mM CaCl 2 ₂ and 1 mM MnCl ₂ ). A total of 179,800 units (nmol / hour) of GUS activity (in 2 L) in culture medium is loaded onto a 22 ml ConA agarose column. Activity is not detectable during flow-through or washing. GUS activity is eluted with column buffer containing 200 mM methylmannopyranoside. The elution fractions containing the active peak are pooled and concentrated. Uptake and competition experiments were performed as described in Reference Examples 3 and 4. However, the organisms were grown in serum-free medium and purified with ConA; about 350-600 units of enzyme were applied to the fibroblasts. The results are shown in FIG.

( Reference Example 6. Competition experiment using denatured IGF-II as competitor)
The experiment in Reference Example 4 is repeated using either normal IGF-II or denatured IGF-II as competitor. As in Reference Example 4, the amount of cell-associated GUS-GILT decreases when co-incubated with a normal IGF-II concentration that is effective for competition, but at the competitive concentration, denatured IGF-II has little effect. Does not have or has no influence.

( Reference Example 7. Enzyme assay)
Assays for GUS activity are performed as described in Reference Example 3 and / or below.

Glass assay tubes are numbered in triplicate and 100 μL of 2 × GUS reaction mixture is added to each tube. A 2 × GUS reaction mixture is prepared by the addition of 100 mg 4-methylmethylumbelliferyl-β-D glucuronide to 14.2 mL 200 mM sodium acetate (pH adjusted to 4.8 with acetic acid). Transfer up to 100 μL of sample to each tube; add water until final reaction volume is 200 μL. The reaction tube is covered with parafilm and incubated in a 37 ° C. water bath for 1-2 hours. The reaction was carried out by dissolving 1.8 mL stop buffer (10.6 g Na ₂ CO ₃ and 12.01 g glycine in a final volume of 500 mL water, adjusting the pH to 10.5 and repeating the repeater (repeat Stop by the addition of (prepared by filter sterilization in the dispenser). The fluorometer is then calibrated using 2 mL of stop solution as a blank and fluorescence is measured from the remaining sample. A standard curve is generated using 1, 2, 5, 10, and 20 μL of 166 μM 4-methylmethylumbelliferone standard in a final volume of 2 mL stop buffer.

A 4-methylmethylumbelliferone standard solution is prepared by dissolving 2.5 mg 4-methylmethylumbelliferone in 1 ml ethanol and adding 99 ml sterile water to a concentration of about 200 nmol / ml. The exact concentration is determined spectrophotometrically. The extinction coefficient at 360 nm is 19.000 cm ⁻¹ M ⁻¹ . For example, 100 μL is added to 900 μL of stop buffer and the absorbance at 360 nm is measured. If the measured value is 0.337, the concentration of the standard solution is 0.337 × 10 (diluted) / 19,000 = 177 μM, which is then diluted to 166 μM by adding an appropriate amount of sterile water. obtain.

( Reference Example 8. Experiments on binding uptake and half-life)
GUS-GILT protein binding to the M6P / IGF-II receptor on fibroblasts was measured and the uptake rate was determined by published methods (York et al. (1999) J. Biol. Chem. 274 (2): 1164 -Evaluate in the same manner as in (71). GM4668 fibroblasts cultured in 12-well culture dishes as described above are washed in ice-cold serum-free medium containing 1% BSA. Ligand (either GUS, GUS-GILT or GUS-ΔGILT, or control protein) is added to cells in chilled serum-free medium supplemented with 1% BSA. Plates were incubated on ice for 30 minutes upon ligand addition. After 30 minutes, the ligand is removed and the cells are quickly washed 5 times with ice-cold medium. Place 1 ml of ice-cold strip buffer (0.2 M acetic acid, pH 3.5, 0.5 M NaCl) into the time zero well. The plate is then floated in a 37 ° water bath and 0.5 ml of pre-warmed medium is added to start uptake. Add 1 ml strip buffer at all stopping points. At the end of the experiment, an aliquot of strip buffer is stored for a fluorometric assay for β-glucuronidase activity as described in Reference Example 3. The cells are then lysed as described above and the lysate is assayed for β-glucuronidase activity. Alternatively, lysates can be tested for the presence of targeted therapeutic proteins using immunological methods.

  Once the temperature is shifted to 37 °, GUS-GILT is immediately taken up by fibroblasts within minutes (York et al., (1999) J. Biol. Chem. 274 (2): 1164-71), and The enzyme activity is expected to persist in the cell for a long time.

(Example 9. Protein production in mammalian cells)
(CHO cells)
GUS-GILTΔ1-7 and GUSΔC18-GILTΔ1-7 were expressed in CHO cells using the system of Ulmasov et al. (2000) PNAS 97 (26): 14212-14217. The appropriate gene cassette was inserted into the EcoRI site of the pCXN vector and electroporated into CHO cells at 50 μF and 1,200 V in a 0.4 cm cuvette. Colony selection and amplification was mediated by 400 μg / mL G418 for 2-3 weeks. CHO cells were grown in MEM medium supplemented with 15% FBS, 1.2 mM glutamine, 50 μg / mL proline, and 1 mM pyruvate. For enzyme production, the cells were plated in DMEM in a multifloor flask. Once the cells reached confluence, recovery medium (Weymouth medium supplemented with 2% FBS, 1.2 mM glutamine, and 1 mM pyruvate) was applied to the cells. Medium containing secreted recombinant enzyme was collected every 24-72 hours. A typical level of secretion of one GUS-GILTΔ1-7 cell line is 4000-5000 units / mL / 24 hours.

  A number of GUSΔC18-GILTΔ1-7 CHO strains were assayed for secreted enzyme production. The six highest producing strains secreted 8600-14900 units / mL / 24 hours. The highest producing strain was selected for protein collection.

(HEK293 cells)
The GUS-GILT cassette was cloned into pCEP4 (Invitrogen) for expression in HEK293 cells. The cassettes used included wild type GUS-GILT; GUS-GILTΔ1-7; GUSGILTY27L; GUSΔC18-GILTΔ1-7; GILTY27L, and GUS-GILTF19S / E12K.

HEK293 cells were cultured to 50-80% confluency in 12-well plates containing DMEM medium containing 4 mM glutamine and 10% FBS. Cells were transfected with the pCEP-GUS-GILT DNA plasmid using FuGENE 6 (Roche) as described by the manufacturer. 0.5 μg DNA and 2 μL FuGENE 6 were added per well. Two to three days after transfection, cells were removed from the wells using trypsin and then cultured in a T25 cm ² culture flask containing the above DMEM medium containing 100 μg / mL hygromycin to stabilize the transfected cells. A population was selected. The medium containing hygromycin was changed every 2-3 days. Within 1-2 weeks, the cultures were spread into T75 cm ² culture flasks. For enzyme production, cells were plated in MEM in multifloor flasks. Once the cells reached confluence, the cells were given recovery medium (Weymouth medium supplemented with 2% FBS, 1.2 mM glutamine, and 1 mM pyruvic acid). This medium is optimized for CHO cells but not for 293 cells; therefore, the secretion level of HEK293 strain can be found to be significantly higher in another medium.

  The levels of secreted enzymes are shown in Table 4.

（実施例１０．ＧＵＳ−ＧＩＬＴ融合タンパク質の精製）
クロマトグラフィー（従来的クロマトグラフィーおよびアフィニティクロマトグラフィーを含む）を使用して、ＧＵＳ−ＧＩＬＴ融合タンパク質を精製し得る。

Example 10. Purification of GUS-GILT fusion protein
Chromatography (including conventional chromatography and affinity chromatography) can be used to purify the GUS-GILT fusion protein.

(Conventional chromatography)
One procedure for purifying the GUS-GILT fusion protein produced in Leishmania is described in Reference Example 2. An alternative procedure is described in the following paragraph.

Culture supernatants from Leishmania mexicana cell lines expressing GUS-GILT fusions were collected, centrifuged, and filtered through a 0.2μ filter to remove cell debris. The supernatant was concentrated using a tandem ultrafilter with a 100,000 molecular weight cut-off and stored at -80 ° C. The concentrated supernatant was loaded directly onto a column containing Concanavalin A (ConA) immobilized on beaded agarose. The column was washed with ConA column buffer (50 mM Tris (pH 7.4), 1 mM CaCl ₂ , 1 mM MnCl ₂ ), after which the mannosylated protein containing the GUS-GILT fusion was washed with 0-0.2 M methyl-α. -D-pyranoside gradient was used to elute in ConA column buffer. Fractions with glucuronidase activity (assayed as described in Example 7) were pooled, concentrated, and buffered for SP column buffer (25 mM sodium phosphate (pH 6), 20 mM) for the next column. (NaCI, 1 mM EDTA). The concentrated fraction was loaded onto an SP fast flow column equilibrated with the same buffer, and the column was washed with additional SP column buffer. The GUS-GILT fusion was eluted from the column in two steps: 1) a gradient of 0-0.15 M glucuronic acid in 25 mM sodium phosphate (pH 6) and 10% glycerol, then 0.2 M glucuronic acid, 25 mM phosphoric acid Sodium (pH 6), 10% glycerol. Fractions containing glucuronidase activity were pooled and the buffer was replaced with 20 mM potassium phosphate (pH 7.4). These pooled fractions were loaded onto a HA-ultrogel column equilibrated with the same buffer. The GUS-GILT fusion protein was eluted with an increasing gradient of phosphate buffer (145-340 mM potassium phosphate, pH 7.4). Fractions containing glucuronidase activity were pooled, concentrated and stored at −80 ° C. in 20 mM Tris (pH 8) containing 25% glycerol.

  Conventional chromatographic methods for purifying GUS-GILT fusion proteins produced in mammalian cells are described in the following paragraphs.

  Mammalian cells overexpressing the GUS-GILT fusion protein are grown to confluency in Nunc Triple flasks and then fed with serum-free medium (Waymouth MB 752/1) supplemented with 2% fetal calf serum, Collect enzyme for purification. The medium is collected and the flask is refeeded at 24 hour intervals. Pool the media from several flasks and centrifuge at 5000 × g for 20 minutes at 4 ° C. to remove detached cells and the like. The supernatant is removed and an aliquot is taken for β-GUS assay. Here, this medium can be used for direct growth or stored at −20 ° C. for later use.

  1 L of secretion medium is thawed at 37 ° C. (if frozen), filtered through a 0.2 μ filter and transferred to a 4 L beaker. The volume of the medium is diluted 4-fold by adding 3 L of dd water to reduce the salt concentration: The pH of the diluted medium is adjusted to 9.0 using 1 M Tris base (Tris base). Add 50 mL of DEAE-Sephacel equilibrated with 10 mM Tris (pH 9.0) to the diluted medium and gently stir with a large stir bar at 4 ° C. for 2 hours. (A small aliquot is removed, microcentrifuged, and the supernatant is assayed to monitor binding.) Once binding is complete, the resin is collected on a fritted glass funnel and multiple times, 750 mL of 10 mM Tris (pH 9. Wash with 0). The resin is transferred to a 2.5 cm column and washed with an additional 750 mL of the same buffer at a flow rate of 120 mL / hour. The DEAE column is eluted with a linear gradient of 0-0.4 M NaCl in 10 mM Tris (pH 9.0). Fractions containing the GUS-GILT fusion protein were detected by 4-methylmethylumbelliferyl-β-D glucuronide assay, pooled and with 25 mM Tris (pH 8), 1 mM β-glycerin phosphate, 0.15 M sodium chloride. Load onto an equilibrated Sephacryl S-200 600 mL column and elute with the same buffer.

  Fractions containing the GUS-GILT fusion protein are pooled and dialyzed against 3 × 4 L of 25 mM sodium acetate (pH 5.5), 1 mM β-glycerol phosphate, 0.025% sodium azide. The dialyzed enzyme is loaded onto a 15 mL column of CM-Sepharose equilibrated with 25 mM sodium acetate (pH 5.5), 1 mM β-glycerol phosphate, 0.025% sodium azide at a flow rate of 36 mL / hour. This is then washed with 10 column volumes of the same buffer. The CM column is eluted in equilibration buffer with a linear gradient of 0-0.3 M sodium chloride. Fractions containing GUS-GILT fusion protein were pooled and 2.4 × 70 cm of Sephacryl S-300 equilibrated with 10 mM Tris (pH 7.5), 1 mM β-glycerol phosphate, 0.15 M NaCl (total volume = 317 mL) column loaded at a flow rate of 48 mL / hour. Fractions containing the fusion protein are pooled; this pool is assayed for GUS activity and protein concentration to determine specific activity. Aliquots are run on SDS-PAGE followed by Coomassie staining or silver staining to confirm purity. If higher concentrations of enzyme are required, use Amicon Ultrafiltration Unit (50,000 molecular weight cut-off) or Centricon C-30 unit (30,000 molecular weight cut-off) using XM-50 membrane to concentrate the fusion protein. obtain. The fusion protein is stored at −80 ° C. in 10 mM Tris (pH 7.5), 1 mM sodium β-glycerin phosphate, 0.15 M NaCl buffer.

(Affinity chromatography)
Affinity chromatography conditions are essentially as described by Islam et al. Biol. Chem. 268 (30): 22627-22633. Conditioned media from mammalian cells overexpressing the GUS-GILT fusion protein (collected and centrifuged as in conventional chromatography above) is filtered through a 0.22μ filter. Sodium chloride (crystals) is added to a final concentration of 0.5M and sodium azide is added (by addition of 1/400 volume of 10% stock solution) to a final concentration of 0.025%. To a 5 mL column of anti-human β-glucuronidase-Affigel 10 (equilibrated with Antibody Sepharose Wash Buffer (10 mM Tris (pH 7.5), 10 mM potassium phosphate, 0.5 M NaCl, 0.025% sodium azide)) Apply medium at 4 ° C. at a rate of 25 mL / hour. Fractions are collected and monitored for any GUS activity in the effluent. The column is washed with a 10-20 column volume of Antibody Sepharose Wash Buffer at 36 mL / hour. Fractions are collected and GUS activity is monitored. The column is eluted with 50 mL of 10 mM sodium phosphate (pH 5.0) +3.5 M MgCl ₂ at 36 mL / hour. 4 mL fractions are collected and assayed for GUS activity. Fractions containing the fusion protein are pooled, diluted with an equal volume of P6 buffer (25 mM Tris (pH 7.5), 1 mM β-glycerol phosphate, 0.15 mM NaCl, 0.025% sodium azide) and BioGel Demineralize on a P6 column (equilibrated with P6 buffer) to remove MgCl ₂ and change the buffer to P6 buffer for storage. The fusion protein is eluted with P6 buffer, fractions with GUS activity are pooled, and the pooled fractions are assayed for GUS activity and protein. Purity is confirmed using an SDS-PAGE gel stained with Coomassie blue or silver. The fusion protein is stored frozen at −80 ° C. in P6 buffer for long-term stability.

Example 11 Uptake Experiment of Protein Produced by Mammals
Culture supernatants from HEK293 or CHO cell lines producing GUS or GUS-GILT constructs were collected and cells were removed through a 0.2 μm filter. GM4688 fibroblasts were cultured at 37 ° C. in 5% CO _{2 in} DMEM supplemented with 15% (v / v) fetal calf serum in 12-well tissue culture plates. The cells were washed once at 37 ° C. with uptake medium (DMEM + 2% BSA (Sigma A-7030)). The fibroblasts were then cultured (1000-21 hours) with 1000-4000 units of enzyme per mL of uptake medium. In some experiments, uptake competitors were added. Mannose-6-phosphate (Calbiochem 444100) was added to several media at a concentration of 2-8 mM, and pure recombinant IGF-II (Cell Sciences OU100) was added to the same media at 2.86 mM (input enzyme). Depending on the amount of A), corresponding to a 10 to 100-fold molar excess). Uptake was typically measured in triplicate wells.

  Following incubation, the media was removed from the wells and assayed in duplicate for GUS activity. Wells were washed 5 times with 1 mL 37 ° C. phosphate buffered saline, then 0.2 mL lysis buffer (10 mM Tris, (pH 7.5), 100 mM NaCl, 5 mM EDTA, and 1% NP− 40) and incubated at room temperature for 15 minutes. The cell lysate was transferred to a microcentrifuge tube and centrifuged at 13,000 rpm for 5 minutes to remove cell debris. Two 10 μL aliquots of lysate were assayed for GUS activity using a standard fluorometric assay. Three 10 μL aliquots of lysate were assayed for protein concentration (Pierce Micro BCA protein assay, Pierce, IL).

  In the first experiment, the uptake of GUS-GILTΔ1-7 produced in CHO was compared to GUSΔC18-GILTΔ1-7 produced in CHO. As shown in Table 5, GUSΔC18-GILTΔ1-7 proteins engineered to eliminate potential protease cleavage sites have significantly higher levels of uptake, which can be inhibited by IGF-II and M6P . In contrast, uptake of recombinant GUS lacking an IGF-II tag produced in mammalian cells was not affected by the presence of excess IGF-II, but was completely attenuated by excess M6P. In this experiment, uptake was performed for 18 hours.

その後の実験により、ＭＰＳＶＩＩ患者由来のヒト線維芽細胞によるＣＨＯおよびＨＥＫ２９３によって産生された酵素の取り込みを評価した。この実験において、取り込みは、２１時間であった。

Subsequent experiments evaluated the uptake of enzymes produced by CHO and HEK293 by human fibroblasts from MPSVII patients. In this experiment, uptake was 21 hours.

さらなる実験により、ＩＧＦ−ＩＩ、８ｍＭ Ｍ６Ｐ、または両インヒビターの存在下での、選択した酵素の取り込みを評価した。取り込みを、２２．５時間の間測定した。

Further experiments evaluated the uptake of selected enzymes in the presence of IGF-II, 8 mM M6P, or both inhibitors. Uptake was measured for 22.5 hours.

上述の実験は、ＣＨＯ産生系およびＨＥＫ２９３産生系は、その機能的組換えタンパク質を分泌する能力において、本質的に同等であることを示す。この実験はまた、過剰なＩＧＦ−ＩＩの存在が、タグ化タンパク質の取り込みを７０〜９０＋％まで減殺するが、非タグ化タンパク質の取り込みにはあまり影響しない（４．５％）ことを示し、これは、哺乳動物によって産生されたタンパク質の、特異的なＩＧＦ−ＩＩ媒介性取り込みを示す。Ｌｅｉｓｈｍａｎｉａによって産生されたタンパク質と違い、哺乳動物細胞によって産生された酵素は、Ｍ６Ｐを含むことが予測される。これらのタンパク質上の２つのリガンドの存在は、Ｍ６Ｐ／ＩＧＦ−ＩＩレセプターを介した取り込みを指向し得、過剰なＩＧＦ−ＩＩも過剰なＭ６Ｐも、取り込みを完全に減殺することはないことを意味する。さらに、２つのリガンドは離れた位置のレセプターに結合するので、１つのリガンドを介したレセプターへの結合は、他方の競合物の過剰な存在によってあまり影響されないはずである。

The above experiments show that the CHO production system and the HEK293 production system are essentially equivalent in their ability to secrete functional recombinant proteins. This experiment also shows that the presence of excess IGF-II reduces the uptake of tagged protein to 70-90 +%, but does not significantly affect uptake of untagged protein (4.5%) This indicates a specific IGF-II mediated uptake of the protein produced by the mammal. Unlike proteins produced by Leishmania, enzymes produced by mammalian cells are expected to contain M6P. The presence of two ligands on these proteins can direct uptake through the M6P / IGF-II receptor, meaning that neither excess IGF-II nor excess M6P completely attenuates uptake To do. Furthermore, since the two ligands bind to distant receptors, binding to the receptor via one ligand should not be significantly affected by the excessive presence of the other competitor.

Example 12. In vivo treatment
First, GUS minus mice can be used to evaluate the effectiveness of GUS-GILT and its derivatives in enzyme replacement therapy. GUS minus mice were obtained from Birkenmeier et al. (1989) J. MoI. Clin. Invest 83 (4): 1258-6, as described by B6. Produced by heterozygous mating of C-H-2 ^bml / ByBIR-gus ^mps / + mice. Preferably, the mouse is resistant to human β-GUS. This mouse carries a transgene with a defective copy of human β-GUS and induces immune tolerance of human proteins (Sly et al. (2001) PNAS 98: 2205-2210). Alternatively, human β-GUS (eg, GUS-GILT protein) can be administered to neonatal mice to induce immune tolerance. However, since the vascular brain barrier is not formed in mice by about 15 days, it is better to determine whether GILT-GUS crosses the cerebral blood vessels when starting injection in mice after 15 days of age. Therefore, transgenic mice are preferred.

  The first experiment is to measure the tissue distribution of the targeted therapeutic protein. At least three mice receive CHO produced GILT-tagged β-GUS protein (referred to herein as GUSΔC18-GILTΔ1-7). In this GUSΔC18-GILTΔ1-7, GUSΔ18, a β-GUS protein from which the last 18 amino acids of the protein have been removed, is fused to the N-terminus of Δ1-7 GILT, an IGF-II protein that lacks the first 7 amino acids of the mature protein. Is done. Other mice receive GUSΔC18-GILTΔ1-7 protein treated with β-GUS, buffer control, or periodate and sodium borohydride as described in Example 14. In general, the preferred dose is in the range of 0.5-7 mg / kg body weight. As one example, the enzyme dose is 1 mg / kg body weight, administered intravenously, and the enzyme concentration is about 1-3 mg / mL. In addition, at least 3 mice receive a 5 mg / kg body weight dose of GUSΔC18-GILTΔ1-7 protein treated with periodate and sodium borohydride. After 24 hours, the mice are sacrificed and the following organs and tissues are isolated: liver, spleen, kidney, brain, lung, muscle, heart, bone, and blood. A portion of each tissue is homogenized and β-GUS enzyme activity per mg of protein is measured as described in Sly et al. (2001) PNAS 98: 2205-2210. A portion of tissue is prepared for histochemistry and / or histopathology, and histochemistry and / or histopathology is performed by published methods (eg, Vogler et al. (1990) Am J. Pathol. 136: 207-217).

  Further experiments include multiple injection protocols, which are injected into mice weekly at a dose of 1 mg / kg body weight. In addition, as described in Example 14, the measurement of the half-life of periodate modifying enzyme is measured relative to untreated enzyme.

  Two other assay formats can be used. In one format, 3-4 animals are given a single injection of 20,000 U of enzyme in 100 μl of enzyme dilution buffer (150 mM NaCl, 10 mM Tris, pH 7.5). Mice are sacrificed 72-96 hours later to verify the efficacy of this therapeutic agent. In the second format, mice are given a weekly injection of 20,000 units for 3-4 weeks and sacrificed one week after the final injection. Histochemical and histopathological analysis of the liver, spleen and brain was performed using published methods (Birkenmeier et al. (1991) Blood 78 (11): 3081-92; Sands et al. (1994) J. Clin. Invest 93 ( 6): 2324-31; Daly et al. (1999) Proc. Natl. Acad. Sci. USA 96 (5): 2296-300). In the absence of therapeutic agents, GUS-deficient mouse cells (eg, macrophages and Kupffer cells) develop a large intracellular storage compartment due to the accumulation of waste in lysosomes. In mice cells treated with the GUS-GILT construct, it is expected that the size of these compartments will be clearly reduced, or that these compartments will contract before they can no longer be seen by light microscopy.

  Similarly, a human suffering from lysosomal storage disease is treated using a construct that targets the appropriate therapeutic moiety to that human lysosome. In some examples, the treatment takes the form of regular (eg weekly) infusion of GILT protein. In other examples, treatment within a patient by administration of a nucleic acid capable of sustaining in vivo expression of the GILT protein, or by administration of a cell that expresses the GILT protein (eg, a human cell or a single cell organ). To achieve. For example, U.S. Patent No. 6,020,144 (issued February 1, 2000); U.S. Provisional Application No. 60 / 250,446; and U.S. Provisional Application Office Processing Number SYM-005PRA, "Protozoan Expression Systems". The Leishmania vector described in “For Lysosomal Storage Disease Genes” (filed May 11, 2001) can be used to express GILT protein in situ.

  Alternatively, the targeted therapeutic proteins of the invention can be administered and using methods previously described for use with GUS (enzyme assays, histochemical assays, neurological assays, survival assays, reproduction assays, etc.). The effect can be monitored. See, for example, Vogler et al. (1993) Pediatric Res. 34 (6): 837-840; Sands et al. (1994) J. MoI. Clin. Invest. 93: 2324-2331; Sands et al. (1997) J. MoI. Clin. Invest. 99: 1596-1605; O'Connor et al. (1998) J. MoI. Clin. Invest. 101: 1394-1400; and Super et al. (1999) 45 (2): 180-186.

Example 13. GILT-modifying enzyme replacement therapy for Fabry disease
The purpose of these experiments is to evaluate the effectiveness of GILT-modified α-galactosidase A (α-GAL A) as an enzyme replacement therapy for Fabry disease.

  Fabry disease is a lysosomal storage disease resulting from insufficient activity of α-GAL A, an enzyme responsible for removing terminal galactose from GL-3 and other neurosphingolipids. This decrease in enzyme activity occurs due to various missense and nonsense mutations in the x-linked genes. GL-3 accumulation is most prevalent in lysosomes of vascular endothelial cells of the heart, liver, kidney, skin and brain, but also occurs in other cells and tissues. The accumulation of GL-3 in vascular endothelial cells ultimately leads to heart disease and renal failure.

  Enzyme replacement therapy is an effective treatment for Fabry disease, and its success depends on the ability of this therapeutic enzyme to be taken up by lysosomes in cells where GL-3 accumulates. The Genzyme product Fabrazyme is recombinant α-GAL A produced in DUKX B11 CHO cells that has been approved in Europe for the treatment of patients with Fabry disease due to this demonstrated efficacy.

  The ability of Fabrazyme to be taken up by cells and delivered to lysosomes is due to the presence of mannose 6-phosphate (M6P) on N-linked carbohydrate. Fabrazyme is delivered to the lysosome via binding to mannose-6-phosphate / IGF-II receptor and subsequent receptor-mediated endocytosis present on the cell surface of most cell types. Fabrazyme reportedly has three N-linked glycosylation sites at ASN residues 108, 161 and 184. The predominant carbohydrates at these three positions are, respectively, the fucosylated biantennary bivalent complex, monophosphorylated mannose-7 oligomannose, and doubly phosphorylated mannose-7 oligomannose. is there.

  The glycosylation-independent lysosome targeting (GILT) technology of the present invention targets therapeutic proteins directly to lysosomes through different interactions with the IGF-II receptor. The targeting ligand is derived from mature human IGF-II, which also binds with high affinity to the IGF-II receptor. In other applications, other configurations involving cross-linking are feasible in current applications, but the IGF-II tag is provided as a c-terminal fusion with a therapeutic protein. The ability for cellular uptake of GILT-modified enzymes is established using GILT-modified β-glucuronidase, which fibroblasts in a process that is competitive with excess IGF-II. Is taken in efficiently. The benefits of GILT modification are increased binding to the M6P / IGF-II receptor, enhanced uptake of target cells into the lysosome, altered or improved pharmacokinetics, and expanded, altered or improved tissue distribution range. Improvement of the tissue distribution range may include delivery of the GILT modified α-GAL A across the blood brain barrier as the IGF protein actually crosses the blood brain barrier.

  Another advantage of the GILT system is the ability to produce uptake-compatible proteins in non-mammalian expression systems where M6P modification does not occur. In certain embodiments, GILT-modified proteins are produced primarily in CHO cells. In certain other embodiments, the GILT tag is placed at the c-terminus of α-GAL A, although the invention is not so limited.

Example 13B. GILT Modifying Enzyme Replacement Treatment for Pamp's Disease
The purpose of these experiments is to evaluate the efficacy of GILT-modified acid α-glucosidase (GAA) as an enzyme replacement therapeutic for Pomp disease.

  The glycosylation-dependent lysosome targeting (GILT) technology of the present invention enables M6P-independent targeting of GAA to patient lysosomes. In one embodiment, GAA or a catalytically active fragment thereof is fused at its N-terminus to a targeting moiety comprising the signal peptide of human IGF-II. A targeting moiety comprising a signal peptide improves secretion of GAA from the host cell, thereby achieving high yield production and facilitating recovery of therapeutic GILT-modified GAA. In one embodiment, a hybrid GAA gene cassette is created in which a mature IGF-II coding sequence containing a human IGF-II signal peptide and the desired modification is placed at the N-terminus of the GAA cassette at His29 ( It fuses at a site where signal peptide cleavage is observed in COS-1 cells) or at another suitable site (Wisselaar et al. (1993) J Biol Chem, 268 (3): 2223-31). Alternatively, the desired GILT tag can be placed at a site downstream of a known cleavage site to prevent removal of the GILT tag by potential proteolytic processing. An exemplary N-terminal GILT-tagged GAA cassette is illustrated in FIG.

  In another embodiment, a hybrid GAA gene cassette is created, in which a 3 ′ DNA segment encoding one of a series of mature human IGF-II peptide tags is added in-frame to the human GAA gene cassette. To do. The fusion can occur either after the C-terminal Cys at position 952 or at a site upstream of Cys952. Placing a GILT tag upstream from the C-terminus also removes a potential proteolytic site and two glycosylation sites that are not present in the mature GAA protein. Polysaccharide particles in clearance through the mannose receptor should increase the half-life of GILT-modified GAA because it reduces glycosylation in GAA by placing the GILT tag upstream from the C-terminus. The construct is designed to place a C-terminal GILT tag every 10 amino acid residues from the C-terminus to residue 816, and these tags label the upstream limit of the predicted mature 70 kDa enzyme C-terminus. . An exemplary C-terminal GILT-tagged GAA cassette is shown in FIG.

  In yet another embodiment, a hybrid GAA gene cassette is created, in which the desired GILT tag is fused to both the N-terminus and C-terminus of the GAA cassette.

  In order to identify which of the GILT-modified GAA constructs described above are the most enzymatically active and uptake-competent, this construct can be expressed in HEK293 cells, CHO cells, or other suitable intracellular The culture supernatant from the pool of clones is assayed for active enzyme and for incorporation into Pompe fibroblasts.

  To test enzyme activity, the fluorescent substrate 4-methylumbelliferyl-α-D-glucopyranoside (Reuser et al. (1978) Am J Hum Genet, 30 (2): 132-43) or the colorimetric substrate p -Culture supernatants are assayed for GAA activity using nitrophenyl-α-D-glucopyranoside. Total enzyme activity is normalized to cell number and the relative levels of expression achieved by various clonal cell lines are compared. In order to test the uptake competence, the culture medium is collected, filtered and subjected to Pompe fibroblasts. To assess the specificity of uptake, cells are incubated for 3-16 hours in the presence or absence of 5 mM M6P or IGF-II, or both. Pompe fibroblasts are collected and cells having GAA activity are measured.

  The GILT modified GAA with the most promising uptake properties and / or enzymatic activity is selected for mass production in CHO cells or other suitable expression systems.

  GAA-deficient mice (available from Jackson Laboratories) are used to assess the ability of GILT-modified GAA to reveal glycogen accumulation from tissues (Bijvoet et al. (1999) J. Pathol., 189 (3): 416-24; Raben et al. (1998) J. Biol. Chem., 273 (30): 19086-92). One exemplary protocol includes injection of 3-5 mice every 4 weeks at doses up to 100 mg / kg. Evaluation of injected mice includes biochemical analysis of glycogen and GAA activity in selected tissues, including various muscle tissues, heart, and liver, and histopathological analysis to visualize glycogen accumulation in tissues. Can be mentioned.

(GAA construct)
Human Image cDNA clone number 4374238 was obtained from Open Biosystems. This clone contains a full-length cDNA encoding human GAA. It has been isolated from the testis-derived NIH MGC 97 library and cloned into pBluescript (pBluescript-GAA). The sequence in GAA clone no. 4374238 is shown in FIG. Sequence analysis of GAA in clone no. 4374238 confirmed the presence of a full-length cDNA with four silent nucleotide changes compared to GenBank sequence NM000152. These three are GAA:

における通常の１ヌクレオチド多型としてこれまでに記載されている。太字の残基は、クローン４３７４２３８中に存在する。これらの３つの多型についての予測されるヘテロ接合性は、集団全体においてそれぞれ約０．２３、０．４、および０．４２である。４つ目の多型は、

Has been previously described as a common single nucleotide polymorphism. Bold residues are present in clone 4374238. The predicted heterozygosity for these three polymorphisms is about 0.23, 0.4, and 0.42, respectively, across the population. The fourth polymorph is

である。

It is.

  An exemplary C-terminal GILT-tagged GAA cassette is illustrated in FIG. The first set of GAA cassettes with C-terminal GILT tags at various positions is constructed by PCR using clone 4374238 as a template. The GAA open reading frame (ORF) is 2859 residues long. To minimize mutations caused by PCR, the 5 'half of the ORF is first amplified by PCR using 5' primer GAA13 (SEQ ID NO: 25) and 3 'primer GAA14 (SEQ ID NO: 26). These primers anneal to a specific SanDI restriction site at residue 1649. This PCR product is ligated between the EcoRI and XbaI sites of pBluescript to create plasmid p5'GAA. The 5 ′ primer GAA 15 (SEQ ID NO: 27) was then used to amplify the 3 ′ portion of the GAA ORF, as described in Table 8, as one of the set of SanDI sites and 3 ′ primers (30 Designed to identify a series of terminal offsets by nucleotides). Each GAA 16-GAA 26 primer contains an AscI site for GILT tag ligation and an XbaI site for cloning into pBluescript. The primers are shown in Table 8: The lower case part of the GAA 16-GAA 26 primer contains the designed AscI and XbaI sites, while the upper case part contains the GAA gene sequence. Primer GAA27, which contains a native stop codon and lacks an AscI site, is used to create a full-length, untagged clone. The amplified 3 'portion is ligated to p5'GAA at the SanDI and XbaI sites. This creates a series of GAA cassettes and inserts a GILT tag into them using AscI and XbaI sites. The complete GILT-tagged GAA cassette is then cloned into the expression vector pCEP4 for transfection into HEK293 cells.

注記：ＧＡＡ １６〜ＧＡＡ ２７およびＧＡＡ１４のプライマーは、下側鎖プライマーである。ＧＡＡ １６〜ＧＡＡ ２６プライマーの小文字の部分は、設計したＡｓｃＩ部位およびＸｂａＩ部位を含む。ＧＡＡ １３プライマーおよびＧＡＡ １５プライマーは、上側鎖のプライマーである。ＧＡＡ １３プライマーおよびＧＡＡ １５プライマーの小文字の部分は、設計ＥｃｏＲＩ部位を含む。各プライマーの大文字部分が、ＧＡＡ遺伝子配列を含む。

Note: GAA 16-GAA 27 and GAA 14 primers are lower side primer. The lower case part of the GAA 16-GAA 26 primer contains the designed AscI and XbaI sites. GAA 13 primer and GAA 15 primer are upper strand primers. The lower case portions of the GAA 13 primer and GAA 15 primer contain a designed EcoRI site. The uppercase part of each primer contains the GAA gene sequence.

  An exemplary N-terminal GILT-tagged GAA cassette is shown in FIG. A 5 'N-terminal tagged GILT cassette construct is obtained in a similar manner. For example, to construct a GAA cassette with a GILT tag fused to amino acid 29 of GAA (the signal peptide cleavage site of COS cells), a 5 ′ primer comprising an AscI site and an EcoRI site and a GAA coding sequence starting at residue His29. , As well as a 3 ′ primer that anneals at a specific StuI site at position 776 of the GAA protein (corresponding to the start position), amplifies the 5 ′ GAA cassette. This fragment is cloned into pBluescript containing full length GAA at the EcoRI and StuI sites (pBluescript-GAA), replacing the native amino terminus. Once this plasmid is made, the desired GILT tag and signal peptide are cloned between the EcoRI and AscI sites.

Example 14. Incompletely glycosylated therapeutic protein
The efficacy of a targeted therapeutic can be increased by extending the serum half-life of the targeted therapeutic. Hepatic mannose and asialoglycoprotein receptors extinguish glycoproteins from the circulation by recognizing specific carbohydrate structures (Lee et al. (2002) Science 295 (5561): 1898-1901; Ishibashi et al. (1994) J Biol.Chem.269 (45): 27803-6). In some embodiments, the present invention allows targeting of therapeutics to lysosomes and / or crossing the vascular brain barrier in a manner that is independent of carbohydrates but dependent on polypeptides or analogs thereof. By minimizing its removal from the circulation by the mannose and asialoglycoprotein receptors, the actual incomplete glycosylation of these proteins is expected to greatly extend their half-life in the circulation. Similarly, functional deglycosylation (eg, by modifying carbohydrate residues in therapeutic proteins by periodate / sodium borohydride treatment) interferes with carbohydrate recognition by one or more clearance pathways. This brings about the same effect. Nevertheless, since in most embodiments protein targeting is based on protein-receptor interactions rather than carbohydrate-receptor interactions, glycosylation modification or loss may be targeted to lysosomes and / or brain. Does not adversely affect the passage of blood vessels.

  Any lysosomal enzyme (eg, IGF-II) that uses peptide targeting signals can be chemically or enzymatically deglycosylated, or desired properties (in addition to specific lysosomal targeting, long-term Can be modified to produce a therapeutic product having a serum half-life of In the case of some lysosomal storage diseases, a fully glycosylated therapeutic may be an incompletely glycosylated targeted therapeutic if it may be important to deliver the therapeutic to macrophages or related cell types via the mannose receptor. Can be used in conjunction with to achieve targeting to a wide variety of cell types.

(Incomplete protein glycosylation during synthesis)
In some cases, it is preferred to first produce the targeted therapeutic protein in a system that does not produce fully glycosylated protein. For example, the targeted therapeutic protein may be E. coli. can be produced in E. coli, thereby producing a fully non-glycosylated protein. Alternatively, non-glycosylated protein is produced in mammalian cells treated with tunicamycin (an inhibitor of Dol-PP-GlcNAc formation). However, if a particular targeted therapeutic is not correctly folded in the absence of glycosylation, it is preferably produced first as a glycosylated protein and then deglycosylated or functionally Undergoes incomplete glycosylation.

  Alternatively, incompletely glycosylated targeted therapeutic proteins can be prepared by manipulating the gene encoding the targeted therapeutic protein and changing the amino acid that normally serves as an acceptor for glycosylation to a different amino acid. obtain. For example, an asparagine residue that serves as an acceptor for N-linked glycosylation can be replaced with a glutamine residue or another residue that is not a glycosylation acceptor. This conservative substitution probably has minimal impact on the enzyme structure, while removing glycosylation at this site. Alternatively, other amino acids near the glycosylation acceptor can be modified to destroy the recognition motif for the glycosylation enzyme without necessarily changing the amino acid that is normally glycosylated.

  In the case of GUS, removal of any one of the four potential glycosylation sites reduces the amount of glycosylation but leaves sufficient enzymatic activity (Shipley et al. (1993) J. Biol. Chem. 268 (16) : 12193-8). Removal of several sets of two glycosylation sites from GUS still allows significant enzyme activity. Removal of all four glycosylation sites by treating cells with tunicamycin abolishes enzyme activity, whereas deglycosylation of the purified enzyme results in enzymatically active material. Thus, the loss of activity associated with removal of the glycosylation site is believed to be due to enzyme misfolding.

  However, other enzymes fold correctly even in the absence of glycosylation. For example, bacterial glucuronidases can be naturally aglycosylated to target mammalian lysosomes using the targeting moiety of the present invention and / or cross the vascular brain barrier. Such enzymes can be synthesized under non-glycosylated conditions, for example, rather than being deglycosylated after being synthesized as a glycosylated protein.

(Deglycosylation)
If the targeted therapeutic is produced in a mammalian cell culture system, it is preferably secreted into the growth medium. This is recovered and then purified of the targeted therapeutic, eg, by a chromatographic purification protocol (eg, ion exchange, gel filtration, hydrophobic chromatography, ConA chromatography, affinity chromatography, or immunoaffinity chromatography). Enable.

  Chemical deglycosylation of glycoproteins can be achieved by a number of methods including trifluoromethanesulfonic acid (TFMS) treatment, or hydrogen fluoride (HF) treatment.

Chemical deglycosylation by TFMS (Sojar et al. (1989) J. Biol. Chem. 264 (5): 2552-9; Sojar et al. (1987) Methods Enzymol. 138: 341-50): 1 mg GILT-GUS in vacuo Allow to dry overnight. The dried protein is treated with 150 μl TFMS with occasional stirring under nitrogen at 0 ° C. for 0.5-2 hours. The reaction mixture is cooled to <−20 ° C. in a dry ice-ethanol bath and the reaction is neutralized by the slow addition of pre-cooled (−20 ° C.) 60% pyrimidine solution in water. The neutralization reaction mixture is then dialyzed at pH 7 at 4 ° C. with several changes of NH ₄ HCO ₃ . Chemical deglycosylation by TFMS can result in modifications that include methylation, succinimide formation, and isomerization of asparagine residues to the processed protein (Douglass et al. (2001) J. Protein Chem. 20 (7): 571-6).

  Chemical deglycosylation by HF (Sojar et al. (1987) Methods Enzymol. 138: 341-50): This reaction is described, for example, in Peninsula Laboratories, Inc. In a closed reaction system available from 10 mg GILT-GUS is dried under reduced pressure, placed in a reaction vessel, and then the reaction vessel is connected to an HF apparatus. After discharging the entire HF line, 10 mL of anhydrous HF is distilled from the reservoir while stirring the reaction vessel. The reaction is continued at 0 ° C. for 1-2 hours. Thereafter, HF is removed with a water suction machine for 15 to 30 minutes. Slight residual HF is removed under high vacuum. The reaction mixture is dissolved in 2 mL 0.2 M NaOH to neutralize any remaining HF and readjusted to pH 7.5 with cold 0.2 M HCl.

  Enzymatic deglycosylation (Thotkura et al. (1987) Methods Enzymol. 138: 350-9): N-linked carbohydrates are completely removed from glycoproteins using protein N-glycosidase (PNGase) A or protein N-glycosidase F. obtain. In one embodiment, the glycoprotein is denatured prior to treatment with glycosidase to facilitate the action of the enzyme on the glycoprotein; as discussed in “In Vitro Refolding” in the previous paragraph, Refolded. In another embodiment, excess glycosidase is used to process native glycoproteins to promote efficient deglycosylation. Some cell types (eg, CHO-derived Lecl cell lines) produce glycoproteins with low or simple glycosylation and facilitate subsequent enzymatic deglycosylation as described in Example 14D To do.

  In the case of targeted therapeutic proteins that are actually incompletely glycosylated, reduced glycosylation can reveal protease sensitive sites in the targeted therapeutic protein and reduce the half-life of the protein. N-linked glycosylation is known to protect a subset of lysosomal enzymes from proteolysis (Kundra et al. (1999) J. Biol. Chem. 274 (43): 31039-46). Such protease sensitive sites are preferably engineered outside the protein (eg, by site-directed mutagenesis). As discussed below, the risk of revealing either protease sensitive sites or potential epitopes can be minimized by incomplete deglycosylation or modification of carbohydrate structure rather than removal of the entire carbohydrate. .

(Carbohydrate structure modification or partial deglycosylation)
In some embodiments, the therapeutic protein is partially deglycosylated. For example, a therapeutic protein can be treated with an endoglycosidase (eg, endoglycosidase H that cleaves N-linked high mannose carbohydrates but not complex carbohydrates), leaving one GlcNAc residue linked to asparagine. A group can be left behind. Therapeutic proteins treated in this way lack high mannose carbohydrates and reduce interaction with hepatic mannose receptors. Even if this receptor recognizes terminal GlcNAc, the probability of a productive interaction with one GlcNAc on the protein surface is not as high as an intact high mannose structure. If the therapeutic protein is produced in mammalian cells, any complex carbohydrates present on the protein remain unaffected by endoglycosidase H (endoH) treatment and are sialylated at the end, thereby May reduce the interaction with hepatic carbohydrate recognition receptors. Thus, such proteins appear to have an extended half-life. At the same time, steric hindrance due to residual carbohydrate protects potential epitopes on the protein surface from the immune system and reduces protease access to the protein surface (eg, in a protease-enriched lysosomal environment).

  In other embodiments, the glycosylation of the therapeutic protein is modified by, for example, oxidation, reduction, dehydration, substitution, esterification, alkylation, sialylation, carbon-carbon bond cleavage, etc. Reduce clearance. In some preferred embodiments, the therapeutic protein is not sialylated. For example, treatment with periodate and sodium borohydride is effective to modify most carbohydrate structures of glycoproteins. Periodate treatment oxidizes vicinal diols, breaks carbon-carbon bonds, and replaces hydroxyl groups with aldehyde groups; borohydride reduces aldehydes to hydroxyls. Many sugar residues contain vicinal diols and are therefore cleaved by this process. As shown in FIG. 9A, proteins can be glycosylated with high mannose carbohydrates, including N-acetylglucosamine residues adjacent to asparagine and mannose residues at any position in the structure, at asparagine residues. As shown in FIG. 9B, the terminal mannose residue has three consecutive carbons with hydroxyl groups; both carbon-carbon bonds involved are cleaved by periodate treatment. Some non-terminal mannose residues also contain vicinal diol, which is similarly cleaved. Nonetheless, this treatment converts cyclic carbohydrates to linear carbohydrates but does not completely remove carbohydrates, potentially minimizing the risk of exposing protein sensitive or antigenic polypeptide sites.

  The half-life of the lysosomal enzyme β-glucuronidase is known to extend more than 10-fold after a series of treatments with periodate and sodium borohydride (Houba et al. (1996) Bioconjug. Chem. 7 (5): 606-11; Stahl et al. (1976) PNAS 73: 4045-4049; Achord et al. (1977) Pediat.Res. 11: 816-822; Achord et al. 78). Similarly, lysine has been treated with a mixture of periodate and sodium cyanoborohydride (Thorpe et al. (1985) Eur. J. Biochem. 147: 197). After injection into rats, the fraction of lysine absorbed into the liver is reduced by chemical treatment from 40% (untreated lysine) to 20% (modified lysine) of the injected dose. In contrast, the amount of lysine in the blood increased from 20% (untreated lysine) to 45% (modified lysine). Thus, the treated lysine was endowed with a broader tissue distribution and longer half-life in the circulation.

  When deglycosylated or modified by sequential treatment with periodate and sodium borohydride, β-glucoronidase constructs (or other glycoproteins) that bind to targeting moieties of the present invention are similar. Blessed with extended half-life (eg, greater than 2 fold, greater than 4 fold, or greater than 10 fold), but still maintains high affinity for the cation-independent M6P receptor and possesses this receptor Allows targeting of this construct to the lysosomes of This construct is also expected to cross the blood brain barrier efficiently. In contrast, β-glucuronidase preparations that rely on M6P for lysosomal targeting are endowed with a longer serum half-life when deglycosylated or treated with periodate and sodium borohydride, Unable to target lysosomes. This is because the M6P targeting signal has been modified by processing.

Carbohydrate modification by sequential treatment with periodate and sodium borohydride can be carried out as follows: Purified GILT-GUS with 40 mM NaIO ₄ in 50 mM sodium acetate (pH 4.5) for 2 hours at 4 ° C. Incubate. The reaction was stopped by the addition of excess ethylene glycol and unreacted reagents were removed by passing the reaction mixture through Sephadex G-25M equilibrated with PBS (pH 7.5). Following this treatment, it is incubated with 40 mM NaBH ₄ in PBS (pH 7.5) for 3 hours at 37 ° C. and then for 1 hour at 4 ° C. The reaction mixture is passed through a Sephadex G-25M column and eluted with PBS (pH 7.5) to stop the reaction.

  Another protocol for treatment of periodate and sodium borohydride is described in Hickman et al. (1974) BBRC 57: 55-61. The purified protein is dialyzed in 0.01M sodium phosphate (pH 6.0), 0.15M NaCl. Sodium periodate is added to a final concentration of 0.01 M and the reaction is allowed to proceed for at least 6 hours at 4 ° C. in the dark. Treatment of β-hexosaminidase with periodate under these conditions sufficiently prevents protein uptake by fibroblasts; uptake is β- with M6P receptors on the surface of fibroblasts. It usually depends on the M6P moiety on hexosaminidase. Periodate oxidation therefore fully modifies M6P and diminishes its ability to interact with the M6P receptor.

  Alternatively, carbohydrates can be obtained from Thorpe et al. (1985) Eur. J. et al. Biochem. 147: 197-206 can be modified by treatment with periodate and sodium cyanoborohydride in a one-step reaction.

  The presence of carbohydrates in a partially deglycosylated protein or a protein with a modified glycosylation pattern should protect a potential polypeptide epitope, which can be exposed by the complete absence of glycosylation. . In the event that a therapeutic protein triggers an immune response, immunosuppressive therapy can be used with the therapeutic protein (Brooks (1999) Molecular Genetics and Metabolism 68: 268-275). For example, it has been reported that approximately 15% of Gaucher patients treated with alglucerase developed an immune response (Beutler et al., The Metabolic and Molecular Bases of Inherited Disease (8th edition) (2001). ), Scriver et al., Pp. 3635-3668). Fortunately, many patients (82/142) that produced antibodies to alglucerase were tolerated by conventional treatment regimens (Rosenberg et al. (1999) Blood 93: 2081-2088). Thus, in order to benefit a small number of patients who can develop an immune response, in some embodiments of the invention, patients receiving therapeutic proteins also receive immunosuppressive therapy.

(test)
This can be tested by exposing it to ConA to confirm that the protein is incompletely glycosylated. Incompletely glycosylated proteins are expected to demonstrate reduced binding to ConA Sepharose when compared to the corresponding fully glycosylated proteins.

  Also, actually incompletely glycosylated proteins can be analyzed by SDS-PAGE and compared to the corresponding fully glycosylated protein. For example, chemically deglycosylated GUS-GILT can be compared to enzymatically deglycosylated GUS-GILT prepared by untreated (glycosylated) GUS-GILT and PNGase A. Incompletely glycosylated proteins are expected to have greater mobility in SDS-PAGE when compared to fully glycosylated proteins.

  Incompletely glycosylated targeted therapeutic proteins exhibit uptake dependent on the targeting domain. Incompletely glycosylated proteins should exhibit reduced uptake (and preferably substantially no uptake) dependent on mannose or M6P. These properties can be confirmed experimentally in cell uptake experiments.

  For example, GUS-GILT protein synthesized in mammalian cells and subsequently treated with periodate and borohydride was tested for functional deglycosylation by testing M6P-dependent and mannose-dependent uptake. obtain. To demonstrate that M6P-dependent uptake is reduced, uptake assays are performed using GM4668 fibroblasts. In the absence of competitors, the treated and untreated enzymes each show significant uptake. The presence of excess IGF-II substantially reduces the uptake of treated and untreated enzymes, but the untreated enzyme retains residue uptake through an M6P-dependent pathway. Excess M6P reduces the uptake of untreated enzyme, but is substantially less efficient in reducing the uptake of functional deglycosylated protein. For treated and untreated enzymes, the simultaneous presence of both competitors substantially eliminates uptake.

  An uptake assay to assess mannose-dependent uptake is performed using J774-E cells, a murine macrophage-like cell line carrying a mannose receptor, and if necessary, using a small amount of the M6P receptor (Diment et al. (1987) J Leukocyte Biol.42: 485-490). Cells are cultured in DMEM, low glucose supplemented with 10% FBS, 4 mM glutamine, and antibiotic antifungal solution (Sigma, A-5955). Uptake assays using these cells are performed in the same manner as assays performed on fibroblasts. In the presence of excess M6P and excess IGF-II, which extinguish uptake due to any residual M6P / IGF-II receptor, the fully glycosylated enzyme is responsible for interacting with the mannose receptor. Due to significant uptake. Incomplete glycosylation enzymes are expected to show substantially reduced uptake under these conditions. Mannose receptor-dependent uptake of fully glycosylated enzymes can be competed by the addition of excess (100 μg / mL) mannan.

  The pharmacokinetics of deglycosylated GUS-GILT can be determined by giving intravenous injections of 20,000 enzyme units to groups of 3 MPSVII mice per time point. For each time point, 50 μL of blood is assayed for enzyme activity.

Example 14A. Deglycosylation of GUSΔC18-GILTΔ1-7
GUSΔC18-GILTΔ1-7 was treated with endoglycosylase F1 to reduce glycosylation. Specifically, 0.5 mg of GUSΔC18-GILTΔ1-7 together with 25 μL of endoglycosidase F1 (Prozyme, San Leandro, Calif.) For 7.5 hours at 37 ° C. in a final volume of 50 mM phosphate buffer at 250-300 μL. Incubated in (pH 5.5). After incubation, GUSΔC18-GILTΔ1-7 was repeatedly concentrated in a 50 kD molecular weight cut-off Centricon spin concentrator to separate GUSΔC18-GILTΔ1-7 from endoglycosidase F1.

  In separation experiments, GUSΔC18-GILTΔ1-7 was treated with both endoglycosidase F1 and endoglycosidase F2. Specifically, 128 μL of GUSΔC18-GILTΔ1-7 (7.79 million units / mL) was combined with 12.8 μL of each endoglycosidase (each from Prozyme) as described above, and a capacity of 325 μL of 50 mM phosphoric acid (pH 5.5). ) At 37 ° C. for 6 hours and then purified with a spin concentrator.

  1-3 μg of GUSΔC18-GILTΔ1-7 treated with endoglycosidase F1 was analyzed by electrophoresis on 7.5% polyacrylamide gel. This gel was stained with Coomassie brilliant blue dye (shown in FIG. 10). ΔΔN refers to untreated GUSΔC18-GILTΔ1-7. HBG5 refers to untagged human β-glucuronidase, which has a lower molecular weight and greater mobility than tagged GUSΔC18-GILTΔ1-7. Treated GUSΔC18-GILTΔ1-7 is denoted by ΔΔN + F1, which has an average moderately increased mobility compared to the untreated protein, consistent with at least partial deglycosylation of the protein. A further increase in mobility was observed in samples of GUSΔC18-GILTΔ1-7 treated with both endoglycosidase F1 and endoglycosidase F2 (data not shown), and only endoglycosidase F1 partially deglycosylated this protein. Suggests that

Example 14B. In Vitro Uptake of Deglycosylated Protein
Cultured mucopolysaccharide VII dermal fibroblasts GM4668 cells were used to evaluate uptake of GUSΔC18-GILTΔ1-7 treated with endoglycosidase F1 or untreated. Specifically, cells were grown at 37 ° C. in 5% CO ₂ in DMEM low glucose + 4 mM glutamine + antibiotic / antimycotic + 15% FBS (not heat inactive). Cell passage was performed with a scraper. Cells were cultured in T75 flasks to approximately 50-80% confluency (75 cm ² surface area). The day before the uptake experiment, cells were split into 12-well tissue culture plates (46 cm ² total surface area). The cells were allowed to reattach for 1-4 hours, then the medium was changed, after which the cells were allowed to grow overnight.

  Uptake experiments were performed in triplicate. The uptake medium is DMEM low glucose + 2% BSA (Sigma A-7030) +4 mM glutamine + antibiotic / antimycotic. The final volume of uptake medium + enzyme is 1 mL / well. Cells are washed with uptake medium and GUSΔC18-GILTΔ1-7 after uptake medium + β-glucuronidase (M6P), untreated GUSΔC18-GILTΔ1-7 (GILT), or endoglycosidase F1 (GILT + F1) (approximately 4000 units). Incubated for 3 hours with or without 2 mM mannose-6-phosphate (+ M6P) or 2.86 mM recombinant IGF-II (+ Tag). The medium was then removed, each well was washed with 4 × 1 mL PBS, and the cells were washed with 200 μL lysis buffer (100 mL lysis buffer: 1 mL 1M Tris pH 7.5, 2 mL NaCI, 1 mL 0). .5M EDTA, 96 mL water; 10 μL NP-40 per 1 mL lysis buffer added immediately prior to use) and incubated at room temperature for 5-10 minutes. All lysates were transferred into microcentrifuge tubes and centrifuged at full speed for 5 minutes. 10 μL of lysate was assayed in duplicate for GUS activity. 10 μL of lysate was assayed in triplicate for total protein concentration using the Pierce's Micro BCA kit. 2 μL of uptake medium was also assayed in duplicate for GUS activity.

  As shown in FIG. 11, GUSΔC18-GILTΔ1-7 (endoglycosidase F1-treated or untreated) is efficiently taken up by fibroblasts, even in the presence of mannose-6-phosphate, while non- Incorporation of tagged β-glucuronidase is essentially eliminated by the presence of mannose-6-phosphate competitor. In contrast, IGF-II successfully inhibited uptake of GUSΔC18-GILTΔ1-7 protein and essentially attenuated uptake of protein treated with endoglycosidase F1. This indicates that the uptake of the processed protein is actually mediated by the GILTΔ1-7 tag.

Example 14C. In vivo uptake experiment
Experiments were performed using cannulated Sprague-Dawley rats to determine the half-life in the circulation of treated or untreated protein. 100,000 to 150,000 units of GUSΔC18-GILTΔ1-7 were administered intravenously and the protein was not treated (GUSΔC18-GILTΔ1-7), treated with endoglycosidase F1 and endoglycosidase F2 (+ F1 + F2), or After treatment with endoglycosidase F1 and endoglycosidase F2, GILT tag was destroyed by incubation with 5 mM dithiothreitol in phosphate buffered saline at 37 ° C. for 45 minutes. At regular intervals thereafter, blood was collected through a cannula, treated for 1 hour at 60 ° C. to inactivate endogenous rat glucuronidase and assayed for glucuronidase activity. The result is shown in FIG. Protein half-life was determined by KaleidaGraph v. The initial time point was evaluated using 3.52 by fitting it to an equation of the form y = e ^−kt . The half-life of the untreated protein is measured for 8 minutes. Treatment with endoglycosidase F1 and endoglycosidase F2 increases the half-life by about 13 minutes. This correlates with improved avoidance of the reticuloendothelial clearance system in deglycosylation. Treatment with DTT prevents uptake of the tagged enzyme through the IGF-II receptor and extends the half-life of the endoglycosidase F1 / endoglycosidase F2 treated protein for an additional approximately 16 minutes.

  In vivo delivery of untreated GUSΔC18-GILTΔ1-7 (GILT), endoglycosidase F1-treated GUSΔC18-GILTΔ1-7 (GILT + F1) to tissues was evaluated and compared to the delivery of untagged β-glucuronidase (M6P). All proteins were produced in CHO cells. 1 mg M6P, GILT or GILT + F1 per kg body weight was administered to MPSVII mice by infusion of this protein into the tail vein. Control animals (control) were injected with buffer only. After 24 hours, the animals were sacrificed and tissue samples were collected for analysis. The data illustrated in FIGS. 13-15 are results from injections of 6-7 animals with each enzyme.

  FIG. 13 shows the levels of various enzymes detected in the liver, spleen, and bone marrow. Macrophages are abundant in these tissues and remove this enzyme from the circulation, primarily through high mannose receptors. GILT + F1 has less high mannose carbohydrate than the untreated enzyme and less accumulation in these tissues is expected. In liver and bone marrow, the accumulation of endoglycosidase F1 processing enzyme was almost as high as that of untreated enzyme. However, there was a notable decrease in GILT + F1 accumulation in the spleen. As shown in FIG. 14, both treated GUSΔC18-GILTΔ1-7 protein and untreated GUSΔC18-GILTΔ1-7 protein efficiently reached the heart, kidney and lung tissue, and an equivalent accumulation of untagged β-glucuronidase. Or have excessive accumulation. Thus, the GILTΔ1-7 tag allows targeting to the same tissue as mannose-6-phosphate with equal or better efficiency. Furthermore, as shown in FIG. 15, both treated and untreated GUSΔC18-GILTΔ1-7 proteins reach brain tissue at levels above those observed for untagged β-glucuronidase. However, the accumulation level in brain tissue was much lower than the accumulation level in other tissues.

  Also, tissues selected from the injected animals were subjected to histology and β-glucuronidase was detected as described in Wolf et al. (1992) Nature 360: 749-753. β-glucuronidase is detected as a red stain; nuclei stain blue. The results of liver histology for untagged β-glucuronidase (HGUS), untreated GUSΔC18-GILTΔ1-7 (Dd-15gilt), and endoglycosidase F1-treated GUSΔC18-GILTΔ1-7 (dd15F1) are shown in FIG. On HGUS slides, the extended Kupffer cells stain strongly red, while the more dense cells, primary hepatocytes, contain relatively little red staining and are blue nuclei surrounded by whitish areas in the figure. Appears as In contrast, the hepatocytes of the other two slides containing untreated GUSΔC18-GILTΔ1-7 or treated GUSΔC18-GILTΔ1-7 stained to some extent strongly, which is a sudden decrease in the number of blue nuclei surrounded by white Proven by This helps to elucidate why even GUSΔC18-GILTΔ1-7 treated with endoglycosidase F1 is still significantly localized in the liver. Tagged proteins reach a more diverse subset of cell types in the liver. Thus, there may be fewer subjects for clearance by high mannose receptors, but nevertheless a large amount of protein reaches the liver. This is because it targets more diverse cells.

  Additional histological data is shown in FIGS. FIG. 17 shows the localization of the kidney to the glomerulus. Histochemical staining is rarely quantitative, but slides from animals injected with GUSΔC18-GILTΔ1-7 (shown as Dd-15 (untreated) and F1 (treated with endoglycosidase F1)) are untagged. Staining more intensely than slides from animals injected with protein is reproducibly shown in the glomeruli. As shown in FIG. 18 and FIG. 19, the tagged protein is treated with endoglycosidase F1 (dd-15F1) or (dd-15), and the same protein as the cell to which the untagged protein (HGUS) arrives. It is considered reachable.

Example 14D. Deglycosylated Lec-1 Producing Protein
GUSΔC18-GILTΔ1-7 protein is produced in Lec1 cells treated with endoglycosidase F1 and tested in vitro and in vivo for targeting and uptake. Lec1 cells produced from ATCC (ATCC accession number CRL1735) are CHO-derived cells lacking β-1,2-N-acetylglucosaminyltransferase I (GlcNAc-T1) activity (Stanley et al. (1975) Cell). 6 (2): 121-8; Stanley et al. (1975) PNAS 72 (9): 3323-7). GIcNAc-T1 adds N-acetylglucosamine to the center of the polysaccharide Man ₅ -GlcNAc ₂ -Asn. This addition is the first essential step in the synthesis of complex N-linked polysaccharides and hybrid N-linked polysaccharides. Therefore, Lec1 cell-derived glycoproteins lack complex polysaccharides and hybrid polysaccharides. Lecl cells produce glycoproteins that possess high mannose polysaccharides, but their structure is susceptible to removal of endoglycosidase F1. Thus, the production of GILT-tagged proteins in Lecl cells that are subsequently subjected to endoglycosidase F1 treatment necessarily results in deglycosylated proteins. When injected into the animal's bloodstream, such proteins avoid clearance by mannose receptors in macrophages resident in the liver and accumulate at higher levels in the target tissue.

  The β-GUSΔC18-GILTΔ1-7 cassette in the expression plasmid pCXN was electroporated into Lecl cells at 50 μF and 1,200 V in a 0.4 cm cuvette. Cells were grown in α-MEM supplemented with 15% FBS, 1.2 mM glutamine, 50 μg / mL proline, and 1 mM pyruvate. Colony selection and amplification were mediated for 2-3 weeks with 400 μg / mL G418. Confluent cultures of pure strains were placed in medium (Weymouth medium supplemented with 2% FBS, 1.2 mM glutamine, and 1 mM pyruvic acid). Medium containing secreted recombinant enzyme was collected every 24-72 hours. One strain Lec1-18 was selected for enzyme production that produced the highest yield of recombinant enzyme.

  Recombinant enzyme from Lecl cells was affinity purified using anti-human β-glucuronidase Affigel 10 resin as described in Example 10. The uptake of GILT-tagged enzyme produced by Lecl is compared to the uptake of tagged or untagged enzyme produced by CHO. As shown in Table 9, the tagged Lecl producing enzyme has a higher M6P independent uptake than the corresponding tagged CHO producing enzyme, and a much higher M6P independent uptake than the untagged (CHO producing) enzyme. Indicates. Treatment of Lec1 producing enzyme with endoglycosidase F1 removes M6P-independent binding but not IGF-II-dependent binding. As shown in FIG. 23, SDS-PAGE of the endoglycosidase F1-treated Lec1 producing enzyme and the untreated Lec1 producing enzyme showed a clear mobility shift approaching that of PNGase F-treated enzyme, which was completely removed. Glycosylated. This suggests that treatment of the Lec1 producing enzyme with endoglycosylase F1 causes a significant loss of glycosylation. In contrast, neither CHO-producing enzyme nor HEK-producing enzyme show a clear shift in mobility after treatment with endoglycosidase F1.

エンドグリコシダーゼＦ１処理Ｌｅｃ１産生酵素およびエンドグリコシダーゼＦ１処理ＣＨＯ産生酵素、ならびに未処理ＨＥＫ２９３産生酵素を、１ｍｇ／ｋｇの容量で、それぞれ３匹の免疫寛容性ＭＰＳＶＩＩマウス内に注入した（Ｓｌｙら（２００１）ＰＮＡＳ ９８：２２０５−２２１０）。２４時間後、このマウスを屠殺し、そして組織を酵素活性についてアッセイした。図２４において示されるように、Ｌｅｃ１産生酵素は、ＣＨＯ産生酵素が蓄積するよりも、肝臓においてより少なく、そして心臓、筋肉および腎臓の標的組織においてより多く蓄積した。ＨＥＫ２９３産生酵素は、Ｌｅｃ１産生酵素より、心臓、筋肉および腎臓において、僅かに高い酵素の蓄積を示した。

Endoglycosidase F1-treated Lec1 producing enzyme, endoglycosidase F1-treated CHO producing enzyme, and untreated HEK293 producing enzyme were injected into 3 immunotolerant MPSVII mice each at a volume of 1 mg / kg (Sly et al. (2001)). PNAS 98: 2205-2210). After 24 hours, the mice were sacrificed and the tissues were assayed for enzyme activity. As shown in FIG. 24, Lecl producing enzyme accumulated less in the liver and more in the heart, muscle and kidney target tissues than did the CHO producing enzyme. HEK293 producing enzyme showed slightly higher enzyme accumulation in heart, muscle and kidney than Lec1 producing enzyme.

  The measured enzyme accumulation in the targeted tissue depends on both the efficiency of targeting to the tissue and the rate of clearance of the enzyme from the tissue. If the Lec-1 producing enzyme is degraded more slowly than the CHO producing enzyme upon proper targeting to the lysosome, this may explain a higher observed level of accumulation. On the other hand, more rapid degradation of the Lec-1 producing enzyme may mask the enhanced targeting efficacy. To address this possibility, the half-life of treated Lec-1 producing enzyme and treated CHO producing enzyme as well as untreated Lec-1 producing enzyme and untreated CHO producing enzyme was tested in an uptake assay by MPSVII fibroblasts. The fibroblasts were incubated with the enzyme for 3 hours, the cells were washed, transferred into fresh media and the media was changed daily. Duplicate wells were lysed for each enzyme on days 0, 3, and 7. The fraction of residual enzyme was plotted against time and this data was fit to an exponential equation. As shown in FIG. 25, endoglycosylase F1-treated Lec-1 producing enzyme has a half-life similar to that of HBG5 in human fibroblasts, and a half-life slightly shorter than that of CHO-producing enzyme It was found to have

(Incorporation by reference)
Patent literature, scientific publications and protein data bank records disclosed herein, and US Provisional Patent Application No. 60 / 250,446 (filed Nov. 30, 2000); No. 287,531 (filed on Apr. 30, 2001); U.S. Provisional Application No. 60 / 290,281 (filed on May 11, 2001); U.S. Provisional Application No. 60 / 304,609, (2001) U.S. Provisional Application No. 60 / 329,461 (filed Oct. 15, 2001); International Patent Application No. PCT / US01 / 44935 (filed Nov. 30, 2001); Provisional application 60 / 351,276 (filed January 23, 2002); US 10 / 136,841 and 10 / 136,639 (filed April 30, 2002), Country 60 / 384,452 (filed May 29, 2002); US 60/386019 (filed June 5, 2002); and US 60 / 408,816 (September 2002) 6 application) are incorporated by reference in their entirety in this application.

FIG. 1 shows several types of incomplete glycosylation. FIG. 2A is a map of the human IGF-II open reading frame (SEQ ID NO: 1) and its encoded protein (SEQ ID NO: 2). Mature IGF-II lacks the signal peptide cleavage region and the COOH cleavage region. The IGF-II signal peptide and mature polypeptide can be fused to the GAA coding sequence. The IGF-II moiety can be modified to incorporate various mutations that enhance the selective binding of IGF-II to the IGF-II receptor. FIG. 2B is a map of the human IGF-II open reading frame (SEQ ID NO: 1) and its encoded protein (SEQ ID NO: 2). Mature IGF-II lacks the signal peptide cleavage region and the COOH cleavage region. The IGF-II signal peptide and mature polypeptide can be fused to the GAA coding sequence. The IGF-II moiety can be modified to incorporate various mutations that enhance the selective binding of IGF-II to the IGF-II receptor. FIG. 3 is IGF-II optimized for Leishmania codon, shown at the XbaI site of pIR1-SAT; the nucleic acid is SEQ ID NO: 3 and the coding protein is SEQ ID NO: 4. FIG. 4A is a depiction of a preferred embodiment of the present invention, incorporating a signal peptide sequence, an adult human β-glucuronidase sequence, a three amino acid bridge, and an IGF-II sequence. The nucleic acid shown is SEQ ID NO: 5 and the coding protein is SEQ ID NO: 6. FIG. 4B is a depiction of a preferred embodiment of the present invention, incorporating a signal peptide sequence, an adult human β-glucuronidase sequence, a three amino acid bridge, and an IGF-II sequence. The nucleic acid shown is SEQ ID NO: 5 and the coding protein is SEQ ID NO: 6. FIG. 4C is a depiction of a preferred embodiment of the present invention, incorporating a signal peptide sequence, an adult human β-glucuronidase sequence, a three amino acid bridge, and an IGF-II sequence. The nucleic acid shown is SEQ ID NO: 5 and the coding protein is SEQ ID NO: 6. FIG. 4D is a depiction of a preferred embodiment of the present invention, incorporating a signal peptide sequence, an adult human β-glucuronidase sequence, a three amino acid bridge, and an IGF-II sequence. The nucleic acid shown is SEQ ID NO: 5 and the coding protein is SEQ ID NO: 6. FIG. 4E is a depiction of a preferred embodiment of the present invention, incorporating a signal peptide sequence, an adult human β-glucuronidase sequence, a three amino acid bridge, and an IGF-II sequence. The nucleic acid shown is SEQ ID NO: 5 and the coding protein is SEQ ID NO: 6. 5, GUS, GUS-IGF-II fusion protein (GILT-GUS), the GILT-GUS with mutations of Δ1-7 and Y27L in IGF-II portion ^(GILT 2-GUS) or negative control (DMEM) FIG. 2 shows β-glucuronidase (GUS) activity in exposed human mucopolysaccharidosis VII dermal fibroblast GM4668 cells. FIG. 6 shows GUS activity in GM4668 cells exposed to GUS (+ βGUS), GUS-GILT (+ GILT), GUS-GILT in the presence of excess IGF-II (+ GILT + IGF-II) or negative control (GM4668). . FIG. 7 is an alignment of human IGF-I (SEQ ID NO: 7) and human IGF-II (SEQ ID NO: 8) showing the A domain, B domain, C domain and D domain. FIG. 8 shows GUS, GUS-GILT, GUS-GILT, GUS-GILT with 7 amino terminal residues deleted (GUS-GILTΔ1-7), GUS-GILT in the presence of excess IGF-II, excess GUS in GM4668 cells exposed to GUS-GILTΔ1-7 in the presence of IGF-II, or negative control (Mock). FIG. 9A illustrates one form of phosphorylated high mannose carbohydrate structure attached to a glycoprotein via an asparagine residue, and also illustrates the structure of mannose and N-acetylglucosamine (GlcNAc). FIG. 9B illustrates portions of the high mannose carbohydrate structure in greater detail and shows portions that are susceptible to cleavage by periodate treatment. The position of the sugar residue in the carbohydrate structure is capital letter A to H labeled with a circle; the phosphate group is indicated by a capital letter P surrounded by a circle. FIG. 10 illustrates SDS-PAGE analysis of GUSΔC18-GILTΔ1-7 treated with GUSΔC18-GILTΔ1-7 (ΔΔN), GUS (HBG), and endoglycosidase F1 (ΔΔN + F1). FIG. 11 shows untagged β-glucuronidase (M6P), GUSΔC18-GILTΔ1-7 (GILT), or in the absence of competitor (enzyme only) or mannose-6-phosphate (+ M6P) or IGF-II ( FIG. 4 illustrates the uptake of GUSΔC18-GILTΔ1-7 (GILT + F1) treated with endoglycosidase F1 in the presence of + Tag). FIG. 12 shows the results of an in vivo half-life measurement experiment of GUSΔC18-GILTΔ1-7, untreated, treated with endoglycosidase F1 and endoglycosidase F2 (+ F1 + F2), or further treated with dithiothreitol (+ F1 + F2 + DTT). FIG. 13 shows measurement experiments on the accumulation of injected untagged β-glucuronidase (M6P), GUSΔC18-GILTΔ1-7 (GILT) or endoglycosidase F1-treated GUSΔC18-GILTΔ1-7 (GILT + F1) in the liver, spleen, or bone marrow. The results of are illustrated. FIG. 14 shows accumulation of injected untagged β-glucuronidase (M6P), GUSΔC18-GILTΔ1-7 (GILT) or endoglycosidase F1-treated GUSΔC18-GILTΔ1-7 (GILT + F1) in heart, kidney, or lung tissue. The results of the measurement experiment are shown. FIG. 15 illustrates the results of an experiment measuring the accumulation of injected untagged β-glucuronidase (M6P), GUSΔC18-GILTΔ1-7 (GILT) or endoglycosidase F1-treated GUSΔC18-GILTΔ1-7 (GILT + F1) in brain tissue. To do. FIG. 16 shows the histology of liver tissue from animals injected with untagged β-glucuronidase (HGUS), GUSΔC18-GILTΔ1-7 (Dd-15gilt), or GUSΔC18-GILTΔ1-7 (dd15F1) treated with endoglycosidase F1. Figure 2 illustrates a dynamic analysis. FIG. 17 shows the renal tissue of animals injected with untagged β-glucuronicidase (HGUS), GUSΔC18-GILTΔ1-7 (Dd-15), or GUSΔC18-GILTΔ1-7 (F1) treated with endoglycosidase F1. The histological analysis is illustrated. FIG. 18 shows bone tissue of animals injected with untagged β-glucuronidase (HGUS), GUSΔC18-GILTΔ1-7 (dd-15), or GUSΔC18-GILTΔ1-7 (dd-15F1) treated with endoglycosidase F1. The histological analysis is illustrated. FIG. 19 shows spleen tissue of animals injected with untagged β-glucuronidase (HGUS), GUSΔC18-GILTΔ1-7 (Dd-15), or GUSΔC18-GILTΔ1-7 (Dd-15F1 :) treated with endoglycosidase F1. The histological analysis of is illustrated. FIG. 20 illustrates an exemplary GILT-tagged GAA cassette having a targeting moiety (GILT) comprising an optimized signal peptide fused to the N-terminus of acid α-glucosidase (GAA). FIG. 21 illustrates an exemplary GILT-tagged GAA cassette with a targeting moiety fused to the C-terminus of GAA. FIG. 22 shows the GAA cDNA sequence of Human Image cDNA clone no. 4374238 (SEQ ID NO: 23) and its encoded GAA protein (SEQ ID NO: 24). FIG. 23 shows SDS-PAGE analysis of GUSΔC18-GILTΔ1-7 produced in endoglycosidase F1 treated (+) or untreated CHO cells (CHO15), Lec1 cells (LEC18), or HEK293 cells (HEK). Is illustrated. Lanes marked with “P” show enzyme treated with PNGase F (expected to fully deglycosylate enzyme) FIG. 24 shows the results of a 24-hour GUSΔC18-GILTΔ1-7 accumulation experiment in immunotolerant MPSVII mice using enzymes produced with endoglycosidase F1-treated CHO or endoglycosidase F1-treated Lec1 and untreated HEK293. Is illustrated. FIG. 25 shows in human MPSVII fibroblasts using GUSΔC18-GILTΔ1-7 produced in untagged human β-glucuronidase (HBG5) or endoglycosidase F1-treated (+ F1) or untreated CHO cells or Lec1 cells. The results of the half-life experiment are illustrated.

Claims (23)

An incompletely glycosylated targeted therapeutic fusion protein comprising:
  A lysosomal enzyme; and
  Lysosomal targeting domain that binds to human cation-independent mannose-6-phosphate receptor in a manner independent of mannose-6-phosphate
Wherein the lysosomal targeting domain is mature human IGF-II or a human IGF-II mutein having an amino acid sequence at least 70% identical to mature human IGF-II protein. 2. The incompletely glycosylated targeted therapeutic fusion protein of claim 1, wherein the lysosomal targeting domain is mature human IGF-II. 2. The incompletely glycosylated targeted therapeutic fusion protein of claim 1, wherein the lysosomal targeting domain is a human IGF-II mutein having an amino acid sequence that is at least 70% identical to mature human IGF-II. 4. The incompletely glycosylated targeted therapeutic fusion protein of claim 3, wherein the human IGF-II mutein comprises amino acids 48-55 of mature human IGF-II. The human IGF-II mutein comprises at least 3 amino acids selected from the group consisting of amino acid position 8, amino acid position 48, amino acid position 49, amino acid position 50, amino acid position 54, and amino acid position 55 of mature human IGF-II 4. The incompletely glycosylated targeted therapeutic fusion protein of claim 3, wherein the relative position of the at least three amino acids corresponds to their position in mature human IGF-II. The human IGF-II mutein comprises an amino acid sequence of IGF-I or an amino acid sequence of an IGF-I mutein, wherein:
  (I) whether amino acids 55 and 56 are changed,
  (Ii) amino acid positions 1 to 4 have been removed or altered, or
  (Iii) amino acids 55 and 56 are changed and amino acids 1-4 are removed or changed,
2. The incompletely glycosylated targeted therapeutic fusion protein of claim 1. 4. The human IGF-II mutein comprises amino acids that are each uncharged or negatively charged at pH 7.4 at positions corresponding to positions 54 and 55 of mature human IGF-II. Incompletely glycosylated targeted therapeutic fusion protein. 4. The incompletely glycosylated targeted therapeutic fusion protein of claim 3, wherein the human IGF-II mutein comprises the amino acid sequence phenylalanine-arginine-serine. 4. The incompletely glycosylated targeted therapeutic fusion protein of claim 3, wherein the human IGF-II mutein comprises amino acids 8-28 and 41-61 of mature human IGF-II. 4. The incompletely glycosylated targeted therapeutic fusion protein of claim 3, wherein the mutein comprises a deletion or substitution at amino acids 1-7 of mature human IGF-II. 4. The incompletely glycosylated targeted therapeutic fusion protein of claim 3, wherein the mutein comprises a human IGF-II Y27L mutation. The incomplete glycosylation targeting of claim 1, wherein the mutein binds to the IGF-I receptor with reduced affinity compared to the affinity of naturally occurring human IGF-II for the IGF-I receptor. Therapeutic fusion protein. 2. The incompletely glycosylated targeted therapeutic fusion protein of claim 1, wherein the cellular or subcellular deficiency in enzyme activity is associated with a human disease. 14. The incompletely glycosylated targeted therapeutic fusion protein of claim 13, wherein the human disease is lysosomal storage disease. 15. The incompletely glycosylated targeted therapeutic fusion protein of claim 14, wherein the lysosomal storage disease is Pomp disease. 15. The incompletely glycosylated targeted therapeutic fusion protein of claim 14, wherein the lysosomal storage disease is Fabry disease. 15. The incompletely glycosylated targeted therapeutic fusion protein of claim 14, wherein the lysosomal storage disease is Gaucher disease. The incompletely glycosylated targeted therapeutic fusion protein of claim 1, wherein the lysosomal enzyme is acid α1,4-glucosidase. 2. The incompletely glycosylated targeted therapeutic fusion protein of claim 1, wherein the lysosomal enzyme is β-glucuronidase. The incompletely glycosylated targeted therapeutic fusion protein of claim 1, wherein the lysosomal enzyme is α-galactosidase A. 21. The incompletely glycosylated targeted therapeutic fusion protein of any one of claims 1-20, wherein the targeted therapeutic fusion protein is deglucosylated. 19. The incompletely glycosylated targeted therapeutic fusion protein of claim 18, wherein the targeted therapeutic fusion protein is deglucosylated by periodate treatment. 23. A pharmaceutical composition for treating a patient comprising the incompletely glycosylated targeted therapeutic fusion protein of any one of claims 1-22.

Images