MXPA97007137A - Stable protein compositions: fosfolipidos ymeto - Google Patents

Abstract

The present invention relates to stable compositions of proteins and related methods, in a protein capable of transition in the dissolved globular state, is brought into contact with a negatively charged lipid vesicle, thus stabilizing the protein against thermal aggregation induced, denaturation and loss of activity. The protein: phospholipid complex directly stabilizes the secondary and tertiary structure of the protein, and the compositions are useful in high temperature formulations and in new delivery vehicles.

Description

STABLE PROTEIN COMPOSITIONS: PHOSPHOLIPIDS AND METHODS FIELD OF THE INVENTION The present invention relates to protein: phospholipid structures which are useful for stabilizing the secondary and tertiary structure of proteins capable of transitioning in the dissolved globular state. More particularly, this invention relates to compositions -G-CSF: phospholipid and MGDF: phospholipid which possess increased stability, exhibit increased shelf life, and are capable of being used in high temperature formulations and as novel delivery vehicles.

BACKGROUND OF THE INVENTION Several proteins have shown transition in a dissolved globular state (MGS). Van der Goot, F. G., Nature -354, 408-410 (1991). The proteins in the disintegrated globular state have a secondary structure which is comparable with the native protein, although they lack - of the rigid tertiary structure. Pitsyn et al., FEBS Le-tters 262: 1, 20-24 (1990). In some cases, the transition in this state is accompanied by exposure of previously hidden hydrophobic regions of the protein. For the exposition of critical hydrophobic residues, the MGS can be an intermediate in the aggregation and precipitation of proteins. The MGS conformation can be detected by co-parasitization of the circular-dichroism in the distant UV region with the spectrum of aromatic side chains (near UV circular dichroism and fluorescence). The dissolved globular state exhibits spectral changes of the aromatic group in the absence of changes in the distant circular UV dichroism, Bychkova et al., FEBS Letters 238: 231-234 (1988), and may be involved in the penetration of the membrane by some proteins, Bychkova et al., FEBS Letters 238: 231-234 (1988); Van der Goot, F. G., Nature 354, 408-410 (1991).

Granulocyte colony stimulating factor (G-CSF) is a known protein for the transition in MGS before aggregation. Recombinant human G-CSF selectively stimulates neutrophils, a type of cell-white in the blood used to fight the infection. Currently, Filgrastim, a recombinant G-CSF, is available for therapeutic use. The structure of low G-CSF - several conditions has been studied extensively; Lu et al., J. Biol. Chem. Vol. 267, 8770-8777 (1992). Given its hydrophobic characteristics, G-CSF is difficult to formulate to extend its shelf life. Formulations of certain hydrophobic proteins lose activity due to the formation of lower order aggregates and higher order aggregates (macro chain) during long term storage. Other chemical changes, such as deamidation and oxidation, may also occur during storage. In order, the formulator G-CSF can protect against denaturation and, in particular, take care of maintaining the stability of the secondary and tertiary structure of the protein.

Human GM-CSF is a 22-kDa glycoprotein continuously re-wanted for in vitro proliferation of macrophages and granulocyte progenitor cells. It also -controlles the irreversible role of these parents to -form granulocytes and macrophages. Other biological activities may include regulation of the functional activity of mature cell typos; Gough et al., Nature, 309, 763-767 (1984), and they increase the chemotaxis towards recognized chemoattractants; Williams et al., Hematology, 4th. ed. ( 1990 ). GM-CSF also stimulates the production of monocytes, and thus may be useful in the treatment of monocytic disorders, such as monocytopenia.

Human GM-CSF can be obtained and purified from a number of sources. Methods for the production of recombinant human GM-CSF have previously been described by Burgess et al., Blood, 69: 1, 43-51 (1987).

The U.S. Patent No. 5,047,504 (Boone), incorporated herein by reference, has allowed the production of large commercial quantities of GM-CSF in non-glycosylated form as a product of the expression of a prokaryotic host cell.

MGDF, or megakaryocyte growth and differentiation factor, is a recently cloned cytosine that appears to be the major regulator of circulating platelet levels. See Bartley, T.D. et al., Cell 77: 1117-1124 - (1994); Lok, S. et al., Nature 369: 565-568 (1994); de Sauvage, F.J. et al., Nature 369: 533-538 (1994); Miyazake, H. et al., Exp. Hematol. 22: 838 (1994); and Kuter, D.J. et al., PNAS USA, 91: 11104-11108 (1994). MGDF is also -called as a thrombopoietin (TP0), as a ligand-mpl, and as a megapoietin. The MGDF of the mature human is a protein that has 332 amino acids in total. The sequence of this protein and the corresponding cDNA are shown in FIGURE 29 here.

The recombinant MGDF produced in both the Ovary of Chinese hamsters (CHO) as well as in E. coli cells, has been shown to possess a biological activity to stimulate or -increment specifically megakaryocytes and / or platelets in vivo in mice, rats and monkeys. See, e.g., Hunt, P. et al., Blood 84 (10): 390A (1994). The human MGDF molecules that have been truncated so that they extend at least 151 amino acids, starting from the amino acid at position 1 in FIGURE 29, retain the biological activity -in vivo. It is also possible to remove up to the first six amino acids at the N-terminus of the human MGDF protein sequence and to preserve the biological activity. Therefore, it appears that the biological activity is conserved within amino acids 7 to 151 (inclusive) of the amino acid-human sequence of MGDF shown in FIGURE 29.

The MGDF that occurs naturally is a glycosylated molecule. The glycosylation model of the natural MGDF is related to two key domains that have been found in the MGDF. The sequence of approximately the first 151 amino acids of the human MGDF, corresponding to an active portion of the molecule, exerts a remarkable homology -to the erythropoietin (EPO), a cytosine capable of stimulating the production of erythrocytes, and is referred to as the "EPO-analogue" domain of the human MGDF. The remaining amino acids of the mature protein constitute a domain called "N-linked carbohydrate", since these amino acids include more, if not all, of the sites for N-linked glycosylation. In human MGDF, there are six N-linked glycosylation sites, all contained in the N-linked glycosylation domain. Both domains contain O-linked glycosylation sites. There is an estimated 12-14 O-linked glycosylation chains in the molecule. Experimental evidence with human MGDF DNA expressed recombinantly in CHO cells reveals that in the EPO-analogous domain, at least two O-linked sites are glycosylated at positions 1 (Ser) and 37 (Thr) .

While proteins such as G-CSF and MGDF can be stabilized under certain defined conditions, there is nevertheless a need to extend the shelf life of these and other proteins by stabilizing the secondary and tertiary structure of the proteins. A pathway that has been tried in the past to work with such proteins is the use of liposomes. The liposomes are completely closed lipid bilayer membranes formed by water-insoluble polar lipids, particularly phospholipids. The liposomal vesicles may have a bilayer membrane -simple (unilamellar) or may have multiple bilayer membranes (multilamellar). The bilayer is composed of two monolayers-of lipid having a hydrophilic "tip" (polar) region and a hydrophobic (non-polar) "tail" region where the hydrophobic tails are oriented toward the center of the bilayer, while Hydrophilic tips are oriented towards the aqueous phase The stability, rigidity, and permeability of the liposomes can be altered by changes in the composition of the phospholipid, by changes in temperature, by the inclusion of a sterol or by the incorporation of a charged -row, The basic structure of liposomes can be elaborated by a variety of techniques known in the art.

In the process of their formation, liposomes can trap water solutes in aqueous channels and release them at variable rates. Given the discovery that lipo-somes can introduce enzymes into cells and alter their metabolism (Gregoriadis, New Engl. J. Med. 295, 704-710, 765-770 (1976)), the liposomes were announced as the response to the search for the supply of the drug - in the objective. As a result, there is a great deal of experimental research in the pharmaceutical industry which involves the use of liposomes as slow-release deposits for drugs, vitamins and proteins sequestered within the hydrophobic layers or within the hydrophobic core of the drug. liposome The successful use of liposomes as drug carriers has been limited because attempts by researchers with such use have encountered various problems. For example, liposomes are known to act as powerful immunological ad-together to entrap antigens and -caution can be exerted when enzymes or other proteins of xenogenic origin are entrapped in liposomes. Also, the diffusion index of the drug is difficult to -control. This is a function of the inherent instability of liposomes and the presence of certain components of the blood which accelerate the diffusion of certain drugs. In addition, by their nature, some substances are poorly entrapped in the liposomes, and, therefore, diffuse rapidly in the circulation. Finally, it has been a problem with some cells or other organs that the liver or the spleen. An excellent analysis of liposomes, substances which have been incorporated into liposomes, and the problems associated with the use of liposomes as carriers of drugs is "Liposomes" by Gregory Gregoriaidis, found in Drug Carriers in Biology and Medicine, Chapter 14 , 287-341 (Academic Press, NY, 1979).

While many attempts have been reported regarding the use of liposomes as carriers of drugs, many attempts have been made with respect to the use of liomasomes for purposes of increasing the shelf life of peptides or therapeutic proteins by stabilizing the structure - of the peptide and / or the structure of the protein. In the international application No. PCT / US90 / 05163, entitled "Therapeutic -Peptides and Proteins", Hostetler, et al., The use of empty liposomes is disclosed as a pharmaceutically acceptable diluent, to solubilize polypeptides and / or proteins to prevent accumulation of the polypeptides and / or proteins in an air / water inter phase, and to prevent adsorption of the polypeptides and / or proteins on the surfaces of the container.

Hostetler et al. reveals that the negatively charged phospholipid can be added to about 50 percent mole, and that phosphatidylcholine, a neutral phospholipid, is the preferred liposome. Hostetler et al. it does not reveal a diluent - which shows stabilizing the structure of a polypeptide and / or - a protein.

In the international application No. PCT / US91 / 07694, entitled M Preparation and Characterization of Liposomal Formula-tions of Tumor Necrosis Factor ", Hung et al., Describes a lipophilically modified tumor necrosis factor (TNF) molecule. in association with the surface or encapsulated within a liposome Lipophilic liposomal TNF molecules are reported to possess an increase in stability in vivo Stability refers to a decrease or a tendency to decrease of the liposome-TNF for , letting out TNF in the in vivo system The preferred liposomes - were neutral lipids Hung et al does not reveal a TNF composition in which the excipients have a stabilizing effect on the structure of the protein.

Nothing can be written from the literature with respect to contacting a protein, e.g. G-CSF or MGDF, with a negatively charged lipid vesicle, by -these directly stabilizing the protein against the thermally induced aggregation, denaturation, loss of activity, and the splitting of the secondary structure. This need exists for such compositions, which provide the benefit of being useful in formulation processes that require high temperatures - eg incorporation of G-CSF and / or MGDF in polymers - as well as being used as new delivery vehicles ( eg oral administration of pegylated G-CSF). The present invention provides such compositions.

BRIEF DESCRIPTION OF THE INVENTION The present invention is directed to the addition of hydrophobic excipients, e.g. smooth-phospholipids or other liposomes, to a protein under conditions in the dissolved globular state, directly stabilizing the secondary and tertiary structure of the protein, thus protecting the protein-against thermally induced aggregation, denaturation and loss of activity In particular, the invention is directed to stable G-CSF: phospholipid and MGDF: phospholipid compositions. Surprisingly, the preferred G-CSF: phospholipid and MGDF phospholipid compositions can be cycled several times between 10-95 ° C with a complete recovery of the secondary structure of the protein in cooling. The compositions are useful for formulation processes that require high temperatures, as well as for use as new supply vehicles. In addition, the compositions have a prolonged shelf life, compared to a single protein, and the interaction of the protein with the phospholipid membrane prevents the adsorption of the protein in glass jars.

In a preferred embodiment, the protein: phospholipid complex comprises a negatively charged liposome which is selected from dioleoylphosphatidylglycerol (DOPG), dimyristoylphosphatidylglycerol (DMPG), dipalmitoylphosphatidylglycerol (DPPG), egphosphatidylglycerol, dioleoylphosphatidylethanolane (DOPE), egphosphatidylethanolamine, dioleoylphosphatidic acid (DOPA), dimyristoylphosphatidic acid (DMPA), dipalmitoylphosphatidic acid (DPPA), dioleoylphosphatidylserine (DOPS), dimyristoylphosphatidylserine (DMPS), dipalmitoylphosphatidylserine (DPPS), egphosphatidylserine, lysophosphatidylglycerol, lysophosphatidylethanolamine, lysophosphatidylserine. DOPG, a negatively charged unsaturated phospholipid, is -especially preferred. In addition, the invention comprises a pH maintained in the range of 3.0-7.5, and a lipid: protein ratio of at least 10: 1.

Additional elements that are preferred embodiments of the invention include the use of chemically modified proteins in the protein: phospholipid complex, as well as the use of one or more of the following: an iso-tonia adjusting agent; a regulatory agent; and a pH adjusting agent.

As will be understood by a person having knowledge in the art, the invention encompasses stable protein-phospholipid compositions having various combinations of these a-tional elements.

BRIEF DESCRIPTION OF THE FIGURES FIGURE 1 represents the fluorescence emission spectrum of rhG-CSF in the presence (curve 1) and in absence (curve 2) of DOPG vesicles. The concentration of rhG-CSF was 0.2 mg / ml. The D0PG: rhG-CSF ratio (curve 1) was 100: 1 (mol: mol).

FIGURE 2 shows the effect in increasing the - lipid: protein ratio on the fluorescence of rhG-CSF. F is the initial fluorescence (not lipid) and F refers to the fluorescence after the addition of the lipid to obtain the indicated molar ratio of lipid: rhG-CSF. Figure 2 (a) shows F / F (f |) and the maximum emission wavelengths (delta) for DOPG mixtures: rhG-CSF. The Figure 2 (b) shows F / F () and the maximum emission wavelengths (delta) for DOPC mixtures: rhG-CSF.

FIGURE 3 shows the Stern-Volmer traces of the fluorescence suppression rhG-CSF by KI in the absence (•) and presence (o) of DOPG vesicles. The deletion experiments were carried out by the addition of KI aliquots to rhG-CSF (0.2 mg / ml) and D0PG: rhG-CSF (100: 1 molar).

FIGURE 4 represents the suppression in fluorescence of tryptophan rhG-CSF by the addition of pyrene-decanoic acid. The wavelength of the emission was 327 nm. The D0PG: rhG-CSF ratio was 100: 1 (molar).

FIGURE 5 is a graph showing a comparison of changes in intensity F for rhG-CSF in the absence and presence of several lipids. In each case, the lipid: protein ratio was 100: 1 (molar).

FIGURE 6 is a graph showing a comparison of the changes in emission maxima for rhG-CSF in the absence and presence of several lipids. In each case, the lipid: protein ratio was 100: 1 (molar).

FIGURE 7 shows the effect of the DMPC (curve 2), of the DMPG (curve 3), and of the DMPA (curve 4) on the Circular Dichroism (CD) of the rhG-CSF (curve 1). In each case, the lipid to protein ratio was 50: 1 (molar) in water, pH 6.0.

FIGURE 8 shows the effect by increasing the -temperature on the Circular Dichroism of the rhG-CSF (curve 1) or the D0PG: rhG-CSF (140: 1 molar) (curve 2). The concentration of rhG-CSF was 80 (mu) g / ml in water, pH 6.0. The temperature was scanned at 10-90 ° C with an index of 100 ° C / hour.

FIGURE 9 shows the differential scanning calorimetry thermograms for rhG-CSF (curve 1) and for -D0PG: rhG-CSF (45: 1 molar) (curve 2). The concentration of rhG-CSF in the samples was 1 mg / ml (pH 7.0 in water). The exploration index was 90 ° C / hour.

FIGURE 10 shows the effect of the -temperature cycles on the Circular Dichroism of the rhG-CSF (curve 1) and of. D0PG: rhG-CSF (140: 1 molar) (curve 2). The samples were rapidly heated to 95 ° C and cooled to 10 ° C as indicated by the arrows. The concentration of rhG-CSF in the samples was 80 (mu) g / ml, pH 6.0.

FIGURE 11 shows the effect of the -temperature cycles on Circular Dichroism (CD) of rhG-CSF (curve 1) and DMPG: rhG-CSF (150: 1 molar) (curve 2). The samples were heated to 95 ° C and cooled to 10 ° C. The concentration of rhG-CSF in the samples was 80 (mu) g / ml, pH 6.0.

FIGURE 12 shows the effect of the -temperature cycles on Circular Dichroism (CD) of rhG-CSF (curve 1) and DPPG: rhG-CSF (150: 1 molar) (curve 2). The samples were heated to 95 ° C and cooled to 10 ° C. The concentration of rhG-CSF in the samples was 80 (mu) g / ml, pH 6.0.

FIGURE 13 is a graph showing the ability of several lipids to stabilize rhG-CSF during freezing drying. The lipid: protein ratio was 100: 1 in each case. The stability was based on the retention of in vitro activity in the test samples consisting of bone marrow. RhG-CSF alone does not survive the secant freezing process so that the control used is rhG-CSF untreated in the absence of the lipid.

FIGURE 14 shows the effects of various lipids on the in vivo activity of rhG-CSF. The activity (WBC count) was measured after a subcutaneous injection to hamsters. The dose of rhG-CSF was 100 (mu) g / kg with a lipid to protein ratio of 100: 1.

FIGURE 15 shows the effects of several lipids on the in vivo activity of rhG-CSF. The activity (WBC count) was measured after a subcutaneous injection to hamsters. The dose of rhG-CSF was 100 (mu) g / kg with a lipid: protein ratio of 50: 1.

FIGURE 16 is a graph showing a comparison of changes in intensity F for CH0-G-CSF in the absence and presence of DOPG with variant pH's. In each case, the lipid: protein ratio was 100: 1 (molar).

FIGURE 17 is a graph showing a comparison of the changes in emission maxima for CHO-G-CSF in the absence and presence of DOPG with variant pH's. In each case, the proportion of lipid protein was 1001 (molar).

FIGURE 18 shows the effect of the -temperature cycles on the Circular Dichroism (CD) of the PEG-G-CSF () and DMPG: PEG-G-CSF (17: 1 molar) (). The samples were heated to 90 ° C and cooled to 10 ° C.

FIGURE 19 shows: (a) the effect of the temperature cycles on the Circular Dichroism (CD) of the GM-CSF in PBS, pH 7.0. The GM-CSF at 10 ° C () is compared to the GM-CSF which was heated to 90 ° C and then cooled to 10 ° C (); (b) the effect of the temperature cycles on the DPPG CD: PEG-C-CSF (17: 1 molar). The DPPG: GM-CSF at 10 ° C () is compared to the DPPG: GM-CSF which was heated to 90 ° C and then cooled to 10 ° C ().

FIGURE 20 shows: (a) the results of the total WBC response to the intraduodenal infusion of rhG-CSF in the absence and presence of DOPG. The dose of rhG-CSF was -750 (mu) g / kg and the lipid: protein ratio was 100: 1; (b) the results of the total WBC response to intraduodenal infusion of PEG-G-CSF in the absence and presence of DOPG.

The dose of PEG-G-CSF was 750 (mu) g / kg and the ratio - lipid: protein was 100: 1.

FIGURE 21 shows the effect of DOPG on the serum levels of PEG-G-CSF after pumping intraduodenal infusion. The dose of PEG-G-CSF was 750 (mu) g / kg and the lipid: protein ratio was 100: 1.

FIGURE 22 represents the fluorescence emission spectrum of MGDF in the presence and absence of vesicles-DMPG. The concentration of the MGDF was 0.1 mg / ml. The MGDF was E. coli derived from MGDF 1-163. The DMPG: MGDF ratio was 100: 1 (mol: mol).

FIGURE 23 shows the effect by increasing the -proportion lipid: protein on the fluorescence of the MGDF. The MGDF was E. coli derived from MGDF 1-163. The maximum emission wavelengths for mixtures of DMPG: MGDF with a pH of 5.0 (-o-) and with a pH of 7.0 (- -) are represented.

FIGURE 24 shows the Stern-Volmer traces of fluorescence suppression of MGDF by KI in the absence (o) and presence (•) of DMPG vesicles. The MGDF was E. coli derived from MGDF 1-163. The deletion experiments were performed by the addition of KI aliquots to the MGDF - (0.1 mg / ml) and to DMPG: MGDF (100: 1 molar).

FIGURE 25 shows the effect of the -temperature cycles on the DC of the MGDF (o) and of the DMPG: MGDF (100: 1 molar) (#). The MGDF was E. coli derived from MGDF 1-163. The% remaining helicity refers to the amount of CD detected at 10 ° C after each cycle (one cycle = samples that were rapidly heated to 95 ° C and cooled to 10 ° C).

FIGURE 26 shows the extent of denaturing of MGDF (+ DMPG) in the presence of various concentrations of urea. The MRE Circular Dichroism for MGDF (-o-) for DMPG: MGDF (- -) is represented, as well as the maximum fluorescence emission of the MGDF (-? J-) and DMPG: MGDF H§-). The MGDF was E. coli derived from MGDF 1-163. The DMPG: MGDF ratio was 100: 1 (mol: mol).

FIGURE 27 is a graphical representation of the maximum fluorescence e-mission for MGDF (+ DMPG) and para-PEG-MGDF (+ DMPG). MGDF was E. coli derived from MGDF 1-163, and PEG-MGDF was mono-pegylated E. coli derived from MGDF 1-163. The DMPG: MGDF and DMPG: PEG-MGDF ratio was -100: 1 (mol: mol).

FIGURE 28 shows the extent of the adsorption-from PEG-MGDF (+ DMPG) to glass bottles at various concentrations of PEG-MGDF. MGDF was E. coli derived from -MGDF 1-163, and PEG-MGDF was mono-pegylated E. coli derived from MGDF 1-163. % Recovery of PEG-MGDF (-O ") and DMPG: PEG-MGDF (-o-) was tested by counting the amount of radiolabelled MGDF recoverable from glass jars after an incubation of 18 hours at room temperature.

FIGURE 29 shows the DNA and amino acid sequence of the human MGDF (SEQ ID Nos: 1 and 2) including - a peptide signal (amino acids -21 to -1) and the mature amino acid sequence (1-332).

FIGURE 30 shows an example of reductive alkylation MGDF at a specific site of the alpha-amino group of the N-terminal residue using mono-methoxy-polyethylene glycol aldehydes to result in a substantially mono-pegylated product.

FIGURE 31 shows the in vivo activity of DMPG: MGDF and DMPG: PEG-MGDF in normal mice, in terms of total platelets. The MGDF was E. coli derived from MGDF 1-163, and the PEG-MGDF was mono-pegylated E. coli derived from MGDF 1-163. The doses of DMPG: MGDF and DMPG: PEG-MGDF were 100 (mu) g / kg and 300 (mu) g / kg, and the lipid: pro-tein ratio was 100: 1.

DETAILED DESCRIPTION OF THE INVENTION The compositions of the present invention are described in greater detail in the discussion that follows, and are illustrated by the examples provided below. The examples show various aspects of the invention, and include test results as to. the stability and biological activity of various protein: phospholipid compositions. Surprisingly, the interaction of the proteins with the lipid vesicle directly stabilizes the structure of the protein, thus exerting a stabilizing effect on the protein even under conditions which -conducts a denaturation of the protein in the absence- of the lipid.

Oral administration of a G-CSF: chemically modified phospholipid composition is also described herein, using G-CSF (as described above) to which polyethylene glycol molecules have been linked.

For use in the practice of the present invention, a variety of proteins capable of making -transition in the dissolved globular state are contemplated. The exemplary proteins contemplated are the cytosines, including several hematopoietic factors such as those mentioned above: G-CSF, GM-CSF, MGDF, M-CSF, interferons (alpha, beta and -game), interleukins (1-11). ), erotropoietin (EPO), fibroblast growth factor, cell-stem factor, nerve growth factor, BDNF, NT3, platelet-derived growth factor, and tumor growth factor ( alpha, beta). Other proteins - can be evaluated in terms of their ability to transition in the MGS. If the protein in question is capable of transitioning in the MGS, the protein in question can then be contacted with a negatively charged liposomal vesicle and in this way evaluate the effects of stabilization.

In general, G-CSF, useful in the practice of this invention, can be a native form isolated in pure form from mammalian organisms or, alternatively, as a product of synthetic chemical procedures or from the expression of hosts. eukaryotic or prokaryotic sequences of exogenous DNA that are obtained by cloning -genomic or cDNA, or by synthesis of the gene. Suitable procaryotic hosts include several bacterial cells (e.g., E. coli). Suitable eukaryotic hosts include yeast cells (e.g., S. cerevisiae) and mammalian cells (e.g., Chinese hamster ovary, monkey). Depending on the host employed, the G-CSF expression product can be -glycosylated with mammalian carbohydrates or other eukaryotic carbohydrates, or this can be non-glycosylated. The present invention contemplates the use of some and all of such forms of G-CSF, although recombinant G-CSF, especially the E. coli derivative, is preferred for reasons of greater commercial feasibility.

The G-CSF to be chemically modified for use - in the present invention can also be the human -natural G-CSF (nhG-CSF) or the product of a process of a recombinant nucleic acid, such as the expression of a cell - prokaryotic or eukaryotic host. In general, the contemplated chemical modification is the fixation of a chemical portion to the G-CSF molecule thereof. An article in a review which describes protein modification and protein fusion is Francis, Focus on Growth Factors 3: 4-10 (May 1992) (published by Mediscript, Mountview -Court, Friern Barnet Lane, London N20 OLD, United Kingdom). For example, see patent EP 0401 384, entitled: "Chemica-lly Modified Granulocyte Colony Stimulating Factor", which describes the materials and methods for preparing the G-CSF to which the polyethylene glycol molecules will be fixed. The binding can be by binding directly to the protein or to a portion which acts as a bridge for the active agent. The covalent bond is preferred as the most stable for fixation. Chemical modification may contribute to the controlled, continuous or prolonged effect of G-CSF. This modification may have the effect, for example, of controlling the amount of time that the modified G-CSF takes - chemically reaching the circulation. An example of a chemical modifier are polyethylene glycol compositions, including derivatives thereof.

Contemplated for use in the practice of this invention, are some chemically modified G-CSF preparations which allow for efficacy in administration. Efficacy can be determined by known methods, as a practitioner in art will recognize it. Pegylated G-CSF, especially pegylated E. coli derived from G-CSF, and more particularly, pegylated tri-tetra E. coli derived from G-CSF is preferred.

G-CSF has been reported to be the most stable under acidic conditions, despite the fact that in the pH range of 2.5 - 5.0, a conformational change occurs, which involves a detachment from the tertiary structure and a increase in the alpha helical content. Narhi et al., J. Protein Chem. 10, 359-367, (1991). This conformational change is characteristic of the dissolved globular state (MGS). Thus, as is the case for a formulator working with other proteins capable of transitioning in the MGS, a formulator working with G-CSF can protect against the thermally induced unfolding of the secondary and tertiary structure to prevent aggregation and denaturation.

The GM-CSF useful in the present invention can be a native form isolated in pure form from mammalian organisms, or a product of the expression of a prokaryotic or eukaryotic host of exogenous DNA sequences obtained by genomic cloning or cDNA cloning, or by-sis of the gene. Suitable prokaryotic hosts include several bacterial cells (e.g., E. coli). Suitable eukaryotic hosts include yeast cells (e.g., S. cerevisiae) and mammalian cells (e.g., Chinese hamster ovary, monkeys). Depending on the host employed, the product of GM-CSF expression can be glycosylated with mammalian carbohydrates or other eukaryotic carbohydrates. or this - it can be non-glycosylated. The present invention contemplates the use of any and all of such forms of GM-CSF, although recombinant GM-CSF, especially the E. coli derivative, is preferred for reasons of greater commercial feasibility.

The term "MGDF", as used herein, includes the naturally occurring MGDF, truncates of the naturally occurring MGDF as well as non-naturally occurring polypeptides which have an amino acid sequence and a sufficiently duplicative glycosylation of the MGDF found naturally to allow the possession of a biological activity-specifically stimulating growth, the development and / or production of megakaryocytes and / or platelets.

In a preferred embodiment, MGDF is the product of the expression of an exogenous DNA sequence, which has been transfected into a cell of a eukaryotic or prokaryotic host; that is, in a preferred embodiment, the MGDF is "recombinant MGDF". The preferred eukaryotic host is a mammal, CHO cells are particularly preferred, and the preferred prokaryotic host is a bacterium, particularly E. coli is preferred. Recombinant MGDF is advantageously produced according to the methods described herein, and in the publications cited herein with respect to the cloning and expression of MGDF.

Some additional preferred MGDF molecules have the following amino acid sequences, based on FIGURE 29 from here: MGDF 1-332 amino acids 1-332 of FIG. 29 MGDF 1-191 amino acids 1-191 of FIG. 29 MGDF 1-183 amino acids 1-183 of FIG. 29 MGDF 1-174 amino acids 1-174 of FIG. 29 MGDF 1-163 amino acids 1-163 of FIG. 29 MGDF 1-153 amino acids 1-153 of FIG. 29 MGDF 1-152 amino acids 1-152 of FIG. 29 MGDF 1-151 amino acids 1-151 of FIG. 29 MGDF 7-332 amino acids 7-332 of FIG. 29 MGDF 7-191 amino acids 7-191 of FIG. 29 MGDF 7-183 amino acids 7-183 of FIG. 29 MGDF 7-174 amino acids 7-174 of FIG. 29 MGDF 7-163 amino acids 7-163 of FIG. 29 MGDF 7-153 amino acids 7-153 of FIG. 29 MGDF 7-152 amino acids 7-152 of FIG. 29 MGDF 7-151 amino acids 7-151 of FIG. 29 In each of the above cases, Met-Lys can also be included in the N-terminal thereof.

Also contemplated for use in the present invention are several analogs of MGDF. As used herein, the phrase "MGDF analogue" refers to the MGDF with one or more changes in the amino acid sequence of the MGDF, which results in a change in the type (N- or 0-linked), number , or location of sites for carbohydrate fixation. The analogue (s) of the MGDF preserves at least the equivalent biological activity compared to that of the natural sequence MGDF (eg, human MGDF), and may possess a substantially higher activity, as measured in tests for the activity biological The resulting analogs may have less or more (preferably more) carbohydrate chains than the recombinant / human-made MGDF.

Also included within the analogs of this invention are analogs which have one or more amino acids deployed from the carboxy terminal end of the MGDF, wherein the carboxy terminal extension provides at least one additional carbohydrate site. The carboxy terminal of the MGDF will vary depending on the particular form of the MGDF used (e.g., amino acids MGDF 1-332, or amino acids MGDF 1-163). An additional carbohydrate site can be added to the terminal carboxy of an MGDF species by the addition of amino acids to the carboxy terminus; such amino acids-they contain one or more N- or 0-linked glycosylation sites.

The present invention also includes, in general, com positions of the chemically modified MGDF. In general, the contemplated chemical modification is a product of the MGDF in which said MGDF protein is linked to at least one molecule of polyethylene glycol (ie, pegylated MGDF). The pegylation of the MGDF can be carried out by any of the pegylation reactions known in the art. See, for example: Focus on Growth Factors 3 (2): 4-10 (1992); EP 0 154316; EP 0401 384; and the other publications cited here with respect to pegylation. Preferably, the pegylation is carried out via an acylation reaction or by an alkylation reaction with a reactive polyethylene glycol molecule (or a water soluble reactive analogue polymer).

PEGylation by acylation generally involves the reaction of an active ester derived from polyethylene glycol (PEG) with an MGDF protein. Any reactive PEG molecule, known or subsequently discovered, can be used to carry out PEGylation of the MGDF. An ester of the preferred activated PEG, the PEG esterified to N-hydroxysuccinimide ("NHS"). As used herein, "acylation" is contemplated to include, without limitation, the following types of linkages between the MGDF and a water soluble polymer, such as PEG: amide, carbamate, urethane, and the like. See Bioconjugate Chem. 5: 133-140 (1994). The reaction conditions can be selected from some known in the art of pegylation, or those developed subsequently, but those conditions, such as temperature, solvent, and pH, which can render the species inactive, should be avoided. MGDF to be modified.

PEGylation by acylation will generally result in a poly-pegylated MGDF product, wherein the epsilon-amino-amine groups are pegylated by an α-cyl linking group. Preferably, the link will be an amide. Also preferably, the resulting product will be substantially unique (e.g., greater than or equal to 95%), mono, di- or tri-pe-gilated. Thus, some species with a high degree of PEGylation (up to the maximum number of ether-amino acid groups of the MGDF plus an alpha-amino group in the amino-terminus of the MGDF) will normally be formed in amounts depending on the reaction conditions. specific used. If desired, the more purified pegylated species can be separated from the mixture, particularly unreacted species, by standard purification techniques, including, among others, dialysis, exterior salting, -tratrafiltration, ion exchange chromatography. , - gel filtration chromatography and electrophoresis.

PEGylation by alkylation generally involves the reaction of a terminal aldehyde derived from PEG with a protein such as MGDF in the presence of a reducing agent. PEGylation by alkylation can also result in poly-pegylated MGDF. In addition, one can manipulate the reaction conditions, as described herein, in favor of the fact that the pegylation is only carried out in the N-terminal alpha-amino group of the MGDF species (i.e., mono-pegylated species). An exemplary reductive alkylation reaction with MGDF to produce a monopegilated product is shown in FIGURE 30. In the case of a monopegylation or a poly-pegylation, the PEG groups are preferably attached to the protein via a -CH-NH group -. With particular reference to the group -CH_-, this type of link will be referred to herein as an "alkyl •" link.

Derivatization via reductive alkylation to produce a monopegilate product explores the differential reactivity of different types of primary amino groups (lysine versus N-terminal) valid for derivatization in the MGDF. The reaction is carried out with a pH (see below) from which one can take advantage of the differences in the pKa between the epsilon-amino groups of the lysine residues and the alpha-amino group of the N-terminal residue of the protein . By such selective derivatization, the attachment of a water-soluble polymer containing a reactive group such as an aldehyde, to a protein, is controlled: the conjugation with -the polymer takes place, predominantly, at the N-terminus of the protein, and there is no significant modification of other reactive groups, such as occurs in the amino groups of the lysine side chain.

Thus, in a preferred aspect, the present invention relates to pegylated MGDF, wherein the PEG group (s) is (are) fixed via acyl or alkyl groups. As discussed above, such products may be mono-pegylated or poly-pegylated (e.g., 2-6, preferably 2-5, PEG groups). The PEG groups are generally fixed to the protein in the alpha or epsilon amino groups of the amino acids, but it is also contemplated that the PEG groups can be fixed to any amino group attached to the protein, which is sufficiently reactive to reach be fixed to a PEG group under suitable reaction conditions.

The polymer molecules used in the acylation and alkylation processes can be selected from water-soluble polymers or a mixture thereof. The selected polymer should be soluble in water, since the protein to which it is fixed does not precipitate in an aqueous environment, such as a physiological environment. The selected polymer should be modified to have a simple reactive group, such as an active ester for acylation or an aldehyde for alkylation, preferably, so that the degree of polymerization can be controlled, as stipulated in the present methods. A preferred reactive PEG aldehyde is polyethylene glycol propionaldehyde, which is water stable, or mono C-C-alkoxy or aryloxy derivatives thereof (see U.S. Patent 5,252,714). The polymer - can be branched or unbranched. Preferably, for the therapeutic use of the preparation of the final product, the polymer will be pharmaceutically acceptable. The water soluble polymer can be selected from the group consisting of, for example, polyethylene glycol, monomethoxy polyethylene glycol, dextran, poly (N-vinyl pyrrolidone) polyethylene glycol, homopolymers of propylene glycol, an oxide co-polymer. of polypropylene / ethylene oxide, polyoxyethylated polyols (eg, glycerol) and polyvinyl alcohol. For the acylation reactions, the selected polymer (s) will have a simple reactive ester group. For the present reductive alkylation, the selected polymer (s) will have a reactive -simple aldehyde group. Generally, the water soluble polymer will not be selected from naturally occurring glycosyl residues, since these are usually more conveniently manufactured by recombinant expression systems in mammals. The polymer can be of any molecular weight, and can be branched or unbranched.

A particularly preferred water-soluble polymer, for use herein, is polyethylene glycol, abbreviated as PEG. As used herein, polyethylene glycol encompasses any of the forms of PEG that have been used to derivatize or-after proteins, such as mono- (C -C) alkoxy- or aryloxy-polyethylene glycol. The methods for preparing the pegylated MGDF will generally comprise the following steps: (a) the reaction of an MGDF polypeptide with polyethylene glycol (such as a reactive ester or an aldehyde derivative of PEG) under conditions whereby the MGDF reaches - be fixed to one or more PEG groups, and (b) obtaining the reaction product (s). In general, the optimal reaction conditions for the acylation reactions will be determined on a case-by-case basis, based on the known parameters and the desired result. For example, the highest proportion of PEG: protein, the highest percentage of poly-pegylated product.

Another important consideration is the molecular weight of the polymer. In general, the polymer's largest molecular weight, the smallest number of polymer molecules which can be attached to the protein. Similarly, the derivation of the polymer will be taken into account when these parameters are perfected. Generally, the largest molecular weight (or the most branches) and the highest polymer: protein ratio. In general, for the pegylation reactions contemplated herein, the preferred average molecular weight is from about 2 kDa to about 100 kDa (the terms "about" and "near" indicate + 1 kDa). The preferred average molecular weight is from about 5 kDa to about 50 kDa, particularly an average molecular weight of from about 12 kDa to about 25 kDa, and more preferably 20 kDa is preferred. The ratio of water-soluble polymer to MGDF protein will generally be in the range of 1: 1 to 100: 1, preferably (for polypegilation) 1: 1 to 20: 1, and (for monopegylation) 1: 1 to 5 :1.

• Using the conditions indicated above, the reductive alkylation will provide selective binding of the polymer to any MGDF protein having an alpha-amino group at the amino terminus, and maintain a substantially homogeneous preparation of the monopolymer / MGDF protein conjugate. The term "monopolymer / MGDF protein conjugate" is used herein to refer to a composition comprising a single polymer molecule attached to a molecule of an MGDF protein. The monopolymer / protein MGDF conjugate, preferably, will have a polymer molecule located at the N-terminus, but not on amino side groups of lysine. The preparation will preferably be greater than 90% of the monopolymer / protein MGDF conjugate, and more preferably greater than 95% of the monopolymer / protein MGDF conjugate, with the remaining observable molecules that remained unreacted (ie, protein that - it lacks the polymer portion). The following examples provide a preparation which is at least about -90% of the monopolymer / protein conjugate, and about 10% of unreacted protein. The monopolymer / protein conjugate -poses biological activity.

Lipid vesicles useful in the compositions of the present invention are those negatively charged liposomes capable of interacting with the protein in question. Liposomes particularly contemplated for use include: t dioleoylphosphatidylglycerol (DOPG), dimyristoylphosphatidylglycerol (DMPG), dipalmitoyl phosphatidyl glycerol (DPPG), egfosfatidilglicerol, dioleoylphosphatidylethanolamine (DOPE), egfosfatidiletanolamina, dioleoilfofatidico acid (DOPA), dimiristoilfosfatidico acid (DMPA), acid dipalmitoylphosphatidic ( DPPA), dioleoylphosphatidylserine (DOPS), dimyristoylphosphatidylserine (DMPS), dipalmitoylphosphatidylserine (DPPS), egphosphatidylserine, lysophosphatidylglycerol, lysophosphatidylethanolamine, lysophosphatidylserine. Depending on the particular liposome used, the amount of liposome will vary.

The protein: phospholipid compositions preferably include a regulatory agent to maintain the pH of the solution within a desired range. Preferred agents include sodium acetate, sodium phosphate, and sodium citrate. Mixtures of these regulatory agents can also be used. The amount of regulatory agent useful in the composition depends largely on the particular regulator used, and the pH of the solution. For example, acetate is a more efficient regulator at a pH of 5 than at a pH of 6, so less acetate can be used in a solution at a pH of 5 than at a pH of 6. The preferred pH range - for the compositions of the present invention is pH 3.0 -7.5.

The compositions of the present invention may also include an isotonic adjusting agent to give an isotonic and more compatible solution for an injection. The most preferred agent is sodium chloride within a concentration range of 0-150 mM.

Also encompassed by the invention are those pharmaceutical compositions which consist of effective amounts of polypeptide products of the invention, together with pharmaceutically acceptable diluents, preservatives, solubilizers, emulsifiers, auxiliaries and / or -carriers. Such compositions will influence the physical state, stability, and bioavailability of the protein. See, e.g., Remingtons Pharmaceutical Sciences, 18th Edition, 1435-1712 (Mack Publishing Co., Easton, PA., 1990) which is incorporated herein by reference. What constitutes an effective amount of the protein in a particular case will depend on a variety of factors which the informed practitioner will take into account, including the desired therapeutic outcome, the severity of the condition or disease to be treated, the condition physical of the subject, and so on.

In a preferred embodiment involving E. coli derived from rhG-CSF, the liposomal vesicle used is -DOPG with a 50: 1 ratio of DOPG: G-CSF, with a pH 4.5, containing 10 mM sodium acetate.

In a preferred embodiment involving E. coli derived from rhGM-CSF, the liposomal vesicle used is -DMPG with a 17: 1 ratio of DMPG: GM-CSF, at pH 7.0, in regulated salt phosphate (PBS).

In a preferred embodiment involving E. coli derived from reG-CSF which has been chemically modified (pegylated), rhG-CSF is tri-tetra pegylated, the liposomal membrane used is DMPG with a 17: 1 ratio of DMPG : PEG-G-CSF, with a pH 4.5.

In a preferred embodiment involving the 'E. coli derived from MGDF 1-163, the liposomal vesicle used is DMPG with a 100: 1 ratio of DMPG: MGDF, with a pH of 5.0, in lOmM sodium acetate and 5% sorbitol .

In a preferred embodiment involving E. coli derived from MGDF 1-163 which has been chemically modified (pegylated), the MGDF is mono-pegylated (20 kDa) via reductive alkylation, the liposomal vesicle used is DMPG with a ratio 100 : 1 of DMPG: MGDF, with a pH 5.0, in 10 mM sodium a-cetate and 5% sorbitol.

In a preferred embodiment involving CHO derived from MGDF 1-332, the liposomal vesicle used is DMPG with a 100: 1 ratio of DMPG: MGDF, at pH 5.0, in 10 mM sodium acetate and 5% sorbitol.

Although the invention has been described and illustrated with respect to specific protein: lipid compositions and methods of treatment, it will be apparent to one of ordinary skill in the art, that a variety of related compositions, and methods of treatment can exist without departing from the field of art. application of the invention.

The following examples will illustrate in greater detail the various aspects of the present invention.

EXAMPLE 1 - Initial experiments were conducted to examine the possibility of incorporating recombinant human G-CSF (rhG-CSF) into a lipid vesicle. RhG-CSF was produced using recombinant DNA technology in which the E. coli cells were transfected with a DNA sequence encoding human G-CSF, as described in U.S. Pat. No. 4,810,643 by Souza. RhG-CSF was prepared as a 4 mg / ml solution in dilute HCl, pH 4.0. All lipids were obtained by means of Avanti Polar Lipids (Alabaster, Ala) and stored at -20 ° C under nitrogen at a final concentration of 100 mg / ml in chloroform.

Preparation of the G-CSF Complexes: Phospholipid To prepare the lipid vesicles for combination with G-CSF, 30 (mu) mol of the appropriate lipid was distributed in a glass tube and dried to a thin film using a stream of nitrogen gas. The lipid films were dried for at least two hours under a vacuum to remove any trace of chloroform.The lipid films were hydrated in 1 ml of any of the following: deionized distilled water (ddHp0), regulated phosphate salt, pH 7.2 (Gibco / BRL "D-PBS") or 150 mM NaCl. The samples were then sonicated in a bath type sonicator (Laboratory Supplies, Hicksville, N. Y.). The sonication was continued until the samples were optically clear (usually between 10-15 minutes). Samples were stored at 4 ° C under nitrogen until use. The final concentration of the lipid was 30 mM. Alternatively, the lipid vesicles were prepared by taking 300 (mu) mol of lipid and drying under nitrogen and drying as described above. The dried lipid films were -hydrated in 10 ml of an appropriate aqueous solution as described above. The samples were then microfluidised in an emulsifier with reference scale (Microfluidics Model 110S, Microfluidics, Inc. Cambridge, MA) operating at -10,000 psi. Samples were recycled through ins-truniento for 10 cycles. The microfluidized samples were then stored at 4 ° C as described above.

The G-CSF: phospholipid complexes were prepared by mixing the G-CSF (as described above) with a particular lipid (as described above). The mixing was accompanied by vortexing, stirring, or gentle shaking. Different molar proportions of lipid: G-CSF were prepared to evaluate membrane insertion and stabilization of the protein. For example, to prepare a 3 ml sample (in water) which is 0.2 mg / ml of G-CSF at a molar ratio of 40: 1 lipid: G-CSF, 150 (mu) l of G- are combined Normal CSF with 44 (mu) l of lipid (30 mM normal, prepared in water by sonication) and water is added to achieve a final sample volume of 3 ml. About five minutes of incubation is recommended (but not necessarily), and this incubation was used here before using or -testing the sample.

G-CSF can also be combined with the hydrated lipid before microfluidization. Subsequent microfluidization of the mixtures, as described above, leads to the incorporation of G-CSF into the membrane of the lipid.

Analysis of the G-CSF Complexes: Phospholipid 1. Spectrum of emission of tryptophan.

There are two residues of tryptophan in the rhG-CSF that are completely sensitive to local environmental conditions. Therefore, the analysis was performed to determine the fluorescence of the tryptophan rhG-CSF when the rhG-CSF is contacted with a liposome. A blue change in the maximum fluorescence emission will suggest that the tryptophanes are in a more hydrophobic environment and therefore, the rhG-CSF was fixed in the lipid membranes. An excellent review of the fluorescence analysis of tryptophan is Principies of Fluorescence Microscopy, by J. Lakowicz, Chapter 11 (Plenum Press, New York, 1983).

The fluorescence of the tryptophan from the G-CSF: lipid complexes (as described above) was tested by exciting the samples at 280 nm and scanning while the emission from 285 nm to 420 nm in 1 nm increments at an index of 1 nm / sec . The sample volume was 3 ml, and the final concentration of G-CSF was 0.2 mg / ml for all samples. The lipid: G-CSF ratios were varied. All fluorescence measurements were made using an Alphascan PTI fluorometer (South Brunswick, NJ). All measurements were made at 25 ° C and this temperature was maintained during the use of a covered water support connected to a circulating water bath. The emission spectra were collected and analyzed using the software data provided by PTI.

The fluorescence spectrum of rhG-CSF in the presence and absence of small unilamellar vesicles composed of -DOPG, is shown in FIGURE 1. The rhG-CSF has a maximum emission at 334 nm in the absence of DOPG vesicles. In the presence of DOPG at a 100: 1 ratio of lipid: protein, the fluorescence of the tryptophan rhG-CSF shows an a-zul change in the maximum emission of fluorescence at 327 nm, and a dramatic increase in the intensity of the fluorescence. The low wavelength in the emission of fluorescence in the presence of DOPG, suggests that tryptophans are found in a more hydrophobic environment than the native protein. As shown in FIGURE 2, the changes in fluoride depend on the molar ratio of DOPG: G-CSF, and the insertion in the membrane is detectable, since a 10: 1 ratio of D0PG: G-CSF is reached . 2. Iodide attenuation experiments.

Iodide is an efficient collisional attenuator of tryptophan fluorescence, but can not penetrate the lipid membranes. Therefore, the efficient attenuation of tryptophan fluorescence by iodide indicates the exposure of the residues to the volume of the aqueous solvent, while the protection of the attenuating iodide occurs when the tryptophan of the protein is sequestered out of the aqueous solvent. In these experiments, the G-CSF and a DOPG.sub.G-CSF composition (100% ratio of lipid-protein) were used.

After the initial readings (F) were taken on or the samples and at their vex these readings were recorded, the "fluorescence intensity" was measured after the addition of increasing amounts of potassium iodide (KI) (normal 5M) Both the sample and the KI solutions were prepared to contain 1 mM Na? S0 (final concentration) as described by Lee et al., Biochem Biophys. Acta, 984¡ 174-182 (1989) and Le Doan et al. Biochem Biophys, Acta, 858: 1-5 (1986) The addition of Na "S0" prevents the formation of Ip, which can be divided into non-polar regions of proteins and membranes. the Stern-Volmer equation (F / F = 1 + Kvt (K?)), where F olo and F are the fluorescence intensities of the samples in the absence and presence, respectively, of KI at a concentration (KI) K is the attenuating constant of Stern-Volmer l for the mitigating KI of the tryptophan residues of G-CSF; Hrer, S., Biochemistry 10: 3254-3263 (1979).

The Stern-Volmer charts of the data are shown in FIGURE 3. In the absence of DOPG vesicles, the fluorescence of rhG-CSF is efficiently attenuated by the KI. In the presence of DOPG, the Stern-Volmer plot of the data is linear, indicating that iodide has poor access to both -tryptophanes. The data show that the tryptophan residue, to which the iodide is accessible in the absence of DOPG, becomes inaccessible to iodide in the presence of DOPG. Therefore, the rhG-CSF portion containing this tryptophan can be fixed in the DOPG bilayer. 3. Energy transfer measurements.

As shown previously, energy transfer can occur between tryptophan donors and lipid-soluble fluorescent acceptors, such as decanoic pyrene, since the excitation spectrum of this test significantly superimposes the emission spectrum of tryptophan. Friere et al., Biochemistry 22: 1675-1680 (1983). If the protein is inserted into the lipid membranes, energy is transferred from the tryptophan to the pyrene, leading to an attenuation in the fluorescence of the tryptophan. In this experiment, the intensity of tryptophan emission from various lipid: G-CSF complexes was recorded before (F) and after (F) the addition of various amounts of decane-co pyrene acid (30 (mu) g / ml normal in tetrahydrofuran). The samples were continuously stirred during the addition of the decanoic pyrene acid, to cause mixing between the decanoic pyrene acid and the sample. The ratio of F / F is proportional to the amount of energy present that is transferred between the tryptophanes of the G-CSF and the "hydrophobic" energy acceptor of the decanoic pyrene acid.

FIGURE 4 shows the attenuation profile for rhG-CSF in the presence of DOPG (100: 1 ratio of lipid: pro-thein) as a function of the addition of pyrene-decanoic acid. The attenuation occurs at very low concentrations of decanoic pyrene acid (less than 1 mol%), so that the effect of the fluorescent test on the structure of the membrane and its operation is minimal. Since decanoic pyrene acid can be rapidly divided into the lipid bilayers, the present data indicates that rhG-CSF is fixed in the DOPG membranes sufficiently deep to allow efficient transfer of energy from tryptophan to the pyrene acceptor. The energy transfer was confirmed by examining the excitation spectrum of decanoic pyrene acid-labeled DOPG vesicles in the presence and absence of rhG-CSF.

The above analysis shows that rhG-CSF can interact closely with an unsaturated phospholipid such as DOPG. In the presence of DOPG vesicles, a tryptophan from rhG-CSF is protected from a water-soluble fluoride attenuator, but is susceptible to attenuation via energy transfer for a hydrophobic fluorescent assay. The data shows that rhG-CSF can be inserted into the composite DOPG membranes. The insertion in the membrane is detectable, since a 10: 1 ratio (lipid: G-CSF) is reached, and this number can represent the number of -lips that surround the inserted portion of the protein.

EXAMPLE 2 - In this example, the ability of rhG-CSF to interact with other phospholipids was determined, using comparisons of the F / F intensity and emission maxima as described above. In each case, the molar ratio of lipid: rhG-CSF was 100: 1.

FIGURE 5 shows the F / F data for the rhG-CSF or in the absence and presence of various lipids. FIGURE 6 - shows the maximum emission data for the same compositions. The data in FIGURE 5 and FIGURE 6 demonstrate that, in addition to DOPG, rhG-CSF can be inserted into DMPG, DPPG, and less efficiently, into phosphatidylethanol amines (PE's) and phosphatidylserines (PS's) . In addition, NG-DOPE (shows DOPE where the PE head group became more negative) was found to provide an improved insertion of rhG-CSF than DOPE.

DOPC, DMPC and DPPC are neutral lipids, and these vesicles have a small effect (if they do not) on the maximum emission or on the fluorescence intensity of rhG-CSF, indicating that the non-interaction takes place with these phospholipids (see FIGURES 5 and 6, and FIGURE 7, curve 2).

The above data show that a protein, able to make a transition in the dissolved globular state, can be inserted in several. lipid vesicles. So, this rhG-CSF-lipid interaction only occurs when a negatively charged lipid particle is used. Among negatively charged lipid vesicles, those vesicles with the highest negative charge appear to provide stronger rhG-CSF-lipid interactions.

EXAMPLE 3 In this example, the effect of the DOPG interaction: rhG-CSF was determined and how it relates to the stability of the protein. The measurements of Circular Dichroism (CD) were made on a Jasco J-720 instrument equipped with a Peltier-type thermostat cell holder and a magnetic stirrer. The circular dichroism at 222 nm was measured using a final concentration of rhG-CSF of 80 (mu) g / ml, pH 6.0. The differential scanning calorimetry measurements were made using a Microcal MC-2 calorimeter. The samples of rhG-CSF (1 mg / ml, in water) or DOPG: rhG-CSF (45: 1 mol / mol, in water), were explored at an index of 90 ° C / hour.

The data was archived and analyzed using the software - from Microcal.

The changes induced by the temperature in the alpha heli-city of the G-CSF can be followed by the measurement of the circular dichroism (222 nm) as a function of the increase in temperatures. The thermally induced cleavage of rhG-CSF at pH 6.0 is shown in FIGURE 8. The curve indicates that a clearly marked transition occurs at approx. 60-70 ° C, which leads to a loss of alpha heli-city. After this transition, rhG-CSF precipitates irreversibly from the solution. The temperature range of the splitting is similar to the melting temperature of rhG-CSF at pH 7.0, as determined by differential scanning calori-metry as shown in FIGU-RA 9.

In contrast, the DOPG¡rhG-CSF samples show a gradual loss of alpha helicity with an increase in temperature and, unlike the rhG-CSF alone, the temperature-induced splitting of the DOPG: rhG-CSF does not appear to be cooperative ( see FIGURE 8). This conclusion is also demonstrated by the lack of a fusion transition, as shown by differential scanning calorimetry (FIGU-RA 9). Notably, DOPG: rhG-CSF samples can recover alpha helicity after heating to 95 ° C, and can be cyclized repeatedly between 95 ° C and 10 ° C with total recovery of helicity during cooling (see FIGURE 10) . The rhG-CSF alone, under these conditions, is irreversibly broken and precipitated from the solution.

The effects of DMPG and DPPG on the circular dichroism of G-CSF were also examined. A 150: 1 ratio of lipid: rhG-CSF was used and, as was the case with DOPG, DMPG and DPPG also stabilize the secondary structure of rhG-CSF (FIGURES 11-13).URI These data demonstrate that the interaction of rhG-CSF with DOPG, with DMPG and with DPPG increases the stability of the protein under conditions where rhG-CSF is only unstable. The interaction directly stabilizes the secondary and tertiary structure of rhG-CSF.

EXAMPLE 4 In this example, the interaction effect of rhG-CSF: DOPG was determined and how this interacts with the biological activity of rhG-CSF. The in vitro activity of -3 rhG-CSF was tested using the G-CSF dependent on (H) -thymidine by me bone marrow cells, as described in Zsebo et al., Immunobiology 172: 175-184 (1986).

All activity tests were performed in triplicate.

The in vivo activity was determined by a subcutaneous injection to hamsters (dose of 100 (mu) g / kg of rhG-CSF), and the measurements of the white blood cell count (WBC). 1. In vitro activity A. The specific activity of rhG-CSF in the absence and presence of DOPG was determined. The heat treated rhG-CSF and the DOPG: rhG-CSF samples were also tested. The results are summarized in TABLE 1.

TABLE 1 Sample Specific Activity (U / mg / protein) rhG-CSF 0.66 + 0.09 rhG-CSF (heated) Not detectable DOPG: rhG-CSFb 0.61 + 0.11 D0PG: rhG-CSFb (heated) to 0.52 + 0.08 The sample was incubated for 10 minutes at 85 ° C in a water bath before performing the test. Proportion D0PG: rhG-CSF of 50: 1 (mol / mol).

As shown in TABLE 1, insertion in the DOPG bilayers does not adversely affect the biological activity of rhG-CSF. After heating at 85 ° C for 10 minutes, the rhG-CSF possesses an undetectable activity, and the protein -precipitate. After a similar treatment, the complex -D0PG: rhG-CSF retains approx. 85% of the unheated rhG-CSF activity, and the secondary structure recovers completely dg cooling.

B. The ability of various lípi dos to stabilize rhG-CSF dg freeze-drying was also studied. The samples of rhG-CSF, in combination with various lipids, were freeze-dried and tested (as described above) for activity. The DOPG, the DMPG and the DPPG, when mixed with the rhG-CSF, allowed approx. 100% retention of the bioactivity of rhG-CSF after freeze drying (FIGURE 14). The rhg-CSF alone does not survive the freeze drying process. 2. In vivo activity The activity (WBC count) of the rhG-CSF in the absence and presence of the lipid was determined. The activity was measured after a subcutaneous injection (dose of 100 (mu) g / kg of rhG-CSF) on day 0. Five different lipid-complexes: rhG-CSF were tested, and in each case, the lipid complex : rhG-CSF maintained the activity in vivo (FIGURES 15 and 16).

Previous studies demonstrate that the insertion -in negatively charged lipid bilayers- does not adversely affect the biological activity of rhG-CSF. Additionally, it appears that the lipid protective effect protects the rhG-CSF dg the freeze drying process.

EXAMPLE 5 In this example, the chemically modified G-CSF (pegylated G-CSF (PEG-G-CSF)) and the G-CSF obtained as a product of the expression of a eukaryotic host cell (CHO-G-CSF) were tested for their ability to interact with negatively charged lipid vesicles. For the CHO-G -CSF, the determinations were made using comparisons of the F / F intensity and the emission maxima (as described in Example 1 above). In each case, the molar ratio of lipid: protein was 100: 1. For PEG-G-CSF, the determination was based on an analysis of circular dichroism.

The CHO-G-CSF used, was produced using recombinant DNA technology, in which the cells of the Chinese Hamster Ovary (CHO) were transfected with a sequence-DNA encoding with human G-CSF, as described in the Patent. US No. 4,810,643 by Souza. The CHO-G-CSF was pre-stopped as a 0.6 mg / ml solution in PBS, pH 7.0. He CHO-G-CSF showed an interaction with DOPG in a manner similar to rhG-CSF, with each sample showing an increase in fluorescence intensity in the presence of DOPG, as well as a blue change in maximum emission in the presence - of DOPG (FIGURES 17 and 18). Therefore, the interaction of the DOPG is not due to any peculiarity of the recommandation form of the G-CSF.

The PEG-G-CSF used in these experiments was tri-tetra pegylated E. coli derived from G-CSF (using PEG 6000). The DMPG: PEG-G-CSF samples (17: 1 mol / mol) were prepared using procedures described above. The -DMPG: PEG-G-CSF samples were characterized by completely recovering the secondary structure after heating (FIGURE 19). Despite the presence of PEG molecules, the derivatized protein was able to interact with the lipid in the same way as the native protein.

The above data show that the stabilizing effects associated with the interaction of G-CSF with a negatively charged lipid vesicle are not unique only to the rhG-CSF obtained as a product of the expression of a prokaryotic host cell. A chemically modified protein capable of transitioning in an MGS and contacted with a liposomal vesicle, here PEG-G-CSF: DMPG, also has stabilizing effects.

EXAMPLE 6 In this example, the effects of DMPG and DPPG on GM-CSF were studied. GM-CSF was recombinant human GM-CSF, as described in U.S. Patent No. 5,047,504 by Boone, and prepared as a 1 mg / ml solution in buffered saline phosphate (PBS), pH 7.0. A pro-lipid: GM-CSF ratio of 17: 1 was used, and the thermal stability was measured using a circular dichroism analysis as described above. The DMPG and the DPPG can lead to a better thermal stability of GM-CSF, ie recovery of the secondary structure after heating (FIGURES 20a &20b).

These data provide another example of a protein, capable of transitioning in the dissolved globular state, by interacting with a negatively charged lipid vesicle to provide better thermal stability to the protein.

EXAMPLE 7 In this example, a DOPG: PEG-G-CSF complex was used to evaluate the possibility of increasing the therapeutic response of G-CSF after enteral administration. For this experiment, the DOPG was prepared as described in EXAMPLE 1, and the PEG-G-CSF was prepared as described in EXAMPLE 5. 100 (mu) mol of lipid (797 (mu) l) were dried under a vacuum and then 1 ml of water-ml is added to obtain a 100 mM solution of the lipid. This solution was sonicated for 5 minutes in a -soning water bath (Model G 112SP1T from Lab. Supply Inc., Hicksville, NY) or until the lipid solution was clear. 9 (mu) mol of the DOPG solution (90 (mu) 1) is added to 90 nmol of native rhG-CSF or PEG-G-CSF in 1 mM HCl. The solution was vortexed and stopped in a final volume of -2 ml with 1 mM HCl. For intraduodenal administration in rats, the material was placed in an osmotic pump, which was implanted in the animal. The release of the material occurs for more than 24 hours.

The results of the analysis for the total WBC for the animals that received both the rhG-CSF and the PEG-G-CSF with and without lipid are shown in FIGURE 21. FIGURE 21a shows that the infusion of the native G-CSF fails to stimulate a WBC response, compared to the control vehicle. The addition has a small impact on the therapeutic response of the animals to the rhG-CSF.

The response of the rats for the pegylated G-CSF is shown in FIGURE 21b. One can see that the PEG-G-CSF has only stimulated a WBC response. The elevation of the WBC is maintained for 48 hours before returning to the baseline. The PEG-G-CSF formulated with DOPG also stimulates a WBC response, and this response is almost twice as large as for the PEG-G-CSF alone. These results are confirmed by serum levels measured from PEG-G-CSF after infusion (FIGURE 22).

These data demonstrate that by including an anionic lipid, such as DOPG, in an oral formulation of PEG-G-CSF, the therapeutic response elicited by the derivatized protein appears to be increased. The mechanism involved is not currently understood.

EXAMPLE 8 In this example, the effects of DMPG on MGDF were studied. The MGDF was recombinant human E. coli derived from MGDF 1-163, and prepared as a 1.0 mg / ml solution in 10 mM sodium acetate, 5% sorbitol, at pH 5.0. The DMPG: MGDF complexes were prepared as described in EXAMPLE 1.

Analysis of the DMPG Complexes: MGDF 1. Spectrum of emission of tryptophan.

There is a tryptophan residue in the MGDF (position 51) that was used as the monitor in the interaction of the MGDF with DMPG vesicles. The fluorescence of the tryptophan from the -complexes DMPG: MGDF was tested as described in EXAMPLE 1, using a MGDF concentration of 0.1 mg / ml. The fluorescence spectrometer of the MGDF in the presence and absence of small unilamellar vesicles composed of DPMG is shown in FIGURE 22. The MGDF has a maximum emission at 336 nm in the absence of DMPG vesicles. In the presence of DMPG at a 100: 1 ratio of lipid: protein, the fluorescence of the tryptophan in the MGDF shows a blue change in the maximum fluorescence e-mission at 328 nm. The low length of the fluorescence emission in the presence of DMPG suggests that tryptophans are found in a more hydrophobic environment than the native protein. And, as shown in FIGURE 23, the changes in fluorescence depend on the molar ratio of DMPG: MGDF, with the insertion in the detectable membrane, since an 8-30: 1 ratio of DMPG: MGDF is reached. The fluorescent change is maximum at molar ratios greater than or equal to 100: 1, and the MGDF has a seemingly higher affinity for DMPG vesicles when the pH is lowered from pH 7.0 to pH 5.0. This suggests that the trituration of certain amino acids (e.g., histidine) can be used to increase or attenuate the interaction. 2. Iodide attenuation experiments.

In these experiments, the MGDF and a DMPG: MGDF composition (100: 1 DMPG: MGDF) were used for the iodide attenuation experiments, as described in -EEMPLE 1. The Stern-Volmer charts of the data, are shown in FIGURE 24. In the absence of DMPG vesicles, the fluorescence of the MGDF is efficiently attenuated by KI. In contrast, in the presence of DMPG, tryptophan is inaccessible to iodide, indicating that the portion of the MGDF containing -this tryptophan, can be fixed in the DMPG bilayer.

The above analysis shows that, when it was the case with G-CSF and GM-CSF, the MGDF can interact closely-with an unsaturated phospholipid, such as DMPG. In the presence of DMPG vesicles, a tryptophan from the MGDF is protected from a water-soluble fluorescence attenuator. The data shows that the MGDF can be inserted into the composite membranes of the DMPG. The insertion in the membrane is detectable, since an 8: 1 ratio (DMPG: MGDF) is reached, and this number can represent the number of lipids surrounding the inserted portion of the protein.

EXAMPLE 9 In this example, the effect of the DMPG: MGDF interaction was determined and how it relates to the stability of the protein. The thermal stability, the stability in the presence of urea, and the shelf life stability of the MGDF (+ DMPG) were evaluated. In each of the studies, a 100: 1 molar ratio of DMPG was used: MGDF. 1. Thermal stability The circular dichroism (CD at 222 nm) of the MGDF alone or inserted into DMPG vesicles, was monitored as a function of the thermal cycles between 95 ° C and 10 ° C, as described in EXAMPLE 3, FIGURE 10. The % of the remaining CD refers to the amount of CD detected (at 10 ° C) after each cycle (one cycle is 10 ° C - 95 ° C - 10 ° C) when compared to the CD of an unheated sample of the indicated composition. While the MGDF loses more than 70% of its helicity after 3 heating cycles, the DMPG: MGDF completely retains its original alpha helicity under the same conditions (see FIGURE 25). 2. Stability in the presence of Urea.

Urea is a chaotropic reagent which can bend and denature proteins. The de-naturalization equilibrium of the MGDF (+ DMPG) was monitored by fluorescence, that is, measurement of the tertiary structure, and by circular dichroism, that is, a measure of the secondary structure. As shown in FIGURE 26, when the tertiary structure of the protein is lost, the tryptophan residues are more exposed to the water phase and the emission of the wavelength of the MGDF is changed to longer-range wavelengths.; and, when the secondary structure is lost, the average residual ellipse (MRE) becomes - less negative when the alpha helicity is lost. In absence of DMPG, a 50% loss in the tertiary structure occurs with urea approx. 3M, while 8M urea is re-wanted to achieve a 50% loss of structure in the presence of DMPG. Similarly, 50% loss of MRE-from MGDF requires 7M urea in the absence of DMPG, compared to 9M ures in the presence of DMPG. 3. Stability during shelf life.

E. coli MGDF 1-163 (+ DMPG) was stored under the conditions indicated in TABLE 2 below, and then examined by size exclusion chromatography (SEC) using a Toso-Haas G3000SWXL column with a mobile phase of regulator 100 mM phosphate, 10% ethanol, 0.2 Tween-20, pH 6.9. The samples were diluted with ethanol and Tween-20 at the same concentration used in the mobile phase, and 10-20 (mu) g of sample were injected per run. The temperature of the column was maintained at 40 ° C. Any aggregate of -MGDF forms eluents earlier than the non-aggregated protein, and is quantified by measuring the area under the -value of the tip of the aggregate and the tip of the monomer. The data refer to the% of the total MGDF at the tip of the aggregate.

TABLE 2 % aggregation when measured by SEC Sample Temperature 5 wk. 11 sem.

MGDF -80 ° C 0 0.2 4 ° C 0.4 0.6 37 ° C 18 39.2 DMPG: MGDF -80 ° C 0 0.14 4 ° C 0 0 37 ° C 1.9 2.5 As shown in TABLE 2, DMPG dramatically reduces aggregate formation during storage. In this way, the DMPG can be used to increase the shelf life of the MGDF.

These data demonstrate that the interaction of MGDF -with DMPG vesicles- increases the stability of the protein under conditions where the MGDF is only unstable. The interaction directly stabilizes the secondary and -terciary structure of the MGDF in the presence of denaturants such as urea, and significantly improves the shelf life of the MGDF at various temperatures.

EXAMPLE 10 In this example, chemically modified MGDF 1-163 (mono-pegylated (20 kDa) MGDF 1-163 (PEG-MGDF)) was tested for its ability to interact with negatively charged lipid vesicles. For the PEG-MGDF, the determinations were made using comparisons of the intensity F / F or emission maxima (as described in EXAMPLES 1 and 8 above). In each case, the molar ratio of lipid: protein was 100: 1.

The PEG-MGDF used in these experiments was mono-pegylated E. coli (20 kDa) derived from MGDF 1-163 using mono-methoxy-polyethylene glycol aldehyde (MePEG) (molecular weight - average: 20 kDa) via reductive alkylation . The homogeneity of the PEG-MGDF conjugates was determined by Sodium Gel Electrophoresis Dodecyl Sulfate Polyacrylamide using 4-20% prefabricated gradient gels (NOVEX). A larger band, corresponding to the position of a 46.9 kDa protein, was revealed.

The DMPG: PEG-MGDF samples (100: 1 mol / mol) were then prepared using procedures described above. The DMPG: PEG-MGDF samples were found to completely recover the secondary structure after heating (FIGURE 27). Despite the presence of the PEG molecules, the derivatized protein was able to interact with the lipid in the same sense as the native protein does.

These data show that the changes in the maximum emission associated with the interaction of the MGDF with a negatively charged lipid vesicle, are not unique only for the MGDF that is obtained as a product of the expression of a prokaryotic host cell. As demonstrated with chemically modified rhG-CSF, the DMPG: PEG-MGDF complex also exhibits changes in emission, and the data show that chemically modified MGDF can be inserted into composite DMPG membranes.

EXAMPLE 11 In this example, the effect of the DMPG: PEG-MGDF interaction was evaluated and how it relates to the problem of adsorption of MGDF to glass jars. About 125 11 pg / ml of (I) -PEG-MGDF were combined with various -concentrations of unlabeled PEG-MGDF, to achieve the indicated final concentrations of PEG-MGDF (see FIGURE 28). Where indicated, the DMPG was included in the -dilution (see also FIGURE 27). 1 ml of the preparations was placed in 3-ce glass jars (Kimble). The recovery% of the PEG-MGDF was tested by counting the amount of radiolabelled MGDF recoverable from the glass jars after an incubation of 18 hours at room temperature. As shown in FIGURE 28, the PEG-MGDF is immediately adsorbed to glass containers - when the concentration of the protein is lowered, and adsorption is especially high in the range of 0.1-50 (mu) g / ml. In contrast, the DMPG: PEG-MGDF samples show almost no non-adsorption to the glass in the range of 0.1-50 (mu) g / ml PEG-MGDF.

EXAMPLE 12 In this example, the effect of the DMPG interaction was determined: MGDF 1-163 and DMPG: PEG-MGDF 1-163 and how it relates to the biological activity of MGDF 1-163 and PEG-MGDF 1-163 . MGDF 1-163 was E. coli derivative, and the - lipid: protein ratio was 100: 1. Platelet calculations from mice treated with 100 (mu) g / kg and 300 (mu) g / kg of MGDF, PEG-MGDF, DMPG: MGDF or DMPG: PEG-MGDF, were measured and the results were measured. presented in FIGURE 31. The indicated concentration of each form was administered subcutaneously in normal female Balb / c mice, once daily for 8 days. Blood tests from a small lateral cut in a vein of the tail were collected 24 hours after the last injection. Cellular blood analyzes were performed with an electronic Sysmex blood cell analyzer (Baxter Diagnostics, Inc. Irvine, CA). The data are represented as the average of the determinations of 4 animals, +/- the standard error of the mean. Other cell parameters in the blood, such as the white cell count in the blood or the red cell count in the blood, were not affected by these treatments (the data showed nothing). The results indicate that the pegylation of E. coli MGDF 1-163, increases the in vivo activity of the molecule. More importantly, previous studies demonstrate that insertion in negatively charged lipid bilayers does not adversely affect the biological activity of the various forms of MGDF.

A-265B 69 SEQUENCE LIST (1) GENERAL INFORMATION: (i) APPLICANT: AMGEN INC. (ii) TITLE OF THE INVENTION: METHODS AND STABLE COMPOSITIONS PROTEINS: PHOSPHOLIPIDS (iii) SEQUENCE NUMBER: 2 (iv) CORRESPONDENCE ADDRESS: (A) RECIPIENT! Amgen Inc. (B) STREET 1840 Dehavilland Drive (C) CITY: Thousand Oaks (D) STATE: California (E) COUNTRY: USA (F) C.P .: 91320-1789 (v) LEGIBLE FORM OF THE COMPUTER: (A) TYPE OF MEDIUM: soft disk (B) COMPUTER: IBM compatible PC (C) OPERATING SYSTEM: PC-DOS / MS-DOS (D) SOFTWARE! Patentln Relay # 1.0, Version # 1.25 (vi) CURRENT APPLICATION DATA: (A) APPLICATION NUMBER: (B) CLASSIFICATION DATE: (C) CLASSIFICATION: RMATION FOR THE SEC. ID. N0: 1: (i) CHARACTERISTICS OF THE SEQUENCE: (A) LENGTH: 1342 base pairs (B) TYPE: nucleic acid (c) HEBRA: simple (D) TOPOLOGY: linear (ii) TYPE OF MOLECULE: cDNA (ix) CHARACTERISTICS: (A) NAME / KEY: CDS (B) LOCATION: 99..621 A-265B? 1 (xi) DESCRIPTION OF THE SEQUENCE: SEC. ID. NO: l: CAGGGAGCCA CGCCAGCCAA GACACCCCGG CCAGAATGGA GCTGACTGAA TTGCTCCTCG 60 TGGTCATGCT TCTCCTAACT GCAAGGCTAA CGCTGTCC AGC CCO GCT CCT CCT 113 Pro Pro Pro Wing 1 5 GCT TGT GAC CTC CGA GTC CTC AGT AAA CTG CTT CGT GAC TCC CAT GTC 161 Ala Cys Asp Leu Arg Val Leu Ser Lya Leu Leu Arg Asp Ser His Val 10 15 20 CTT CAC AGC AGA CTG AGC CAG TGC CCA GAG GTT CAC CCT TTG CCT ACA 209 Leu His Ser Arg Leu Ser Gln Cys Pro Glu Val His Pro Leu Pro Thr 25 30 35 CCT GTC CTG CTG CCT GCT GTG GAC TTT AGC TTG GGA GAA TGG AAA ACC 257 Pro Val Leu Leu Pro Wing Val Asp Phe Ser Leu Gly Glu Trp Lys Thr 40 45 SO CAG ATG GAG GAG ACC AAG GCA CAG GAC ATT CTG GGA GCA GTG ACC CTT 305 Gln Mee Giu Giu Thr Lys Ala Gln Asp lie Leu Gly Wing Val Thr Leu 55 60 65 CTG GG GGA GTG ATG GCA OCA CGG GGA CAG CTG GGA CCC ACT TGC 353 Leu Leu Glu Gly Val Mee Wing Wing Arg Gly G n Leu Gly Pro Thr Cys 70 75 SO 85 CTC TCA TCC CTC CTG GGG CAG CTT TCT GGA CAG GTC CGT CTC CTC CTT 401 Leu Ser Ser Leu Leu Gly G n Leu Ser Gly Gln Val Arg Leu Leu Leu 90 95 100 GGG GCC CTG CAG AGC CTC CTT GGA ACC CAG CTT CCT CCA CAG GGC AGG 449 Gly Ala Leu Gln Ser Leu Leu Gly Thr Gln Leu Pro Pro Glp Gly Arg 105 110 115 ACC ACCT GCT CAC AAG GAT CCC AAT GCC ATC TTC CTG AGC TTC CAA CAC 497 Thr Thr Wing His Lys Asp Pro Asp Ala le Phe Leu Ser Phe Gln His 120 125 130 CTG CTC CGA CGA GGA AAG GTG CGT TTC CTG ATG CTT GTA GCA GGG TCC ACC 545 Leu Leu Arg Gly Lys Val Arg Phe Leu Mee Leu Val Gly C and Ser Thr 135 140 145 CTC TGC GTC AGG CGG GCC CCA CCC ACC GCT GTC CCC AGC AGA ACC 593 Leu Cys Val Arg Arg Ala Pro Pro Thr Thr Ala Val Pro Ser Arg Thr 150 155 160 165 TCT CTA GTC CTC ACÁ CTG AAC GAG CTC C CAAACAGGAC TTCTGCATTG 641 Ser Leu Val Leu Thr Leu Asn Giu Leu 170 TTCGAGACAA ACTTCACTGC CTCAGCCAGA ACTACTCGCT CTsGGCTTCT GAAGTGGCAG 701 CAGGGATTCA GAGCCAAGAT TCCTGGTCTG CTGAACCAAA CCTCCAGGTC CCTGGACCAA 761 ATCCCCGGAT ACCTGAACAG GATACACGAA CTCTTGAA C AACTCGTGG ACTCTTTCCT 821 A-265B 72 GGACCCTCAC GCAGGACCCT AGGAGCCCCG GACATTTCCT CAGGAACATC AGACACAGGC 881 TCCCTGCCAC CCAACCTCCA GCCTGGATAT TCTCCTTCCC CÁACCCATCC TCCTACTGGA 941 CAGTATACGC TCTTCCCTCT TCCACCCACC TTGCCCACCC CTGTGCTCCA GCTCCACCCC 1001 CTCCTTCCTC ACCCTTCTCC TCCAACGCCC ACCCCTACCA GCCCTCTTCT AAACACATCC 1061 TACACCCACT CCCAGAATCT GTCTCAGCAA GGCTAACCTT CTCACACACT CCCOACATCA 1121 GCATTsTCTC GTGTACAGCT CCCTTCCCTG CAGCGCGCCC CTGGCACACA ACTOCACAAG 1181 ATTTCCTACT TTCTCCTGAA ACCCAAAGCC CTGGTAAAAG GCATACACAC GACTGAAAAC 1241 GGAATCATTT TTCACTGTAC ATTATAAACC TTCAGAAGCT ATTTTTTTTAA GCTATCAGCA 1301 ATACTCATCA GAGCAGCTAG CTCTTTGGTC TATTTTCTGC A 1342 (2) INFORMATION FOR SEC. ID. NO: 2: (i) CHARACTERISTICS OF THE SEQUENCE: (A) LENGTH: 174 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) TYPE OF MOLECULE: protein (xi) DESCRIPTION OF THE SEQUENCE: SEC. ID. NO: 2: Pro Pro Pro Wing Ala Cys Asp Leu Arg Val Leu Ser Lys Leu Leu 1 5 10 15 Arg Asp Ser His Val Leu His Ser Arg Leu Ser Gln Cyn Pro Glu Val 20 25 30 His Pro Leu Pro Thr Pro Val Leu Leu Pro Ala Val Asp Phe Ser Leu 35 40 45 Gly Glu Trp Lys Thr Gln Mee Glu Glu Thr Lys Wing Cln Aßp lie Leu 50 55 60 Gly Wing Val Thr Leu Leu Leu Glu Giy Val Mee Ala Wing Arg Gly Gln 65 70 75 80 Leu Gly Pro Thr Cyß Leu Ser Ser Leu Leu Gly Gln Leu Ser Gly Gln 85 90 95 Val Arg Leu Leu Leu Gly Ala Leu Gln Ser Leu Leu Gly Thr Gln Leu 100 105 110 Pro Pro Gin Gly Arg Thr Thr Wing His Lys Asp Pro Asn Ala Phe 115 120 125 Leu Ser Phe Gln His Leu Leu Arg Gly Lys Val Arg Phß Leu Mee Leu 130 135 140 A-265B 73 Val Cly Gly Ser Thr Leu Cys Val Arg Arg Ala Pro Pro Thr Thr Ala 145 150 155 160 Val Pro Ser Arg Thr Ser Leu Val Leu Thr Leu Asn Glu Leu 165 170 It is noted that in relation to this date, the best method known by the applicant to bring to practice the aforementioned invention, is that which is clear from the present description of the invention. Having described the invention as above, the content of the following is claimed as property.

Claims (46)

1. A composition comprising a protein capable of transitioning in the mixed dissolved globular state - with an intact phospholipid liposomal vesicle, characterized in that the liposomal vesicle is composed of negatively charged phospholipids, to form a liposome-protein complex where only one portion of the protein is inserted into the lipid portion of the liposomal vesicle.

2. The composition according to claim 1, characterized in that the composition has a pH of -3.0 - 7.5 and has at least a lipid: protein ratio of 10: 1.

3. The composition according to claim 1, characterized in that the liposomal vesicle is selected from the group consisting of: dioleoylphosphatidylglycerol (DOPG), dimyristoylphosphatidylglycerol (DMPG), dipalmitoylphosphatidylglycerol (DPPG), egphosphatidylglycerol, dioleoylphosphatidylethanolamine (DOPE), egphosphatidylethanolamine, dioleoylphosphatidic acid ( DOPA), dimyristoylphosphatidic acid (DMPA), dipalmitoylphosphatidic acid (DPPA), dioleoylphosphatidylserine (DOPS), dimyristoylphosphatidylserine (DMPS), dipalmitoylphosphatidylserine (DPPS), egphosphatidylserine, lysophosphatidylglycerol, lysophosphatidylethanolamine, and lysophosphatidylserine.

4. The composition according to claim 1, characterized in that the protein is a cytosine.

5. The composition according to claim 4, characterized in that the cytosine is a hematopoietic factor.

6. The composition according to claim 5, characterized in that the hematopoietic factor is selected from the group consisting of G-CSF, GM-CSF and MGDF.

7. The composition according to claim 6, characterized in that the haematopoietic factor is G-CSF.

8. The composition according to claim 7, characterized in that the G-CSF is the natural human G-CSF, or is obtained as a product of the expression of a prokaryotic or eukaryotic host cell.

9. The composition according to claim 7, characterized in that the G-CSF is chemically modified G-CSF.

10. The composition according to claim 9, characterized in that the chemically modified G-CSF is pegylated G-CSF.

11. The composition according to claim 6, characterized in that the hematopoietic factor is the MGDF.

12. The composition according to claim 11, characterized in that the MGDF is the natural human MGDF, or obtained as a product of the expression of a prokaryotic or eukaryotic host cell.

13. The composition according to claim 12, characterized in that the MGDF possesses an amino acid sequence of: MGDF 1-163 amino acids 1-163 of FIGURE 29.

14. The composition according to claim 12, characterized in that the MGDF possesses an amino acid sequence of: MGDF 1-332 amino acids 1-332 of FIGURE 29.

15. The composition according to claim 11, characterized in that the MGDF is the chemically modified MGDF.

16. The composition according to claim 15, characterized in that the chemically modified MGDF is the pegylated MGDF (PEG-MGDF).

17. The composition according to claim 16, characterized in that the PEG-MGDF is pegylated with polyethylene glycol.

18. The composition according to claim 17, characterized in that the PEG-MGDF is the mono-pegylated MGDF (mPEG-MGDF).

19. The composition according to claim 18, characterized in that the PEG group is fixed to the N-terminal thereof.

20. The composition according to claim 1, characterized in that the composition contains a pharmaceutically acceptable carrier.

21. The composition according to claim 1, characterized in that the protein is E. coli derived from rhG-CSF, wherein said liposomal vesicle is DOPG, and wherein said composition has a 50: 1 ratio of D0PG: rhG-CSF, has a pH of 4.5 and contains 10 mM sodium acetate.

22. The composition according to claim 1, characterized in that the protein is E. coli derived from MGDF 1-163, wherein said liposomal vesicle is DMPG, and wherein said composition has a 100: 1 ratio of DMPG: MGDF, has a pH of 5.0 and contains sodium acetate-10 mM and 5% sorbitol.

23. The composition according to claim 1, characterized in that the protein is mono-pegi-lated E. coli derived from the MGDF (mPEG-MGDF), wherein said liposomal vesicle is DMPG, and wherein said composition has a 100: 1 ratio of DMPG: mPEG-MGDF, has a pH of 5.0 and has 10mM sodium acetate and 5% sorbitol.

24. The composition according to claim 1, characterized in that the protein is CHO derived from MGDF 1-332, wherein said liposomal vesicle is DMPG, and wherein said composition has a 100: 1 ratio of DMPG: MGDF, has a pH of 5.0 and contains 10 mM sodium acetate and 5% sorbitol.

25. A method for preparing a liposome-to-protein composition which comprises the mixing of a protein, characterized in that the protein is capable of transitioning in the dissolved globular state, with a liposomal vesicle of intact phospholipid, wherein said liposomal vesicle is - negatively charged, such that only a portion of said protein is inserted into the lipid portion of said liposomal vesicle, and said liposome-pro-thein composition is obtained.

26. A method according to claim 25, characterized in that the composition has a pH of 3.0-7.5, and has at least a lipid to protein ratio of 10: 1.

27. A method according to claim 25, characterized in that the liposomal vesicle is prepared from a lipid selected from the group consisting of: dioleoylphosphatidylglycerol (DOPG), dimyristoylphosphatidylglycerol (DMPG), dipalmitoylphosphatidylglycerol (DPPG), egphosphatidylglycerol, dioleoylphosphatidylethanolamine (DOPE), egfosfatidiletanolamina, dioleoilfosfatidico acid (DOPA), dimiristoilfosfatidico acid (DMPA), dipalmitoylphosphatidic acid (DPPA), dioleoylphosphatidylserine (DOPS), dimiristoilfosfatidilserina (DMPS), dipalmitoylphosphatidylserine (DPPS), egfosfatidilserina, lysophosphatidylglycerol, lysophosphatidylethanolamine, lysophosphatidylserine and.

28. A method according to claim 25, characterized in that the protein is a cytosine.

29. A method in accordance with the claim 28, characterized in that the cytosine is a hematopoietic factor.

30. A method according to claim 29, characterized in that the hematopoietic factor is selected from the group consisting of G-CSF, GM-CSF and MGDF.

31. A method in accordance with the claim 30, characterized in that the hematopoietic factor is G-CSF.

.32. A method of conformity with the claim 31, characterized in that the G-CSF is the natural human G-CSF, or is obtained as a product of the expression of a prokaryotic or eukaryotic host cell.

33. A method according to claim 31, characterized in that the G-CSF is the chemically modified G-CSF.

34. A method in accordance with the claim 33, characterized in that the chemically modified G-CSF is the pegylated G-CSF.

35. A method according to claim 30, characterized in that the hematopoietic factor is the MGDF.

36. A method according to claim 35, characterized in that the MGDF is the natural human MGDF, or is obtained as a product of the expression of a prokaryotic or eukaryotic host cell.

37. A method according to claim 36, characterized in that the MGDF has an amino acid sequence of: MGDF 1-163 amino acids 1-163 of FIGURE 29.

38. A method according to claim 36, characterized in that the MGDF has an amino acid sequence of MGDF 1-332 amino acids 1-332 of FIGURE 29.

39. A method according to claim 35, characterized in that the MGDF is the chemically modified MGDF.

40. A method in accordance with the claim 39, characterized in that the chemically modified MGDF is the Pegylated MGDF (PEG-MGDF).

41. A method according to claim 40, characterized in that the PEG-MGDF is pegylated with polyethylene glycol.

42. A method according to claim 41, characterized in that the PEG-MGDF is the MGDF ono-pegilata-do (mPEG-MGDF).

43. A method according to claim 42, characterized in that the PEG group is fixed to the N-terminal thereof.

44. A method in accordance with the claim 25, characterized in that the composition contains a pharmaceutically acceptable carrier.

45. A method according to claim 21, characterized in that the insert includes the insertion of a portion of said protein, in a bilayer of the lipid of said liposomal vesicle.

46. A method according to claim 21, characterized in that the liposome-protein composition is -directly stabilized against the cleavage of the secondary structure of said protein.