Indian Patents. 260881:A PHARMACEUTICAL COMPOSITION HAVING STRUCTURE OF THERE-BRANCHED PEG-G-CSF CONJUGATE COMPRISING THREE-BRANCHED PEG AND G-CSF AND METHOD OF PREPARING THE SAME

Title of Invention	A PHARMACEUTICAL COMPOSITION HAVING STRUCTURE OF THERE-BRANCHED PEG-G-CSF CONJUGATE COMPRISING THREE-BRANCHED PEG AND G-CSF AND METHOD OF PREPARING THE SAME
Abstract	The present invention relates to three-branched PEG-G-CSF conjugate of general formula (1) in which the bonding ratio of three-branched polyethylene glycol(PEG) and G-CSF is 1:1 (mol/mol), wherein PEG has an average molecular weight of from 200 to 45,000 daltons; a pharmaceutical composition comprising the same, and a preparing method thereof.

Full Text	TECHNICAL FIELD The present invention relates to a three-branched PEG-G-CSF conjugate. BACKGROUND ART Colony stimulating factor(CSF), as a glycoprotein acting on production, division, and activity of hematopoietic cell, is classified into granulocyte colony stimulating factor(G-CSF), macrophage colony stimulating tactor(M-CSF), granulocyte-macrophage colony stimulating factor(GM-CSF), multi colony stimulating factor(multi-CSF), etc. It is known that G-CSF increases releasing of matured neutrophils into peripheral blood by facilitating production and division of neutrophilic precursor at bone marrow; induces a phagocytic activity by activating matured neutrophils; represents an antibody dependent cell mediated cytotoxicity; generates superoxides; and increases reactivity against chemotactic factor. Human granulocyte colony stimulating factor(hG-CSF) was firstly separated from bladder carcinoma cell line 5637 by Welt et al. in 1985, and recombinant hG-CSF(rhG-CSF) was produced by cloning of gene from cDNA library of a cancer cell by Nagata et al., Souza et al. in 1986. rhG-CSF currently used in clinical trials is filgrastim binding methionine at N- terminal of protein without glycosylation, produced in Escherichia coli, and lenograstim with glycosylation, produced in animal cell. In case of using rhG-CSF clinically, rhG-CSF has to be administrated once or more a day for decreasing leukocyte because rhG-CSF has a short duration of pharmacological effect. Thus, many studies have been conducted to increase a half-life of circulation of rhG-CSF using water-soluble polymer, thereby developing the drug to have a long duration of activity and high stability, and decreasing the frequency of administration. And, polyethylene glycol in the water-soluble polymer is strongly hydrophilic, and can increase solubility at the time of bonding with protein for treatment. Also, polyethylene glycol is effective for increasing the molecular amount of protein bonded thereto, with maintaining main biological functions such as enzyme activity and receptor binding. Thus, polyethylene glycol can decrease the filtration of kidney, and effectively protect protein from proteolytic enzyme to decompose the protein. Therefore, many studies have been conducted to find out modifying methods of protein by using polyethylene glycol because it has the advantages to prevent protein decomposition, increase the stability and circulation time of protein, and decrease immunogenicity. Korean Patent No. 0508358 disclosed conjugates and a preparing method thereof, wherein the conjugates have biological activities and are in the form of binding a biocompatibility polymer to thiol of cysteine residue of G-CSF, in the molar ratio of the polymer: G-CSF to stoichiometrically 0.7~1.3:1, preferably 1:1. But, since the cysteine residue of G-CSF which does not form disulfide bonds in G-CSF is main bonding part with G-CSF receptor, the conjugate using the cysteine residue has a drawback that its actual proliferation effect for neutrophil is little. And, because the conjugates immediately aggregate to bind a PEG to cysteine residues of G-CSF, it has a disadvantage that a small amount of SDS should be added to the conjugate to prevent the aggregation for safekeeping and the SDS was removed by ultrafiltration for administrating in vivo. Korean Patent No. 0507796 disclosed PEG-homodimer conjugates binding a biological active polypeptide and a PEG, to increase the half-life in vivo of polypeptide. Particularly, it described that the homodimer binds an amino group of lycine residues of bioactive polypeptide of two molecules to a PEG to increase the residual time and sustain biological activities for a long time. But, the conjugates have less activity than mono-PEG conjugates because of conjugation of excessive PEGs and physicochemical and biological characteristics of the conjugates are not uniform because of non-specific conjugation. And, the method of binding rhG-CSF to linear SCM-MPEG(Succinimidyl carboxymethyl ester of methoxy PEG) is known. The mono-PEG-G-CSF conjugates prepared by the method have 3 types of position isomers modified at N-terminal, lycine 35, and lycine 45 residues. Particularly, the conjugate modified at lycine 35 has a disadvantage leaving PEGs from the conjugate (Korean Patent No. 0248111, United State Patent No. 5,824,784). United State Patent No. 5,951,974 disclosed conjugates of binding a linear SC- PEG(polyethylene glycol-N-succinimide carbonate) to gene recombination alpha interferon. The conjugates consist of covalent conjugates of urethane bond at e-amino group of lycine residues of interferon, a-amino group of N-terminal cystaine residues, and an amino group of histidine residues. But, the conjugates have a disadvantage leaving PEGs from the conjugate because the urethane bond of conjugates are unstable. Commonly used linear polyethylene glycol has a molecular weight of about 1,000~ 25,000 daltons, but has a limitation in binding many linear high molecules to protein or peptide, with maintaining their activities, due to limited biological active regions of protein and peptide. Korean Patent No. 0254097 disclosed conjugates binding a two-branched PEG of a lycine skeletal structure to gene recombination alpha interferon. The conjugates have a merit to prevent PEG from binding to multi parts of interferon, and have 2 times the molecular weight of PEG by that of the linear PEG because two linear PEGs bind to a single part of interferon. But, the two-branched PEG can be hydrolyzed to a single chain for safekeeping or for reacting under an alkali condition because the two-branched PEG having a lycine skeletal structure has two urethane bonds in the PEG. Generally, it is known that biological activity, durability, etc. of conjugates of binding a PEG to gene recombination protein is dependent on size and part modified in the PEG. But, it has not been known yet whether a size of PEG in conjugates binding PEG and rhG-CSF affects biological activities of a protein. Therefore, there has been a need in the art to overcome the disadvantages of the known method through controlling a size of PEG and binding to various skeletal structures of PEG not to decrease bioactivities of G-CSF and to increase the stability of binding parts. DISCLOSURE OF THE INVENTION TECHNICAL PROBLEM The object of the present invention is to provide three-branched polyethylene glycol(PEG)-G-CSF conjugates prepared by conjugating three-branched PEG and G-CSF, having high stability and same or more bioactivities of G-CSF than G-CSF and PEG-G-CSF conjugates known in the art; a preparation method of the same; and a pharmaceutical composition containing the same. TECHNICAL SOLUTION To achieve the above object, the present invention provides three-branched polyethylene glycol(PEG)-G-CSF conjugates conjugated between three-branched PEG derivatives and G-CSF. The present invention also provides a method of preparing three-branched PEG-G- CSF conjugates comprising conjugating three-branched PEG derivatives and G-CSF. The present invention also provides a pharmaceutical composition comprising the conjugates. The present invention also provides a method of treating neutropenia, preventing neutropenia, or facilitating increase of the number of neutrophil at the time of hematopoietic stem cell mobilization to peripheral blood and hematopoietic stem cell transplatation, comprising administrating the conjugate of the present invention as an effective ingredient. The present invention is explained in detail below. Three-branched PEG-G-CSF conjugates of the present invention may be prepared by conjugating three-branched polyethylene glycoI(PEG), represented by the following general formula(l): wherein, n is an integer of 1 to 1,000; m is an integer of 10 to 1,000; Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an integer of 1 to 6; Y is an amide bond formed by combining NH2 functional group in G-CSF and a functional group of PEG derivative. The PEG has an average molecular weight of from 200 to 45,000 daltons, preferably from 20,000 to 45,000 daltons, more preferably from 23,000 to 43,000 daltons. Because a medicinally effective time can be changed according to an average molecular weight of PEG, the average molecular weight of PEG used in the present invention can be changed, depending on the time required for treatment. The method of preparing three-branched PEG-G-CSF conjugate comprising conjugating three-branched PEG derivatives and G-CSF of the present invention is explained in detail below. The three-branched PEG-G-CSF conjugates of general formula (1) are prepared by forming a covalent bond between three-branched PEG derivative of the following general formula (2) and G-CSF wherein PEG has an average molecular weight of from 200 to 45,000 daltons, preferably from 20,000 to 45,000 daltons, more preferably from23,000 to 43,000 daltons: wherein, n is an integer of 1 to 1,000; m is an integer of 10 to 1,000; Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an integer of 1 to 6; and the functional group which can chemically react with proteins and peptides containing G-CSF is N-hydrosuccinimide. Three-branched PEG derivatives of the present invention activate high molecules having branched structure in which three linear biological receptive high molecules are combined. All of three OH (hydroxy) regions in the glycerol skeletal structure are polymerized with ethylene glycol unit molecules, and the end of one region is activated as a functional group. The other two regions except for the activated region are substituted with monomethoxy, to prevent additional reactions. When the above branched PEG derivatives are prepared, the size of each linear PEG can be controlled freely by a method known in the art. Three-branched PEG derivatives in the present invention can be used as PEG derivatives, with the functional group chemically reacting with protein and peptide, known in the art. PEG derivatives of general formula (2) (with N-hydroxysuccinaimde) are preferable in terms of yield of the conjugate of the present invention. The G-CSF in the present invention can be separated from mammalian organism or synthesized by a method known in the art such as gDNA cloning, cDNA cloning, etc. And, the G-CSF can be commercialized in the market. And, the covalent bond between G-CSF and the three-branched PEG derivatives can be formed at a low temperature. The reaction is completed by adding acids, and the prepared three-branched PEG-G-CSF conjugates can be purified by a method known in the art such as the purification method using a cation exchange resin. In the present invention, the reaction molar ratio of the G-CSF to the three-branched PEG derivative is from 1: 0.5 to 1: 50. A preferable molar ratio of the G-CSF to the three- branched PEG derivative is from 1: 0.5 to 1: 5 because the yield of mono PEG-G-CSF conjugate is decreased, as the molar ratio of polyethylene glycol to the G-CSF is increased. The present invention also provides a pharmaceutical composition for treating or preventing symptoms caused by decreased hematopoietic function or decreased immunologic function. Particularly, clinical trials using the conjugate of the present invention as an effective showed that the number of neuctrophil was decreased, and the symptoms were relieved or controlled, for diseases caused in a treatment such as cancer chemotherapy or radiation therapy; infective diseases caused by bacteria, virus and fungus; other infective diseases; diseases caused by genetic or environmental reason such as severe chronic neutropenia and leukemia; or geriatric diseases caused by again. For example, symptoms caused by decreased hematopoietic function or decreased immunologic function are neutropenia caused by cancer chemotherapy for blood tumor or solid cancer, neutropenia caused by myelodysplastic syndrome, neutropenia caused by aplastic anemia, congenital idiopathic neutropenia, and neutropenia caused from treating human immunodeficiency virus. The composition can be composed of PEG-G-CSF conjugates of the present invention or pharmaceutically acceptable diluent, antiseptics, solubilizer, emulsifier, adjuvant, and/or carrier. The compositions of the present invention can be formulated to injection agent, capsule, tablet, liquid drug, pill, ointment, oculentum, coUyrium, transdermal absorptive agent, paste, cataplasm, aerosols, etc. And, the effective dosage of G-CSF is very small such as 0.1 -500 μg (preferably, 5-50 μg). And, the drug comprising 0.1 -500 μg of G-CSF may be administrated to an adult, generally from 1 to 7 times a week. Thus, the effective dosage of the pharmaceutical composition of the present invention can be calculated by a known administration amount of G-CSF and a molecular weight of PEG used in the present invention. The effective dosage of pharmaceutical composition of the present invention may be varied, but generally once a week, and the composition can be administrated once or many times a day within a daily effective dosage range. The following examples are intended to further illustrate the present invention, and the scope of the present invention is not intended to be limited thereby in any way. ADVANTAGEOUS EFFECTS The three-branched PEG-G-CSF conjugates of the present invention are more pharmaceutically stable than linear- or two-branched PEG-G-CSF conjugates in the aspect of leaving PEGs from PEG-G-CSF conjugates and forming the aggregates of conjugates. And, the conjugates of the present invention can be used without additional separations of position isomers because the conjugates consist of position isomers with similar activities. And, the composition of the present invention has the effects of continuous production of neutrophilias and increased in-vivo half-life. DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic drawing illustrating the analytic results of a mixture formed to prepare conjugates of three-branched PEG having MW 23,000Da and G-CSF according to Example 1, by size-exclusion high performance liquid chromatography. Fig. 2 is a schematic drawing illustrating the analytic results of a mixture formed to prepare conjugates of three-branched PEG having MW 43,000Da and G-CSF according to Example 2, by size-exclusion high performance liquid chromatography. Fig. 3 is a schematic drawing illustrating the analytic results of a mixture formed to prepare conjugates of linear PEG having MW 10,000Da and G-CSF according to Comparative Example 1, by size-exclusion high performance liquid chromatography. Fig. 4 is a schematic drawing illustrating the analytic results of the purities of mono three-branched PEG-G-CSF conjugates separated from the mixture of Example 1, by size- exclusion high performance liquid chromatography. Fig. 5 is a schematic drawing illustrating the analytic results of the purities of mono three-branched PEG-G-CSF conjugates separated from the mixture of Example 2, by size- exclusion high performance liquid chromatography. Fig. 6 is a schematic drawing illustrating the analytic results of the purities of mono linear PEG-G-CSF conjugates separated from the mixture of Comparative Example 1, by size-exclusion high performance liquid chromatography. Fig. 7 is a schematic drawing illustrating the analytic results of mono three-branched PEG-G-CSF conjugates separated from the mixture of Example 1, by Matrix-Assisted Laser Desorption Ionization (MALDI-TOF) Mass Spectrometry. Fig. 8 is a schematic drawing illustrating the analytic results of mono three-branched PEG-G-CSF conjugates separated from the mixture of Example 2, by Matrix-Assisted Laser Desorption Ionization (MALDI-TOF) Mass Spectrometry. Fig. 9 is a schematic drawing illustrating the analytic results of mono linear PEG-G- CSF conjugates separated from the mixture of Comparative Example 1, by Matrix-Assisted Laser Desorption Ionization (MALDI-TOF) Mass Spectrometry. Fig. 10 is a schematic drawing illustrating the analytic results of the position isomers comprising mono three-branched PEG-G-CSF conjugates separated from the mixture of Example 1, by ion exchange chromatography. Fig. 11 is a schematic drawing illustrating the analytic results of biological activity of the position isomers of mono three-branched PEG-G-CSF conjugates separated from the mixture of Example 1, according to Experimental Example 1, by the method of Experimental Example 2. MODE FOR INVENTION Preparation of three-branched polyethylene glycol (PEG MW 23,000 Da)-G-CSF conjugates 10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of sodium borate buffer solution. Then 9.2mg of three-branched PEG having N- hydroxysuccinimde (NOF corporation, Japan) and the molecular weight of 23,000 daltons was added to the solution in the molar ratio of rhG-CSF : three-branched PEG to 1: 0.75. And, the solution was stirred at 4 °C for 90min. The reaction was stopped by adding lOOmM of HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the volume of the solution. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate (pH4.0) buffer solution, and three-branched PEG-G-CSF conjugate was separated. The mixture was fractioned by using 0 ~ IM of concentration gradient of sodium chloride(NaCl). The molar ratio of three-branched PEG to G-CSF in the three-branched PEG- G-CSF conjugates was confirmed by HPLC and SDS-PAGE. And, di- or more (tri-, tetra- ...) three-branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono-three-branched PEG-G-CSF conjugate in which a G-CSF and a three-branched PEG (MW 23,000Da) are conjugated (or called as mono-three-branched PEG-G-CSF conjugate). It was confirmed that the reaction mixture consisted of 35.4% of mono-three-branched PEG-G-CSF conjugate(mono-PEG-G-CSF conjugate), about 59.4% of G-CSF unmodified by PEG, and the others [di-three-branched PEG-G-CSF conjugate(di-PEG-G-CSF conjugate), and N-hydroxysuccinimide(NHS)], by a size-exclusion high performance chromatography (see Fig. 1, wherein 1 and 2 represent di-PEG-G-CSF conjugate, 3 represents mono-PEG-G- CSF conjugate, 4 represents G-CSF unmodified by PEG, and 5 represents NHS leaving from three-branched PEG). And, the purity of mono-three-branched PEG-G-CSF separated from the mixture was measured by size-exclusion high performance liquid chromatography (see Fig. 4, wherein 2 represents di-PEG-G-CSF conjugate, and 3 represents mono-PEG-G-CSF conjugate) and the molecular weight was measured by matrix assisted laser desorption/ionization time-of-flight(MALDI-TOF) mass spectrometry (see, Fig. 7: the analytic result of mono-three-branched PEG-G-CSF by MALDI-TOF/MS). And, the position isomers were separated by cation exchange resin analytic column (SP-5WP, TOSOH) to confirm that the ratio for each position isomer is about 54%: 33%: 10% (see. Fig. 10). And, It was confirmed that each position isomer has very similar activity, in the experiment for biological activity of each position isomer (see. Fig. 11) Preparation of three-branched polyethylene glycol (PEG MW 43,000 Da)-G-CSF conjugates 10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of sodium borate buffer solution. Then, 17.2mg of three-branched PEG with M- hydroxysuccinimde (NOF corporation, Japan) and the molecular weight of 43,000 daltons was added to the solution to make the molar ratio of the rhG-CSF : the three-branced PEG to 1: 0.75. And, the solution was stirred at 4 °C for 90min. And, the reaction was stopped by adding 100mM of HCl to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume. And, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer solution, and three-branched PEG-G-CSF conjugate was separated therefrom. The mixture was fractioned by using 0 ~ IM of concentration gradient of sodium chloride(NaCl). The molar ratio of three-branched PEG to G-CSF in the three- branched PEG-G-CSF conjugates was confirmed by HPLC and SDS-PAGE. And di- or more (tri, tetra...)-three-branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono-three-branched PEG-G-CSF conjugate in which a G-CSF and a three- branched PEG (MW 43,000Da) are conjugated(or called as mono-three-branched PEG-G- CSF conjugate). It was confirmed that the mixture of the reaction consisted of 18.7% of mono-three- branched PEG-G-CSF conjugate (mono-PEG-G-CSF conjugate), about 40.1% of G-CSF unmodified by PEG, and the others [di-three-branched PEG-G-CSF conjugate(di-PEG-G- CSF conjugate), and N-hydroxysuccinimide(NHS)] by size-exclusionhigh performance chromatography (see Fig. 2, wherein 2 represents di-PEG-G-CSF conjugate, 3 represents mono-PEG-G-CSF conjugate, and 4 represents G-CSF unmodified by PEG). And, the purity of mono-three-branched PEG-G-CSF separated from the mixture was measured by size- exclusion high performance liquid chromatography (see Fig. 5, wherein 1 represents oligo- PEG-G-CSF conjugate, 2 represents di-PEG-G-CSF conjugate, and 3 represents mono-PEG- G-CSF conjugate), and the molecular weight was measured by matrix assisted laser desorption/ionization time-of-flight(MALDI-TOF) mass spectrometry (see. Fig. 8; the analytic result of mono-three-branched PEG-G-CSF by MALDI-TOF/MS). Preparation of linear polyethylene glycol (PEG MW 10,000 Da)-G-CSF conjugates 10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of sodium borate buffer solution. And, 5.2mg of linear PEG with N-hydroxysuccinimde (NOF corporation, Japan) and the molecular weight of 13,000 daltons was added to the solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. And, the solution was stirred at 4 °C for 90min. And, the reaction was stopped by adding 100mM of HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume. And, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride(NaCl). The molar ratio of linear PEG to G-CSF in the linear PEG-G-CSF conjugate was confirmed by HPLC and SDS-PAGE. And, di- or more (tri, tetra...)-linear PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono-linear PEG-G-CSF conjugate in which a G-CSF and a linear PEG(MW 13,000Da) are conjugated(or called as mono-linear PEG-G- CSF conjugate). It was confirmed that the mixture of the reaction consisted of 45.6% of mono-linear PEG-G-CSF conjugate(mono-PEG-G-CSF conjugate), about 45.9% of G-CSF unmodified by PEG, and the others [di-linear PEG-G-CSF conjugate(di-PEG-G-CSF conjugate), and N- hydroxysuccinimide(NHS)], by a size-exclusion high performance chromatography (see Fig. 3, wherein 1 represents oligo-PEG-G-CSF conjugate, 2 represents di-PEG-G-CSF conjugate, 3 represents mono-PEG-G-CSF conjugate, and 4 represents G-CSF unmodified by PEG). And, the purity of mono-linear-branched PEG-G-CSF separated from the mixture was measured by size-exclusion high performance liquid chromatography (see Fig. 6, wherein 1 represents oligo-PEG-G-CSF conjugate, 2 represents di-PEG-G-CSF conjugate, 3 represents mono-PEG-G-CSF conjugate, and 4 represents G-CSF unmodified by PEG), and the molecular weight was measured by matrix assisted laser desorption/ionization time-of- flight(MALDI-TOF) mass spectrometry (see, Fig. 9). Preparation of linear polyethylene glycol (PEG MW 20,000 Da)-G-CSF conjugates 10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of sodium borate buffer solution, and then 8.0mg of linear PEG having N-hydroxysuccinimde (NOF corporation, Japan) and the molecular weight of 20,000 daltons was added to the solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. And, the solution was stirred at 4 °C for 90min. The reaction was stopped by adding 100mM of HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride (NaCl). The molar ratio of linear PEG to G-CSF in the linear PEG-G-CSF conjugate was confirmed by HPLC and SDS-PAGE. And, di- or more(tri, tetra...)-linear PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono-linear PEG-G-CSF conjugate in which a G-CSF and a a linear PEG(MW 20,000Da are conjugated(or called as mono-linear PEG-G- CSF conjugate). Preparation of two-branched polyethylene glycol (PEG MW 20,000 Da, lycine skeletal structure)-G-CSF conjugates 10 mg of rhG-CSF(Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of sodium borate buffer solution, and then 9.1mg of two-branched PEG having N- hydroxysuccinimde(Nektar, USA) and the molecular weight of 20,000 daltons was added to the solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. And, the solution was stirred at 4 °C for 90min. The reaction was stopped by adding 1 OOmM of HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride(NaCl). The molar ratio of two-branched PEG to G-CSF in the two-branched PEG- G-CSF conjugate was confirmed by HPLC and SDS-PAGE. And, di- or more(tri, tetra.. .)- two branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono- two-branched PEG-G-CSF conjugate in which a G-CSF and a two-branched PEG(MW 20,000Da are conjugated(or called as mono-linear PEG-G-CSF conjugate). Preparation of two-branched polyethylene glycol (PEG MW 20,000 Da, glyerine skeletal structure)-G-CSF conjugates 10 mg of rhG-CSF(Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of sodium borate buffer solution. And, 8.0mg of two-branched PEG having N- hydroxysuccinimde(NOF, Japan) and the molecular weight of 20,000 daltons was added to the solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. The solution was stirred at 4 °C for 90min. And, the reaction was stopped by adding 100mM of HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride(NaCl). The molar ratio of two-branched PEG to G-CSF in the two-branched PEG- G-CSF conjugate was confirmed by HPLC and SDS-PAGE. And, di- or more(tri, tetra...)- two branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono- two-branched PEG-G-CSF conjugate in which a G-CSF and a two-branched PEG(MW 20,000Da) are conjugated(or called as mono-linear PEG-G-CSF conjugate). Preparation of mono-methoxypolyethylene glycol-G-CSF conjugates binding at N-terminal a-amino residue The title conjugate was prepared by a method described in Korean Patent No. 0248111, and United State Patent No. 5,824,784. The method as briefly described is as follows. 100mM of sodium phosphate containing 20mM of NaCNBHs, and 5mg/ml of rhG-CSF were stirred at 4 °C. And, 5 times mol of linear methoxylpolyethylene glycol(mPEG) aldehyde(Nektar, USA) was added to the solution, which was stirred for 10hs. The reaction was stopped by adding 100mM of HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride(NaCl). And, the eluates were concentrated and sterilized-filtered to obtain the mono- mPEG-G-CSF conjugate in which a G-CSF and a two-branched PEG(MW 20,000Da) are conjugated(or called as mono-linear PEG-G-CSF conjugate). Preparation of mono-methoxypolyethylene glycol-G-CSF conjugates binding at thiol group of cysteine residue of G-CSF The title conjugate was prepared by a method described in Korean Patent No. 0508358. The method as briefly described is as follows. 1 mg of rhG-CSF(Dong-A Pharm. Co., Ltd.) was added to 1mL of sodium phosphate buffer solution(0.1M) having pH8.5, and 52.6mg of polyethylene glycol maleimide(NOF, Japan) was added to the solution. And, the solution was stirred for 1h at room temperature. Then, PEG derivatives remaining after the reaction were removed from the solution by centricon 30(Amicon, USA). And, the solution was concentrated and sterilized-filtered to obtain the mono-mPEG-G-CSF conjugate. The property and pharmacological activity tests were conducted by using the conjugates prepared above, and the results are as follows. Analytic separation of position isomers of three-branched PEG-G-CSF conjugates 100 μg of Mono-three-branched PEG-G-CSF conjugates of Example 1 was input to a cation exchange resin analytic column(SP-5WP, TOSOH) equalized with 25mM of sodium acetate buffer(pH 4.0, controlled with glacial acetic acid), and each position isomer of the conjugates was separated by 100mM of sodium acetate buffer, pH 7.8). It was confirmed that the ratio of each position isomer is about 54%: 33%: 10% (see. Fig. 10). Analysis of biological activities of position isomers of three-branched PEG-G-CSF conjugates Relative Biological activities of position isomers of three-branched PEG-G-CSF conjugates of Example 1 were measured by using that of G-CSF. NFS-60 cell strain cultured at growth RPMI 1640 containing 10% FBS and 1ng/ml of rmIL-3 was washed 3 times with test RPMI 1640 containing 5% FBS. And, the cell strain was divided into 100ml(2xl05 cells/ml) at 96 well plate. Then the sample was diluted with test medium to prepare a sample of 200ng/ml, and 9 more samples of 200ng/ml by 5 times gradient, and the samples were divided into 3 wells, each 100ml, per concentration in 96well plates containing the cell solutions. Each well was cultured at 37°C and 5% CO2 incubator for 48hs, 40 μl of MTS was added to each well. After 2hs, an absorbance of the sample was measured at 490nm by an ELISA reader. EC50 was calculated by a dose response curve and a linear regression analysis of point comprised straight line of a standard curve, and the activities of position isomers were determined (see, Fig.l 1). The three-branched PEG-G-CSF conjugates of the present invention have position isomers having similar activities. That is, it is not necessary to remove some position isomers having relatively lower activity because the three-branched PEG-G-CSF conjugates of the present invention are composed of conjugates with uniformed activities Pharmacokinetics test of the three-branched PEG-G-CSF conjugates The pharmacodynamic test was conducted to inject hypodermically each of G-CSF, the PEG-G-CSF conjugates of Example 1, and the PEG-G-CSF conjugates of Example 2, into test rats (Sprague Dawley) which had 240 ~ 260g of body weights. After injecting them in the amount of 400μg per rat, the blood samples were collected from the rats at 0 min, 30min, 1hr, 2hr, 3hr, 4hr, 8hr, 12hr, 1day, 2days, 3days, 4days, 5days, 6days, and 7days, after the injection. And, the concentrations of the G-CSF conjugates in the sample were quantitatively analyzed by an enzyme-linked immunosorbent assay (ELISA) kit(Biosource, USA). The half-life in blood of the PEG-G-CSF conjugates of Example 1 and Example 2 was increased 3 times and 10.8, respectively, compared with that of G-CSF, and the areas under the concentration time curve in blood of the conjugates of Example 1 and Example 2 were also increased by 7.6 times and 23.4 times, respectively, compared with that of G-CSF [Table 1 : Pharmacokinetics test of the three-branched PEG-G-CSF conjugates and G-CSF at rat(Sprague Dawley rat)]. It was confirmed by this test that pharmacokinetics of PEG-G-CSF conjugate is dependent on a size of PEG. Stability test of the three-branched PEG-G-CSF conjugates on leaving PEGs from the conjugates PEG-G-CSF conjugates of Examples 1 and 2, and Comparative Examples 2, 3, and 4 were deposited under pH 7 and 37 °C for 7days, and the aspects of leaving PEGs in length of time were analyzed by reversed-phase high-performance liquid chromatography, to estimate the degrees of leaving PEGs on a skeletal structure, molecular weights, and branched types of PEG. In the PEG'S skeletal structure, the leaving PEGs occurred most frequently in Comparative Example 2 using linear PEG. However, the leavings PEGs from the conjugates of Examples 1 and 2, and Comparative Example 4 occurred less frequently than those of Comparative Examples 2 and 3 (Table 2: Comparison of generation rates of G-CSF in PEG- G-CSF conjugates having various PEG skeletal structures). This test confirmed that the three- branched PEG-G-CSF conjugate is more stable than the linear-, two-branched PEG-G-CSF conjugates, and the conjugate with a glycerol skeletal structure is more stable than the conjugates with a iycine skeletal structure. conjugates against forming aggregations according to a skeletal structure of PEG and a type ofbinding PEG to G-CSF Mono three-branched PEG-G-CSF conjugate of Example 1; mono PEG-G-CSF conjugate of Comparative Examples 2 and 4, having a similar molecular weight to the conjugate of Example 1 and different skeletal structure from the conjugate of Example 1; and mono PEG-G-CSF conjugate of Comparative Example 5 having different binding type from the conjugate of Example 1 were tested to measure stabilities against forming aggregations. Each sample was collected by 500μl, and the solvent of the samples were exchanged 3 times with 100mM of phosphate buffer solution having pH7. And, the samples were deposited at 37°C for 14 days, and formation of aggregation of the conjugates was analyzed by size- exclusion high performance liquid chromatography. The analysis results of forming of the aggregation represent that mono-PEG-G-CSF conjugate of Example 1 had most stability in the aspect of leaving PEGs, and formed the least aggregation (Table 3: Generation rates of aggregation conjugates according to a skeletal structure of PEG and a type of binding PEG to G-CSF). Mono-PEG-G-CSF conjugates of Comparative Examples 2 and 4 using PEG having the same glycerol skeletal structure as PEGs in Example 1, and using the same binding type to Example 1, as well as the conjugate of Example 1, formed some aggregation. However, mono-PEG-G-CSF conjugates of Comparative Example 3 using PEG having different skeletal structure from PEGs in Example 1 (although using the same binding type as Example 1) formed much aggregation. And, mono-PEG-G-CSF conjugates of Comparative Example 5 using PEG having different skeletal structure from PEGs in Example 1 and using the different binding type from Example 1 formed 2 times or more of aggregation, compared with the conjugates of Example 1. In conclusion, mono-three-branched PEG-G-CSF conjugate of Example 1 is most stable in both leaving PEGs from the conjugates and forming aggregation. That is, this advantageous effect is from different skeletal structures between branched PEG and branched PEG used in the present invention. Neutrophil proliferation activities assay of PEG-G-CSF conjugates at animal model to which an anti-cancer medicine is administrated. Neutrophil proliferation activities of mono-three-branched PEG-G-CSF conjugates of Example 1, mono-mPEG(methoxy polyethylene glycol)-G-CSF conjugates of Comparative Examples 5 and 6 having different binding type and different molecular weight from the conjugates of Example 2 were assayed by mouse model to which an anti-cancer agent was administrated. Mice(BDFl) of 20 ~ 25g were administrated with 200 mg/kg of an anti-cancer agent(cyclophosphamide) to decrease the number of neutrophil for 2days. And, each of the conjugate of Example I and the conjugates of Comparative Examples 5 and 6 was administrated subcutaneously once, 1mg/kg per mouse, and G-CSF was administrated 8 times by 0.125 mg/kg per mouse to make the total administration amount 1 mg/kg. The mice' bloods were collected at 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and 12 days, and the total numbers of leukocyte and neutrophil were measured by automatic hemocyte measuring instrument (HEMAVET 850, Drew Scientific Ltd., UK). The periods that started the recovery of neutrophils decreased by an anti-cancer agent and the times required for the recovery of neutrophil, in the conjugate of Example 1 and the conjugates of Comparative Examples 5 and 6 were similar. However, the time reaching the maximum number of neutrophil in the conjugate of Example 1 was about 12hs faster than those of the conjugates of Comparative Examples 5 and 6. The area under the increased neutrophil count-time curve(AU-NC) of the conjugate of Example 1 was largest (Table 4: Neutrophil proliferation activities assay of PEG-G-CSF conjugates at animal model to which an anti-cancer medicine is administrated). Therefore, it was confirmed that mono-three-branched PEG-G-CSF conjugate has a rapid effect of Neutrophil proliferation. [Table 4] t: Time for neutrophil to be increased after administration of a medicine *: Time to reach l,000/mm3 or more of neutrophil after administration of an anti-cancer medicine. INDUSTRIAL APPLICABILITY The three-branched PEG-G-CSF conjugates of the present invention are more pharmaceutically stable than linear- or two-branched PEG-G-CSF conjugates in terms of leaving PEGs from PEG-G-CSF conjugates and forming the aggregates of conjugates. And, the conjugates of the present invention can be used without additional separations of position isomers because the conjugates consist of the position isomers with similar activities. And, the composition of the present invention has the effects of continuous production of neutrophilias and increased in vivo half-life. WE CLAIM: 1 . Three-branched PEG-G-CSF conjugate of the following general formula (1): in which a bonding ratio of three-branched polyethylene glycol(PEG) and G-CSF is 1:1 (mol/mol), and PEG has an average molecular weight of from 200 to 45,000 daltons, wherein, n is an integer of I to 1,000; m is an integer of 10 to 1,000; Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an integer of 1 to 6; Y is an amide bond formed by combining NH2 functional group in G-CSF and a functional group of PEG derivative 2. The conjugate of Claim 1, wherein the PEG has an average molecular weight of from 20,000 to 45,000 daltons. 3. The conjugate of Claim 1, wherein the PEG has an average molecular weight of from 23,000 to 43,000 daltons. 4. A method of preparing three-branched PEG-G-CSF conjugate of the general formula (1) of Claim 1 comprising forming a covalent bond of three-branched PEG derivative of the following, general formula (2) and G-CSF wherein PEG(polyethylene glycol) has an average molecular weight of from 200 to 45,000 daltons; wherein, n is an integer of 1 to 1,000; m is an integer of 10 to 1,000; Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an integer of 1 to 6; and the functional group which can chemically react with proteins and peptides containing G- CSF is N-hydrosuccinimide. 5. The method of Claim 4, wherein the PEG has an average molecular weight of from 20,000 to 45,000 daltons. 6. The method of Claim 4, wherein the PEG has an average molecular weight of from 23,000 to 45,000 daltons. 7. The method of Claim 4, wherein the molar ratio of G-CSF to three-branched PEG derivative is from 1: 0.5 to 1: 50. 8. The method of Claim 7, wherein the molar ratio of G-CSF to three branched PEG derivative in the reaction is from 1: 0.5 to 1: 5. 9. A pharmaceutical composition for treating or preventing symptoms caused by decreased hematopoietic function and decreased immunologic function, comprising the conjugate of any of claim 1, 2 or 3 as effective ingredient. 10. The pharmaceutical composition of claim 9, wherein the symptoms caused by decreased hematopoietic function and decreased immunologic function are solid cancer, neutropenia caused by cancer chemotherapy to blood tumor, neutropenia caused by myelodysplastic syndrome, neutropenia caused by aplastic anemia, congenital idiopathic neutropenia, and neutropenia caused by treating human immunodeficiency virus. 1. A method of treating neutropenia, preventing neutropenia, or facilitating increase of the number of neutrophil at the time of hematopoietic stem cell mobilization to peripheral blood and hematopoietic stem cell transplatation, comprising administrating the conjugate of any of claim 1, 2 or 3 as an effective ingredient. The present invention relates to three-branched PEG-G-CSF conjugate of general formula (1) in which the bonding ratio of three-branched polyethylene glycol(PEG) and G-CSF is 1:1 (mol/mol), wherein PEG has an average molecular weight of from 200 to 45,000 daltons; a pharmaceutical composition comprising the same, and a preparing method thereof.

Full Text

TECHNICAL FIELD
The present invention relates to a three-branched PEG-G-CSF conjugate.
BACKGROUND ART
Colony stimulating factor(CSF), as a glycoprotein acting on production, division, and
activity of hematopoietic cell, is classified into granulocyte colony stimulating factor(G-CSF),
macrophage colony stimulating tactor(M-CSF), granulocyte-macrophage colony stimulating
factor(GM-CSF), multi colony stimulating factor(multi-CSF), etc.
It is known that G-CSF increases releasing of matured neutrophils into peripheral
blood by facilitating production and division of neutrophilic precursor at bone marrow;
induces a phagocytic activity by activating matured neutrophils; represents an antibody
dependent cell mediated cytotoxicity; generates superoxides; and increases reactivity against
chemotactic factor.
Human granulocyte colony stimulating factor(hG-CSF) was firstly separated from
bladder carcinoma cell line 5637 by Welt et al. in 1985, and recombinant hG-CSF(rhG-CSF)
was produced by cloning of gene from cDNA library of a cancer cell by Nagata et al., Souza
et al. in 1986. rhG-CSF currently used in clinical trials is filgrastim binding methionine at N-
terminal of protein without glycosylation, produced in Escherichia coli, and lenograstim with
glycosylation, produced in animal cell.
In case of using rhG-CSF clinically, rhG-CSF has to be administrated once or more a day for
decreasing leukocyte because rhG-CSF has a short duration of pharmacological effect.
Thus, many studies have been conducted to increase a half-life of circulation of rhG-CSF
using water-soluble polymer, thereby developing the drug to have a long duration of activity
and high stability, and decreasing the frequency of administration.
And, polyethylene glycol in the water-soluble polymer is strongly hydrophilic, and can
increase solubility at the time of bonding with protein for treatment. Also, polyethylene
glycol is effective for increasing the molecular amount of protein bonded thereto, with

maintaining main biological functions such as enzyme activity and receptor binding. Thus,
polyethylene glycol can decrease the filtration of kidney, and effectively protect protein from
proteolytic enzyme to decompose the protein. Therefore, many studies have been conducted
to find out modifying methods of protein by using polyethylene glycol because it has the
advantages to prevent protein decomposition, increase the stability and circulation time of
protein, and decrease immunogenicity.
Korean Patent No. 0508358 disclosed conjugates and a preparing method thereof, wherein
the conjugates have biological activities and are in the form of binding a biocompatibility
polymer to thiol of cysteine residue of G-CSF, in the molar ratio of the polymer: G-CSF to
stoichiometrically 0.7~1.3:1, preferably 1:1. But, since the cysteine residue of G-CSF which
does not form disulfide bonds in G-CSF is main bonding part with G-CSF receptor, the
conjugate using the cysteine residue has a drawback that its actual proliferation effect for
neutrophil is little. And, because the conjugates immediately aggregate to bind a PEG to
cysteine residues of G-CSF, it has a disadvantage that a small amount of SDS should be
added to the conjugate to prevent the aggregation for safekeeping and the SDS was removed
by ultrafiltration for administrating in vivo.
Korean Patent No. 0507796 disclosed PEG-homodimer conjugates binding a biological
active polypeptide and a PEG, to increase the half-life in vivo of polypeptide. Particularly, it
described that the homodimer binds an amino group of lycine residues of bioactive
polypeptide of two molecules to a PEG to increase the residual time and sustain biological
activities for a long time. But, the conjugates have less activity than mono-PEG conjugates
because of conjugation of excessive PEGs and physicochemical and biological characteristics
of the conjugates are not uniform because of non-specific conjugation.
And, the method of binding rhG-CSF to linear SCM-MPEG(Succinimidyl carboxymethyl
ester of methoxy PEG) is known. The mono-PEG-G-CSF conjugates prepared by the
method have 3 types of position isomers modified at N-terminal, lycine 35, and lycine 45
residues. Particularly, the conjugate modified at lycine 35 has a disadvantage leaving PEGs
from the conjugate (Korean Patent No. 0248111, United State Patent No. 5,824,784).

United State Patent No. 5,951,974 disclosed conjugates of binding a linear SC-
PEG(polyethylene glycol-N-succinimide carbonate) to gene recombination alpha interferon.
The conjugates consist of covalent conjugates of urethane bond at e-amino group of lycine
residues of interferon, a-amino group of N-terminal cystaine residues, and an amino group of
histidine residues. But, the conjugates have a disadvantage leaving PEGs from the conjugate
because the urethane bond of conjugates are unstable.
Commonly used linear polyethylene glycol has a molecular weight of about 1,000~ 25,000
daltons, but has a limitation in binding many linear high molecules to protein or peptide, with
maintaining their activities, due to limited biological active regions of protein and peptide.
Korean Patent No. 0254097 disclosed conjugates binding a two-branched PEG of a lycine
skeletal structure to gene recombination alpha interferon. The conjugates have a merit to
prevent PEG from binding to multi parts of interferon, and have 2 times the molecular weight
of PEG by that of the linear PEG because two linear PEGs bind to a single part of interferon.
But, the two-branched PEG can be hydrolyzed to a single chain for safekeeping or for
reacting under an alkali condition because the two-branched PEG having a lycine skeletal
structure has two urethane bonds in the PEG.
Generally, it is known that biological activity, durability, etc. of conjugates of binding a PEG
to gene recombination protein is dependent on size and part modified in the PEG. But, it has
not been known yet whether a size of PEG in conjugates binding PEG and rhG-CSF affects
biological activities of a protein. Therefore, there has been a need in the art to overcome the
disadvantages of the known method through controlling a size of PEG and binding to various
skeletal structures of PEG not to decrease bioactivities of G-CSF and to increase the stability
of binding parts.

DISCLOSURE OF THE INVENTION
TECHNICAL PROBLEM
The object of the present invention is to provide three-branched polyethylene
glycol(PEG)-G-CSF conjugates prepared by conjugating three-branched PEG and G-CSF,
having high stability and same or more bioactivities of G-CSF than G-CSF and PEG-G-CSF
conjugates known in the art; a preparation method of the same; and a pharmaceutical
composition containing the same.
TECHNICAL SOLUTION
To achieve the above object, the present invention provides three-branched
polyethylene glycol(PEG)-G-CSF conjugates conjugated between three-branched PEG
derivatives and G-CSF.
The present invention also provides a method of preparing three-branched PEG-G-
CSF conjugates comprising conjugating three-branched PEG derivatives and G-CSF.
The present invention also provides a pharmaceutical composition comprising the
conjugates.
The present invention also provides a method of treating neutropenia, preventing
neutropenia, or facilitating increase of the number of neutrophil at the time of hematopoietic
stem cell mobilization to peripheral blood and hematopoietic stem cell transplatation,
comprising administrating the conjugate of the present invention as an effective ingredient.
The present invention is explained in detail below.
Three-branched PEG-G-CSF conjugates of the present invention may be prepared by
conjugating three-branched polyethylene glycoI(PEG), represented by the following general
formula(l):

wherein,
n is an integer of 1 to 1,000;
m is an integer of 10 to 1,000;
Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an
integer of 1 to 6;
Y is an amide bond formed by combining NH2 functional group in G-CSF and a
functional group of PEG derivative.
The PEG has an average molecular weight of from 200 to 45,000 daltons, preferably
from 20,000 to 45,000 daltons, more preferably from 23,000 to 43,000 daltons. Because a
medicinally effective time can be changed according to an average molecular weight of PEG,
the average molecular weight of PEG used in the present invention can be changed,
depending on the time required for treatment.
The method of preparing three-branched PEG-G-CSF conjugate comprising
conjugating three-branched PEG derivatives and G-CSF of the present invention is explained
in detail below.
The three-branched PEG-G-CSF conjugates of general formula (1) are prepared by
forming a covalent bond between three-branched PEG derivative of the following general
formula (2) and G-CSF wherein PEG has an average molecular weight of from 200 to 45,000
daltons, preferably from 20,000 to 45,000 daltons, more preferably from23,000 to 43,000
daltons:

wherein,
n is an integer of 1 to 1,000;
m is an integer of 10 to 1,000;
Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an
integer of 1 to 6; and
the functional group which can chemically react with proteins and peptides containing
G-CSF is N-hydrosuccinimide.
Three-branched PEG derivatives of the present invention activate high molecules
having branched structure in which three linear biological receptive high molecules are
combined. All of three OH (hydroxy) regions in the glycerol skeletal structure are
polymerized with ethylene glycol unit molecules, and the end of one region is activated as a
functional group. The other two regions except for the activated region are substituted with
monomethoxy, to prevent additional reactions. When the above branched PEG derivatives
are prepared, the size of each linear PEG can be controlled freely by a method known in the
art.
Three-branched PEG derivatives in the present invention can be used as PEG
derivatives, with the functional group chemically reacting with protein and peptide, known in
the art. PEG derivatives of general formula (2) (with N-hydroxysuccinaimde) are preferable
in terms of yield of the conjugate of the present invention.
The G-CSF in the present invention can be separated from mammalian organism or
synthesized by a method known in the art such as gDNA cloning, cDNA cloning, etc. And,

the G-CSF can be commercialized in the market.
And, the covalent bond between G-CSF and the three-branched PEG derivatives can
be formed at a low temperature. The reaction is completed by adding acids, and the prepared
three-branched PEG-G-CSF conjugates can be purified by a method known in the art such as
the purification method using a cation exchange resin.
In the present invention, the reaction molar ratio of the G-CSF to the three-branched
PEG derivative is from 1: 0.5 to 1: 50. A preferable molar ratio of the G-CSF to the three-
branched PEG derivative is from 1: 0.5 to 1: 5 because the yield of mono PEG-G-CSF
conjugate is decreased, as the molar ratio of polyethylene glycol to the G-CSF is increased.
The present invention also provides a pharmaceutical composition for treating or
preventing symptoms caused by decreased hematopoietic function or decreased immunologic
function. Particularly, clinical trials using the conjugate of the present invention as an
effective showed that the number of neuctrophil was decreased, and the symptoms were
relieved or controlled, for diseases caused in a treatment such as cancer chemotherapy or
radiation therapy; infective diseases caused by bacteria, virus and fungus; other infective
diseases; diseases caused by genetic or environmental reason such as severe chronic
neutropenia and leukemia; or geriatric diseases caused by again. For example, symptoms
caused by decreased hematopoietic function or decreased immunologic function are
neutropenia caused by cancer chemotherapy for blood tumor or solid cancer, neutropenia
caused by myelodysplastic syndrome, neutropenia caused by aplastic anemia, congenital
idiopathic neutropenia, and neutropenia caused from treating human immunodeficiency virus.
The composition can be composed of PEG-G-CSF conjugates of the present invention
or pharmaceutically acceptable diluent, antiseptics, solubilizer, emulsifier, adjuvant, and/or
carrier.
The compositions of the present invention can be formulated to injection agent, capsule,
tablet, liquid drug, pill, ointment, oculentum, coUyrium, transdermal absorptive agent, paste,
cataplasm, aerosols, etc.

And, the effective dosage of G-CSF is very small such as 0.1 -500 μg (preferably,
5-50 μg). And, the drug comprising 0.1 -500 μg of G-CSF may be administrated to an
adult, generally from 1 to 7 times a week. Thus, the effective dosage of the pharmaceutical
composition of the present invention can be calculated by a known administration amount of
G-CSF and a molecular weight of PEG used in the present invention. The effective dosage of
pharmaceutical composition of the present invention may be varied, but generally once a
week, and the composition can be administrated once or many times a day within a daily
effective dosage range.
The following examples are intended to further illustrate the present invention, and the
scope of the present invention is not intended to be limited thereby in any way.
ADVANTAGEOUS EFFECTS
The three-branched PEG-G-CSF conjugates of the present invention are more
pharmaceutically stable than linear- or two-branched PEG-G-CSF conjugates in the aspect of
leaving PEGs from PEG-G-CSF conjugates and forming the aggregates of conjugates. And,
the conjugates of the present invention can be used without additional separations of position
isomers because the conjugates consist of position isomers with similar activities. And, the
composition of the present invention has the effects of continuous production of neutrophilias
and increased in-vivo half-life.
DESCRIPTION OF THE DRAWINGS
Fig. 1 is a schematic drawing illustrating the analytic results of a mixture formed to
prepare conjugates of three-branched PEG having MW 23,000Da and G-CSF according to
Example 1, by size-exclusion high performance liquid chromatography.
Fig. 2 is a schematic drawing illustrating the analytic results of a mixture formed to
prepare conjugates of three-branched PEG having MW 43,000Da and G-CSF according to
Example 2, by size-exclusion high performance liquid chromatography.

Fig. 3 is a schematic drawing illustrating the analytic results of a mixture formed to
prepare conjugates of linear PEG having MW 10,000Da and G-CSF according to
Comparative Example 1, by size-exclusion high performance liquid chromatography.
Fig. 4 is a schematic drawing illustrating the analytic results of the purities of mono
three-branched PEG-G-CSF conjugates separated from the mixture of Example 1, by size-
exclusion high performance liquid chromatography.
Fig. 5 is a schematic drawing illustrating the analytic results of the purities of mono
three-branched PEG-G-CSF conjugates separated from the mixture of Example 2, by size-
exclusion high performance liquid chromatography.
Fig. 6 is a schematic drawing illustrating the analytic results of the purities of mono
linear PEG-G-CSF conjugates separated from the mixture of Comparative Example 1, by
size-exclusion high performance liquid chromatography.
Fig. 7 is a schematic drawing illustrating the analytic results of mono three-branched
PEG-G-CSF conjugates separated from the mixture of Example 1, by Matrix-Assisted Laser
Desorption Ionization (MALDI-TOF) Mass Spectrometry.
Fig. 8 is a schematic drawing illustrating the analytic results of mono three-branched
PEG-G-CSF conjugates separated from the mixture of Example 2, by Matrix-Assisted Laser
Desorption Ionization (MALDI-TOF) Mass Spectrometry.
Fig. 9 is a schematic drawing illustrating the analytic results of mono linear PEG-G-
CSF conjugates separated from the mixture of Comparative Example 1, by Matrix-Assisted
Laser Desorption Ionization (MALDI-TOF) Mass Spectrometry.
Fig. 10 is a schematic drawing illustrating the analytic results of the position isomers
comprising mono three-branched PEG-G-CSF conjugates separated from the mixture of
Example 1, by ion exchange chromatography.

Fig. 11 is a schematic drawing illustrating the analytic results of biological activity of
the position isomers of mono three-branched PEG-G-CSF conjugates separated from the
mixture of Example 1, according to Experimental Example 1, by the method of Experimental
Example 2.
MODE FOR INVENTION
Preparation of three-branched polyethylene glycol (PEG MW 23,000
Da)-G-CSF conjugates
10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of
sodium borate buffer solution. Then 9.2mg of three-branched PEG having N-
hydroxysuccinimde (NOF corporation, Japan) and the molecular weight of 23,000 daltons
was added to the solution in the molar ratio of rhG-CSF : three-branched PEG to 1: 0.75. And,
the solution was stirred at 4 °C for 90min. The reaction was stopped by adding lOOmM of
HCL to make the solution pH 4 and by diluting with sterile distilled water of 5 times the
volume of the solution. Then, the mixture was inputted in SP-Sepharose Fast Flow cation
exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium
acetate (pH4.0) buffer solution, and three-branched PEG-G-CSF conjugate was separated.
The mixture was fractioned by using 0 ~ IM of concentration gradient of sodium
chloride(NaCl). The molar ratio of three-branched PEG to G-CSF in the three-branched PEG-
G-CSF conjugates was confirmed by HPLC and SDS-PAGE. And, di- or more (tri-, tetra-
...) three-branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction,
etc. were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the
mono-three-branched PEG-G-CSF conjugate in which a G-CSF and a three-branched PEG
(MW 23,000Da) are conjugated (or called as mono-three-branched PEG-G-CSF conjugate).
It was confirmed that the reaction mixture consisted of 35.4% of mono-three-branched
PEG-G-CSF conjugate(mono-PEG-G-CSF conjugate), about 59.4% of G-CSF unmodified by
PEG, and the others [di-three-branched PEG-G-CSF conjugate(di-PEG-G-CSF conjugate),
and N-hydroxysuccinimide(NHS)], by a size-exclusion high performance chromatography
(see Fig. 1, wherein 1 and 2 represent di-PEG-G-CSF conjugate, 3 represents mono-PEG-G-
CSF conjugate, 4 represents G-CSF unmodified by PEG, and 5 represents NHS leaving from
three-branched PEG). And, the purity of mono-three-branched PEG-G-CSF separated from

the mixture was measured by size-exclusion high performance liquid chromatography (see
Fig. 4, wherein 2 represents di-PEG-G-CSF conjugate, and 3 represents mono-PEG-G-CSF
conjugate) and the molecular weight was measured by matrix assisted laser
desorption/ionization time-of-flight(MALDI-TOF) mass spectrometry (see, Fig. 7: the
analytic result of mono-three-branched PEG-G-CSF by MALDI-TOF/MS). And, the position
isomers were separated by cation exchange resin analytic column (SP-5WP, TOSOH) to
confirm that the ratio for each position isomer is about 54%: 33%: 10% (see. Fig. 10). And, It
was confirmed that each position isomer has very similar activity, in the experiment for
biological activity of each position isomer (see. Fig. 11)
Preparation of three-branched polyethylene glycol (PEG MW 43,000
Da)-G-CSF conjugates
10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of
sodium borate buffer solution. Then, 17.2mg of three-branched PEG with M-
hydroxysuccinimde (NOF corporation, Japan) and the molecular weight of 43,000 daltons
was added to the solution to make the molar ratio of the rhG-CSF : the three-branced PEG to
1: 0.75. And, the solution was stirred at 4 °C for 90min. And, the reaction was stopped by
adding 100mM of HCl to make the solution pH 4 and by diluting with sterile distilled water
of 5 times the solution volume. And, the mixture was inputted in SP-Sepharose Fast Flow
cation exchange chromatography (Amersham Pharmacia Biotech) equalized with 20mM of
sodium acetate(pH4.0) buffer solution, and three-branched PEG-G-CSF conjugate was
separated therefrom. The mixture was fractioned by using 0 ~ IM of concentration gradient
of sodium chloride(NaCl). The molar ratio of three-branched PEG to G-CSF in the three-
branched PEG-G-CSF conjugates was confirmed by HPLC and SDS-PAGE. And di- or
more (tri, tetra...)-three-branched PEG-G-CSF conjugates, unreacted G-CSF remaining after
the reaction, etc. were excluded. And, the eluates were concentrated and sterilized-filtered to
obtain the mono-three-branched PEG-G-CSF conjugate in which a G-CSF and a three-
branched PEG (MW 43,000Da) are conjugated(or called as mono-three-branched PEG-G-
CSF conjugate).
It was confirmed that the mixture of the reaction consisted of 18.7% of mono-three-
branched PEG-G-CSF conjugate (mono-PEG-G-CSF conjugate), about 40.1% of G-CSF

unmodified by PEG, and the others [di-three-branched PEG-G-CSF conjugate(di-PEG-G-
CSF conjugate), and N-hydroxysuccinimide(NHS)] by size-exclusionhigh performance
chromatography (see Fig. 2, wherein 2 represents di-PEG-G-CSF conjugate, 3 represents
mono-PEG-G-CSF conjugate, and 4 represents G-CSF unmodified by PEG). And, the purity
of mono-three-branched PEG-G-CSF separated from the mixture was measured by size-
exclusion high performance liquid chromatography (see Fig. 5, wherein 1 represents oligo-
PEG-G-CSF conjugate, 2 represents di-PEG-G-CSF conjugate, and 3 represents mono-PEG-
G-CSF conjugate), and the molecular weight was measured by matrix assisted laser
desorption/ionization time-of-flight(MALDI-TOF) mass spectrometry (see. Fig. 8; the
analytic result of mono-three-branched PEG-G-CSF by MALDI-TOF/MS).
Preparation of linear polyethylene glycol (PEG MW
10,000 Da)-G-CSF conjugates
10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of
sodium borate buffer solution. And, 5.2mg of linear PEG with N-hydroxysuccinimde (NOF
corporation, Japan) and the molecular weight of 13,000 daltons was added to the solution to
make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. And, the solution was stirred
at 4 °C for 90min. And, the reaction was stopped by adding 100mM of HCL to make the
solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume.
And, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography
(Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer
solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture
was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride(NaCl). The
molar ratio of linear PEG to G-CSF in the linear PEG-G-CSF conjugate was confirmed by
HPLC and SDS-PAGE. And, di- or more (tri, tetra...)-linear PEG-G-CSF conjugates,
unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were
concentrated and sterilized-filtered to obtain the mono-linear PEG-G-CSF conjugate in which
a G-CSF and a linear PEG(MW 13,000Da) are conjugated(or called as mono-linear PEG-G-
CSF conjugate).
It was confirmed that the mixture of the reaction consisted of 45.6% of mono-linear
PEG-G-CSF conjugate(mono-PEG-G-CSF conjugate), about 45.9% of G-CSF unmodified by

PEG, and the others [di-linear PEG-G-CSF conjugate(di-PEG-G-CSF conjugate), and N-
hydroxysuccinimide(NHS)], by a size-exclusion high performance chromatography (see Fig.
3, wherein 1 represents oligo-PEG-G-CSF conjugate, 2 represents di-PEG-G-CSF conjugate,
3 represents mono-PEG-G-CSF conjugate, and 4 represents G-CSF unmodified by PEG).
And, the purity of mono-linear-branched PEG-G-CSF separated from the mixture was
measured by size-exclusion high performance liquid chromatography (see Fig. 6, wherein 1
represents oligo-PEG-G-CSF conjugate, 2 represents di-PEG-G-CSF conjugate, 3 represents
mono-PEG-G-CSF conjugate, and 4 represents G-CSF unmodified by PEG), and the
molecular weight was measured by matrix assisted laser desorption/ionization time-of-
flight(MALDI-TOF) mass spectrometry (see, Fig. 9).
Preparation of linear polyethylene glycol (PEG MW
20,000 Da)-G-CSF conjugates
10 mg of rhG-CSF (Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of
sodium borate buffer solution, and then 8.0mg of linear PEG having N-hydroxysuccinimde
(NOF corporation, Japan) and the molecular weight of 20,000 daltons was added to the
solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. And, the solution
was stirred at 4 °C for 90min. The reaction was stopped by adding 100mM of HCL to make
the solution pH 4 and by diluting with sterile distilled water of 5 times the solution volume.
Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange chromatography
(Amersham Pharmacia Biotech) equalized with 20mM of sodium acetate(pH4.0) buffer
solution, and linear-branched PEG-G-CSF conjugate was separated therefrom. The mixture
was fractioned by using 0 ~ 1M of concentration gradient of sodium chloride (NaCl). The
molar ratio of linear PEG to G-CSF in the linear PEG-G-CSF conjugate was confirmed by
HPLC and SDS-PAGE. And, di- or more(tri, tetra...)-linear PEG-G-CSF conjugates,
unreacted G-CSF remaining after the reaction, etc. were excluded. And, the eluates were
concentrated and sterilized-filtered to obtain the mono-linear PEG-G-CSF conjugate in which
a G-CSF and a a linear PEG(MW 20,000Da are conjugated(or called as mono-linear PEG-G-
CSF conjugate).
Preparation of two-branched polyethylene glycol (PEG

MW 20,000 Da, lycine skeletal structure)-G-CSF conjugates
10 mg of rhG-CSF(Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of
sodium borate buffer solution, and then 9.1mg of two-branched PEG having N-
hydroxysuccinimde(Nektar, USA) and the molecular weight of 20,000 daltons was added to
the solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. And, the
solution was stirred at 4 °C for 90min. The reaction was stopped by adding 1 OOmM of HCL
to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution
volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange
chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium
acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated
therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium
chloride(NaCl). The molar ratio of two-branched PEG to G-CSF in the two-branched PEG-
G-CSF conjugate was confirmed by HPLC and SDS-PAGE. And, di- or more(tri, tetra.. .)-
two branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc.
were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono-
two-branched PEG-G-CSF conjugate in which a G-CSF and a two-branched PEG(MW
20,000Da are conjugated(or called as mono-linear PEG-G-CSF conjugate).
Preparation of two-branched polyethylene glycol (PEG
MW 20,000 Da, glyerine skeletal structure)-G-CSF conjugates
10 mg of rhG-CSF(Dong-A Pharm. Co., Ltd.) was filtrated with pH8.5 and 50 mM of
sodium borate buffer solution. And, 8.0mg of two-branched PEG having N-
hydroxysuccinimde(NOF, Japan) and the molecular weight of 20,000 daltons was added to
the solution to make the molar ratio of the rhG-CSF : the linear PEG to 1: 0.75. The solution
was stirred at 4 °C for 90min. And, the reaction was stopped by adding 100mM of HCL to
make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution
volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange
chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium
acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated
therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium
chloride(NaCl). The molar ratio of two-branched PEG to G-CSF in the two-branched PEG-
G-CSF conjugate was confirmed by HPLC and SDS-PAGE. And, di- or more(tri, tetra...)-

two branched PEG-G-CSF conjugates, unreacted G-CSF remaining after the reaction, etc.
were excluded. And, the eluates were concentrated and sterilized-filtered to obtain the mono-
two-branched PEG-G-CSF conjugate in which a G-CSF and a two-branched PEG(MW
20,000Da) are conjugated(or called as mono-linear PEG-G-CSF conjugate).
Preparation of mono-methoxypolyethylene glycol-G-CSF
conjugates binding at N-terminal a-amino residue
The title conjugate was prepared by a method described in Korean Patent No. 0248111,
and United State Patent No. 5,824,784. The method as briefly described is as follows.
100mM of sodium phosphate containing 20mM of NaCNBHs, and 5mg/ml of rhG-CSF
were stirred at 4 °C. And, 5 times mol
of linear methoxylpolyethylene glycol(mPEG) aldehyde(Nektar, USA) was added to
the solution, which was stirred for 10hs. The reaction was stopped by adding 100mM of HCL
to make the solution pH 4 and by diluting with sterile distilled water of 5 times the solution
volume. Then, the mixture was inputted in SP-Sepharose Fast Flow cation exchange
chromatography (Amersham Pharmacia Biotech) equalized with 20mM of sodium
acetate(pH4.0) buffer solution, and linear-branched PEG-G-CSF conjugate was separated
therefrom. The mixture was fractioned by using 0 ~ 1M of concentration gradient of sodium
chloride(NaCl). And, the eluates were concentrated and sterilized-filtered to obtain the mono-
mPEG-G-CSF conjugate in which a G-CSF and a two-branched PEG(MW 20,000Da) are
conjugated(or called as mono-linear PEG-G-CSF conjugate).
Preparation of mono-methoxypolyethylene glycol-G-CSF
conjugates binding at thiol group of cysteine residue of G-CSF
The title conjugate was prepared by a method described in Korean Patent No. 0508358.
The method as briefly described is as follows.
1 mg of rhG-CSF(Dong-A Pharm. Co., Ltd.) was added to 1mL of sodium phosphate
buffer solution(0.1M) having pH8.5, and 52.6mg of polyethylene glycol maleimide(NOF,
Japan) was added to the solution. And, the solution was stirred for 1h at room temperature.

Then, PEG derivatives remaining after the reaction were removed from the solution by
centricon 30(Amicon, USA). And, the solution was concentrated and sterilized-filtered to
obtain the mono-mPEG-G-CSF conjugate.
The property and pharmacological activity tests were conducted by using the
conjugates prepared above, and the results are as follows.
Analytic separation of position isomers of three-branched
PEG-G-CSF conjugates
100 μg of Mono-three-branched PEG-G-CSF conjugates of Example 1 was input to a
cation exchange resin analytic column(SP-5WP, TOSOH) equalized with 25mM of sodium
acetate buffer(pH 4.0, controlled with glacial acetic acid), and each position isomer of the
conjugates was separated by 100mM of sodium acetate buffer, pH 7.8). It was confirmed that
the ratio of each position isomer is about 54%: 33%: 10% (see. Fig. 10).
Analysis of biological activities of position isomers of
three-branched PEG-G-CSF conjugates
Relative Biological activities of position isomers of three-branched PEG-G-CSF
conjugates of Example 1 were measured by using that of G-CSF.
NFS-60 cell strain cultured at growth RPMI 1640 containing 10% FBS and 1ng/ml of
rmIL-3 was washed 3 times with test RPMI 1640 containing 5% FBS. And, the cell strain
was divided into 100ml(2xl05 cells/ml) at 96 well plate. Then the sample was diluted with
test medium to prepare a sample of 200ng/ml, and 9 more samples of 200ng/ml by 5 times
gradient, and the samples were divided into 3 wells, each 100ml, per concentration in 96well
plates containing the cell solutions.
Each well was cultured at 37°C and 5% CO2 incubator for 48hs, 40 μl of MTS was
added to each well. After 2hs, an absorbance of the sample was measured at 490nm by an
ELISA reader. EC50 was calculated by a dose response curve and a linear regression analysis
of point comprised straight line of a standard curve, and the activities of position isomers

were determined (see, Fig.l 1). The three-branched PEG-G-CSF conjugates of the present
invention have position isomers having similar activities. That is, it is not necessary to
remove some position isomers having relatively lower activity because the three-branched
PEG-G-CSF conjugates of the present invention are composed of conjugates with uniformed
activities
Pharmacokinetics test of the three-branched PEG-G-CSF
conjugates
The pharmacodynamic test was conducted to inject hypodermically each of G-CSF, the
PEG-G-CSF conjugates of Example 1, and the PEG-G-CSF conjugates of Example 2, into
test rats (Sprague Dawley) which had 240 ~ 260g of body weights. After injecting them in
the amount of 400μg per rat, the blood samples were collected from the rats at 0 min, 30min,
1hr, 2hr, 3hr, 4hr, 8hr, 12hr, 1day, 2days, 3days, 4days, 5days, 6days, and 7days, after the
injection. And, the concentrations of the G-CSF conjugates in the sample were
quantitatively analyzed by an enzyme-linked immunosorbent assay (ELISA) kit(Biosource,
USA).
The half-life in blood of the PEG-G-CSF conjugates of Example 1 and Example 2 was
increased 3 times and 10.8, respectively, compared with that of G-CSF, and the areas under
the concentration time curve in blood of the conjugates of Example 1 and Example 2 were
also increased by 7.6 times and 23.4 times, respectively, compared with that of G-CSF [Table
1 : Pharmacokinetics test of the three-branched PEG-G-CSF conjugates and G-CSF at
rat(Sprague Dawley rat)]. It was confirmed by this test that pharmacokinetics of PEG-G-CSF
conjugate is dependent on a size of PEG.

Stability test of the three-branched PEG-G-CSF
conjugates on leaving PEGs from the conjugates
PEG-G-CSF conjugates of Examples 1 and 2, and Comparative Examples 2, 3, and 4
were deposited under pH 7 and 37 °C for 7days, and the aspects of leaving PEGs in length of
time were analyzed by reversed-phase high-performance liquid chromatography, to estimate
the degrees of leaving PEGs on a skeletal structure, molecular weights, and branched types of
PEG.
In the PEG'S skeletal structure, the leaving PEGs occurred most frequently in
Comparative Example 2 using linear PEG. However, the leavings PEGs from the conjugates
of Examples 1 and 2, and Comparative Example 4 occurred less frequently than those of
Comparative Examples 2 and 3 (Table 2: Comparison of generation rates of G-CSF in PEG-
G-CSF conjugates having various PEG skeletal structures). This test confirmed that the three-
branched PEG-G-CSF conjugate is more stable than the linear-, two-branched PEG-G-CSF
conjugates, and the conjugate with a glycerol skeletal structure is more stable than the
conjugates with a iycine skeletal structure.

conjugates against forming aggregations according to a skeletal structure of PEG and a type
ofbinding PEG to G-CSF
Mono three-branched PEG-G-CSF conjugate of Example 1; mono PEG-G-CSF
conjugate of Comparative Examples 2 and 4, having a similar molecular weight to the
conjugate of Example 1 and different skeletal structure from the conjugate of Example 1; and
mono PEG-G-CSF conjugate of Comparative Example 5 having different binding type from
the conjugate of Example 1 were tested to measure stabilities against forming aggregations.
Each sample was collected by 500μl, and the solvent of the samples were exchanged 3 times
with 100mM of phosphate buffer solution having pH7. And, the samples were deposited at
37°C for 14 days, and formation of aggregation of the conjugates was analyzed by size-
exclusion high performance liquid chromatography.
The analysis results of forming of the aggregation represent that mono-PEG-G-CSF
conjugate of Example 1 had most stability in the aspect of leaving PEGs, and formed the least
aggregation (Table 3: Generation rates of aggregation conjugates according to a skeletal
structure of PEG and a type of binding PEG to G-CSF). Mono-PEG-G-CSF conjugates of
Comparative Examples 2 and 4 using PEG having the same glycerol skeletal structure as
PEGs in Example 1, and using the same binding type to Example 1, as well as the conjugate
of Example 1, formed some aggregation. However, mono-PEG-G-CSF conjugates of
Comparative Example 3 using PEG having different skeletal structure from PEGs in Example
1 (although using the same binding type as Example 1) formed much aggregation. And,
mono-PEG-G-CSF conjugates of Comparative Example 5 using PEG having different
skeletal structure from PEGs in Example 1 and using the different binding type from
Example 1 formed 2 times or more of aggregation, compared with the conjugates of Example 1.

In conclusion, mono-three-branched PEG-G-CSF conjugate of Example 1 is most
stable in both leaving PEGs from the conjugates and forming aggregation. That is, this
advantageous effect is from different skeletal structures between branched PEG and branched
PEG used in the present invention.
Neutrophil proliferation activities assay of PEG-G-CSF
conjugates at animal model to which an anti-cancer medicine is administrated.
Neutrophil proliferation activities of mono-three-branched PEG-G-CSF conjugates of
Example 1, mono-mPEG(methoxy polyethylene glycol)-G-CSF conjugates of Comparative
Examples 5 and 6 having different binding type and different molecular weight from the
conjugates of Example 2 were assayed by mouse model to which an anti-cancer agent was
administrated.
Mice(BDFl) of 20 ~ 25g were administrated with 200 mg/kg of an anti-cancer
agent(cyclophosphamide) to decrease the number of neutrophil for 2days.
And, each of the conjugate of Example I and the conjugates of Comparative Examples
5 and 6 was administrated subcutaneously once, 1mg/kg per mouse, and G-CSF was
administrated 8 times by 0.125 mg/kg per mouse to make the total administration amount 1
mg/kg. The mice' bloods were collected at 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and 12 days, and the
total numbers of leukocyte and neutrophil were measured by automatic hemocyte measuring
instrument (HEMAVET 850, Drew Scientific Ltd., UK). The periods that started the
recovery of neutrophils decreased by an anti-cancer agent and the times required for the
recovery of neutrophil, in the conjugate of Example 1 and the conjugates of Comparative
Examples 5 and 6 were similar. However, the time reaching the maximum number of
neutrophil in the conjugate of Example 1 was about 12hs faster than those of the conjugates
of Comparative Examples 5 and 6. The area under the increased neutrophil count-time
curve(AU-NC) of the conjugate of Example 1 was largest (Table 4: Neutrophil proliferation
activities assay of PEG-G-CSF conjugates at animal model to which an anti-cancer medicine
is administrated). Therefore, it was confirmed that mono-three-branched PEG-G-CSF
conjugate has a rapid effect of Neutrophil proliferation.
[Table 4]

t: Time for neutrophil to be increased after administration of a medicine
*: Time to reach l,000/mm3 or more of neutrophil after administration of an anti-cancer
medicine.

INDUSTRIAL APPLICABILITY
The three-branched PEG-G-CSF conjugates of the present invention are more
pharmaceutically stable than linear- or two-branched PEG-G-CSF conjugates in terms of
leaving PEGs from PEG-G-CSF conjugates and forming the aggregates of conjugates. And,
the conjugates of the present invention can be used without additional separations of position
isomers because the conjugates consist of the position isomers with similar activities. And,
the composition of the present invention has the effects of continuous production of
neutrophilias and increased in vivo half-life.

WE CLAIM:
1 . Three-branched PEG-G-CSF conjugate of the following general formula (1):
in which a bonding ratio of three-branched polyethylene glycol(PEG) and G-CSF is 1:1
(mol/mol), and PEG has an average molecular weight of from 200 to 45,000 daltons,
wherein,
n is an integer of I to 1,000;
m is an integer of 10 to 1,000;
Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an
integer of 1 to 6;
Y is an amide bond formed by combining NH2 functional group in G-CSF and a
functional group of PEG derivative
2. The conjugate of Claim 1, wherein the PEG has an average molecular weight of from
20,000 to 45,000 daltons.
3. The conjugate of Claim 1, wherein the PEG has an average molecular weight of from
23,000 to 43,000 daltons.
4. A method of preparing three-branched PEG-G-CSF conjugate of the general formula (1) of
Claim 1 comprising forming a covalent bond of three-branched PEG derivative of the
following, general formula (2) and G-CSF wherein PEG(polyethylene glycol) has an
average molecular weight of from 200 to 45,000 daltons;

wherein,
n is an integer of 1 to 1,000;
m is an integer of 10 to 1,000;
Z is (CH2)s or (CH2)sNHCO(CH2)s as a linker of G-CSF and PEG wherein S is an integer
of 1 to 6; and
the functional group which can chemically react with proteins and peptides containing G-
CSF is N-hydrosuccinimide.
5. The method of Claim 4, wherein the PEG has an average molecular weight of from 20,000
to 45,000 daltons.
6. The method of Claim 4, wherein the PEG has an average molecular weight of from 23,000
to 45,000 daltons.
7. The method of Claim 4, wherein the molar ratio of G-CSF to three-branched PEG
derivative is from 1: 0.5 to 1: 50.
8. The method of Claim 7, wherein the molar ratio of G-CSF to three branched PEG
derivative in the reaction is from 1: 0.5 to 1: 5.
9. A pharmaceutical composition for treating or preventing symptoms caused by decreased
hematopoietic function and decreased immunologic function, comprising the conjugate of
any of claim 1, 2 or 3 as effective ingredient.
10. The pharmaceutical composition of claim 9, wherein the symptoms caused by decreased

hematopoietic function and decreased immunologic function are solid cancer, neutropenia
caused by cancer chemotherapy to blood tumor, neutropenia caused by myelodysplastic
syndrome, neutropenia caused by aplastic anemia, congenital idiopathic neutropenia, and
neutropenia caused by treating human immunodeficiency virus.
1. A method of treating neutropenia, preventing neutropenia, or facilitating increase of the
number of neutrophil at the time of hematopoietic stem cell mobilization to peripheral
blood and hematopoietic stem cell transplatation, comprising administrating the conjugate
of any of claim 1, 2 or 3 as an effective ingredient.

The present invention relates to three-branched PEG-G-CSF conjugate of general formula (1) in which the bonding ratio of three-branched polyethylene glycol(PEG) and G-CSF is 1:1 (mol/mol), wherein PEG has an average molecular weight of from 200 to 45,000 daltons; a pharmaceutical composition comprising the same, and a preparing method thereof.

A PHARMACEUTICAL COMPOSITION HAVING STRUCTURE OF THERE-BRANCHED PEG-G-CSF CONJUGATE COMPRISING THREE-BRANCHED PEG AND G-CSF AND METHOD OF PREPARING THE SAME

Documents:

Inventors:

PCT Conventions: