(en)The invention relates to derivatives of recombinant proteins, comprising homo-multimers of genetically fused recombinant biologically active protein monomer units, connected via selected peptide linker moiety; and the method of preparation thereof. Derivative of recombinant protein is preferably dimer of human granulocyte colony-stimulating factor, characterised by increased circulation time in vivo.

1.ApplicationNumber: US-201114237807-A
1.PublishNumber: US-2014193357-A1
2.Date Publish: 20140710
3.Inventor: MISTINIENE EDITA
PESLIAKAS JONAS HENRIKAS
PLECKAITYTE MILDA
ZVIRBLIS GINTAUTAS

4.Inventor Harmonized: MISTINIENE EDITA(LT)
PESLIAKAS JONAS HENRIKAS(LT)
PLECKAITYTE MILDA(LT)
ZVIRBLIS GINTAUTAS(LT)

5.Country: US
6.Claims:
(en)The invention relates to derivatives of recombinant proteins, comprising homo-multimers of genetically fused recombinant biologically active protein monomer units, connected via selected peptide linker moiety; and the method of preparation thereof. Derivative of recombinant protein is preferably dimer of human granulocyte colony-stimulating factor, characterised by increased circulation time in vivo.

7.Description:
(en)FIELD OF THE INVENTION
This invention belongs to the field of protein biotechnology; actually, the present invention provides one of the solutions to the problem of increasing in vivo circulation time of recombinant proteins of therapeutic values. More particularly, the present invention relates to the possibility of preparation of genetically fused homo-multimers of interferon a-2a or granulocyte colony-stimulating factor with extended circulation half-life, comprising at least two monomer units of the protein genetically connected via the linkers of defined length, and efficient isolation and purification of biologically active G-CSF dimers produced. The present invention also relates to a synthetic gene coding for dimer form of human G-CSF containing linker of defined length between G-CSF monomer units for the expression in host cells.
BACKGROUND OF THE INVENTION
In this application the term “protein of therapeutic value” means pharmacologically active protein, suitable for use in therapy and produced through genetic engineering, including mammalian antibodies, blood product substitutes, vaccines, hormones, cytokines. Potential therapeutic proteins are different in a number of important aspects from classical new drug entities, which are generally low-molecular-weight biologically active compounds. To be efficacious, protein drugs are used in concentration and circumstances which differ markedly from their native counterparts and this may lead to undesirable effects in vivo. To avoid or minimise toxicity and increase efficacy both the physicochemical and biological properties of proteins are being altered generally by some form of protein modification e.g., covalent conjugation with other macromolecules, antibody binding, mutagenesis and glycosylation.
The term “biologically active protein” means the protein molecule which exhibit the same detectable biological activity as the respective naturally existing protein.
The term “monomer unit” means a single polypeptide chain of naturally occurring biologically active protein.
The term “homo-multimer” means a linear polypeptide chain consisting of more than one identical biologically active polypeptide subunits, which are connected to each other via peptide linker molecule in a such manner that connection of polypeptide subunit through disulfide bonds is excluded.
The term “genetically fused” means that homo-multimer protein is obtained using recombinant DNA methodology: construction of artificial DNA fragment consisting of two or more connected genes (coding sequences) of monomer proteins via linker DNA sequence, introduction of the DNA fragment into the vector and expression of the protein in the selected cell type. The recombinant protein consists from monomer proteins connected with linker peptide sequences (protein could be isolated after applying special purification procedure) in contrast with “chemically conjugated” protein that is obtained by chemical joining of two or more monomer proteins using specific chemical methods and “chemically modified” protein that is obtained by chemical modification of monomer protein using chemical agents or polymer residues.
Advances in biochemistry, protein chemistry and molecular biology over the last twenty-five years have spurred the increased use and development of recombinant proteins as injectable therapeutic agents. Protein and peptide biopharmaceuticals have been successfully used as very efficient drugs in therapy of many pathophysiological states since the first recombinant product insulin was approved in 1982. One group of approved first generation protein biopharmaceuticals mimics native proteins and serves as replacement therapy, while another group represents monoclonal antibodies for antagonist therapy or activating malfunctioning body proteins [Jev{hacek over (s)}evar S. et al., PEGylation of therapeutic proteins, Biotechnol. J., 5, 113-128 (2010)]. The main drawbacks of the first-generation biopharmaceuticals are their suboptimal physicochemical, pharmacokinetic and pharmacodynamic (PK/PD) properties. Main limitations are physicochemical instability, limited solubility, proteolytic instability, relatively short elimination half-life, immunogenicity and toxicity. Consequently, protein therapeutics are mainly administrated parenterally [Jev{hacek over (s)}evar S. et al., Biotechnol. J., 5, 113-128 (2010)].
Wide range of technologies have been developed during the last decade directed to the development of second-generation biopharmaceuticals encompassing the main products of first-generation protein drugs by conferring to them improved PK/PD characteristics. Extension of in vivo circulating half-life time is the main goal of such therapeutics. Long-acting forms of erythropoietin (EPO, containing two N-linked oligosaccharide chains, under the name of Aranesp), granulocyte colony-stimulating factor (pegylated-G-CSF under the name of Neulasta) and interferons (pegylated IFN alfa-2b under the name of PEG-Intron or pegylated-IFN alfa-2a under the name of Pegasys) are successful examples of second-generation biopharmaceuticals which gained the status of blockbuster drugs.
One of widely used method to prolong plasma half-life time of the target therapeutic protein is chemical modification focusing on increasing the size of therapeutic protein by conjugation with natural or synthetic polymers using well established procedures known as PEGylation (chemical conjugation with polyethylene glycol, Veronese F. M. et al., Protein PEGylation, basic science and biological application in PEGylated Protein Drugs: Basic Science and Clinical Applications, ed. F. M. Veronese, 11-31(2009), polysialylation (chemical conjugation with polysialic acid, Sanjay Jain et al. Polysialylation: The natural way to improve the stability and pharmacokinetics of protein and peptide drugs. dds&s Vol 4 No 1 May (2004), HESylation (chemical conjugation with hydroxyethylstarch, International Patent Application WO 02/080979 and International Patent Application WO 03/000738) and others.
Another method of modification of clearance of therapeutic proteins is through chemical cross-linking or genetic modification to fuse proteins of interest with long-living plasma proteins like albumin, immunoglobulin, or portions of these proteins (Sheffield, Modification of clearance of therapeutic and potentially therapeutic proteins, Curr. Drug Targets Cardiovasc Haematol Disord., 1,1-22. (2001).
Modification of N- and C-terminus of a therapeutic protein or replacement of amino acids which are known to be susceptible for enzymatic cleavage is also used as a strategy to reduce immunogenicity and proteolytic instability and therefore to improve plasma half-life time (Werle M and Bernkop-Schmurrch, Amino acids, 30, 351-367 (2006).
However, all these modifications often cause significant reduction of the biological activity of the protein of interest or elicit antibody formation. Due to this a relatively limited range of proteins with improved plasma half-life time are in clinical use. Therefore, search of new means for alteration of circulation half-life of therapeutic protein remains a challenging task of current biotechnology.
Patent U.S. Pat. No. 5,580,853 describes methods of preparing multimeric erythropoietin derivatives comprising two or more erythropoietin molecules covalently linked together by one or more thioether bond(s). These erythropoietin multimers exhibit increased biological activity and prolonged circulation time. In this method a first erythropoietin derivative was produced by chemically reacting wild type erythropoietin with a hetero-bifunctional cross-linking reagent containing a cleavable disulfide bond group. The disulfide bond was reduced to produce erythropoietin containing a free sulfhydryl group. A second erythropoietin derivative was produced by reacting wild type erythropoietin with a hetero-bifunctional cross-linking reagent containing a maleimido group. The first and second erythropoietin derivatives were reacted together, thereby forming at least one thioether bond between the sulfhydryl and maleimido groups, thus forming a homodimer or homotrimer of erythropoietin. These multimeric erythropoietin molecules exhibit biological activity comparable to wild type erythropoietin and prolonged circulating half-life in vivo, relative to wild type erythropoietin. However, chemical conjugation of protein molecules, e.g., erythropoietin into multimeric form may be accompanied by non-specific chemical modification via functional group of amino acid residue participating in receptor binding. This could decrease the biological activity of the multimeric product. In general, chemical modification may generate unfavourably linked products which must be separated from the correctly linked target product and other by-products. Such modification process is expensive and requires additional purification process steps trying to obtain desired product with reproducible activity and quality characteristics.
EP1334127 discloses single-chain multimeric polypeptides comprising at least two units of a monomeric polypeptide covalently linked via a peptide bond or a peptide linker. Monomeric polypeptide belongs to the protein type that is biologically active in monomeric form. Here, at least one monomeric unit of the construct differs from the corresponding wild-type monomeric polypeptide by at least one added or removed amino acid residue (Lys, Cys, Asp, Glu or His) which serves as an attachment site for further chemical conjugation with non-polypeptide type moiety. The polypeptide is preferably a G-CSF dimer with attached polymer molecule, preferably polyethylene glycol molecules. This patent does not provide any information on the impact of peptide linker on the biological activity of multimeric peptides.
More attractive approach related to production of biologically active multimeric protein derivatives is described in U.S. Pat. No. 5,705,484 that discloses biologically active multimeric polypeptide molecule in which two or more monomeric subunits are linked together as a single polypeptide and prepared as genetically fused multimer. The fusion multimers specifically include PDGF fusion dimers in which protein monomer units are linked via spacer moiety selected from the pre-pro region of a PDGF precursor protein. These fusion multimers are more easily and rapidly refolded than unfused multimers and eliminate the simultaneous formation of undesirable polypeptide by-products during refolding. However, the use of pre-pro sequence as a linker is not favourable due to possible digestion by proteases which commonly participated in the processing of pre-pro domains of a protein molecule.
WO 01/03737 discloses fusion proteins comprising a cytokine or growth factor fused to an immunoglobulin domain, in particular IgG. Also disclosed are multimeric fusion proteins comprising two or more members of the growth hormone superfamily joined with or without a peptide linker. The peptide linker is SerGly, Ser(GlyGlySer) n , wherein n is 1 to 7 or sequences Ser(GlyGlySer) or Ser(GlyGlySer) 2 . However examples are only provided for fusions with IgG which are regarded as hetero-multimeric fusion proteins.
For separation domains of a bifunctional fusion proteins several lengths of helix-forming peptides linkers with sequence A(EAAAAK) n A (n=2-5) were introduced (R. Arai et all., Prot. Eng., 14 529-532 (2001) showing that such linkers allow controlling the distance and reduce the interference between the domains. This approach is used for separation of different domains of the same type of proteins.
WO2005/034877 describes a polypeptide comprising a G-CSF domain linked to a transferrin domain through Leu-Glu linker. Such fusion can be used for treating various diseases as orally delivered pharmaceutical, but transferrin domain activity is very poor.
Patent U.S. Pat. No. 7,943,733 discloses the role of linker introduced between the partners of hetero-multimers of transferrin-based fusion proteins demonstrating that the insertion of alpha-helical linker between two fusion partners increase the expression level of the fusion. LEA(EAAAK) 4 ALE sequence between the partner of the fusion comprising a carrier protein (transferrin, serum albumin, antibody and sFV) and therapeutic protein (G-CSF, an interferon, a cytokine, a hormone, a lymphokine, an interleukin, a hematopoietic growth factor and a toxin). However, this patent is dedicated for the production of hetero-fusion construct where protein partners are different protein molecules, representing different activities.
A program to generate all possible linker sequences for fusion proteins has been recently published (C. J. Crasto and J. a. Feng, Prot. Eng., 13, 309-312 (2000). It serves as a tool for rational design of desired linker sequence and its length. However this program could not predict the biologic activity of the final fusion proteins due to complexity of the factors enabling the active structure of the fusion proteins. So, only experimental testing of the different selected linker structures is capable to give the evidence of protein activity.
Thus, among the existing methods for production of long-acting biopharmaceuticals a method of linking monomer unit of the protein of interest into multimers is less time-consuming and allowing to protect multimer construction from the lost of biological activity. However, known method of chemical conjugation of protein monomer into homo-multimers albeit protected biological activity and prolonged circulating half-life did not assure reproducibility of multimer quality parameters.
Method for production of single-chain multimeric polypeptides by introduction into one monomeric unit of amino acid residue for subsequent attachment of PEG in principle mimics well-known methodology of therapeutic protein PEGylation with the only difference that dimer form of the protein instead of monomer is used.
With regards to this, the present invention proposes a method for production of biopharmaceuticals with increased circulation half-life, using genetic fusion of protein monomer units into homo-multimer construct via linkers of defined length. Sequence and structure of the linker is designed to achieve accessibility of each monomer unit to interact with specific receptor of selected monomer of therapeutic protein. Owing to this, no further attachment of polymer, e.g. PEG molecules is required.
Therefore, it is an object of the present invention to provide a multimeric form of therapeutic proteins of interest having biologically active monomer units and improved PK/PD characteristics.
It is a further object of the present invention to provide a multimeric form of therapeutic proteins that can be produced via recombinant DNA technology without formation of undesirable dimer and oligomer forms cross-linked via disulfide bonds.
When molecular mass of the recombinant protein is increased the technologic problems frequently multiply during preparation of the target protein to the purity level in compliance with the requirements for pharmaceutical substance. On the other side the increase of molecular mass of the target protein is associated with the risk of increased propensity for aggregation especially when high expression level of the protein leads to its accumulation in insoluble form (inclusion bodies). In this case technology approaches applied for preparation of monomer protein are no more suitable for isolation and purification of multimeric proteins. Here serious attention should be paid for the proper selection not only of each process stage but also for the compatibility of whole process steps assuring as high as possible biologic activity of the protein with its increased molecular mass at the end of purification cycle. So, this proves the necessity of a new approaches to the preparation of multimeric proteins.
SUMMARY OF THE INVENTION
The present invention provides a new approach for increasing in vivo circulation time of recombinant proteins of therapeutic values, using directed multimerization procedure by genetic fusion of target protein polypeptide chain connected via linker of defined length into dimer, trimer or multimer of desirable length.
Derivatives of recombinant proteins according to present invention comprise homo-multimers of genetically fused recombinant biologically active protein monomer units, connected via suitable peptide linker moiety, wherein protein monomer unit has amino acid sequence of native biologically active protein or essentially the same sequence of at least 95% identity. The suitable linker moiety has an amino acid sequence, selected from the group, consisting of (S-G 4 ) n -S, wherein n=2-7, and SGLEA-(EAAAK) m -ALEA-(EAAAK) m -ALEGS, wherein m=2-8. In preferred embodiment of present invention said suitable linker moiety has an amino acid sequence, selected from the group, consisting of SEQ ID No:1, SEQ ID No:2, SEQ ID No:3, SEQ ID No:4 and SEQ ID No:5 below.
In the main aspect of present invention said homo-multimer contains single chain of at least two protein monomer units and said biologically active protein monomer unit is selected from the group, consisting of cytokines, growth factors and hormones, wherein cytokines comprise interleukines, colony stimulating factors or interferons, preferably granulocyte colony stimulating factor (G-CSF) or interferon.
In one preferable embodiment of present invention said biologically active protein monomer unit is recombinant interferon alpha-2a, homo-multimer is dimer, preferably having amino acid sequence SEQ. ID No:6 below.
In another preferred embodiment said biologically active protein monomer unit is recombinant human granulocyte colony-stimulating factor (rhG-CSF).
More specifically homo-multimers of granulocyte colony-stimulating factor according to present invention comprise of least two genetically fused recombinant granulocyte colony-stimulating factor protein monomers, connected via suitable linker moiety defined. Preferred homo-multimer of granulocyte colony-stimulating factor according to present invention is dimer of granulocyte colony-stimulating factor, having amino acid sequence, selected from the group, consisting of SEQ ID No:7, SEQ ID No: 8, SEQ ID No: 9, SEQ ID No:10 and SEQ ID No: 11.
Homo-multimers of granulocyte colony-stimulating factor according to present invention are characterised by increased circulation time in vivo.
Yet another preferred embodiment is homo-multimers of granulocyte colony-stimulating factor for use in therapy. More specifically, homo-multimers of granulocyte colony-stimulating factor according to present invention are intended for the manufacture of the medicament (pharmaceutical substance) for treating of diseases and conditions in indications where monomeric granulocyte colony-stimulating factor protein is used. The related object of present invention is pharmaceutical composition, comprising an therapeutically effective amount of homo-multimer of granulocyte colony-stimulating factor above in combination with pharmaceutically acceptable carrier, diluent, excipient and/or auxiliary substances.
Second main aspect of present invention is a method for preparation of homo-multimers of granulocyte colony-stimulating factor, comprising the steps of:
a) cultivation of microorganism of producer which host cell comprise a nucleotide sequence encoding the target genetically fused protein in a suitable culture medium under conditions, permitting expression of the target proteins;
b) lysing the microorganism of producer and separating the fraction of insoluble target proteins;
c) solubilizing of the fraction of insoluble proteins;
d) renaturation of the target protein;
e) chromatographic purification of the target protein.
The main distinguishing features of the method according to present invention are:
solubilising of the fraction of insoluble proteins is performed in a buffer system, containing urea or guanidine hydrochloride as chaotropic agent; renaturation is oxidizing renaturation of the target protein from inclusion bodies in the presence of a pair reduced/oxidized glutathione, or metal ions in solution, or immobilized metal ions on chromatography support, and purification by chromatography consisting of the purification by immobilized metal ions affinity chromatography, and further additional purification by at least one ion exchange and/or gel-filtration chromatography, or by combination of the ion-exchange chromatography with sequential use of anion-exchange and cation-exchange chromatography.

Finally, dimer of granulocyte colony-stimulating factor according to present invention comprises one, produced by the method according to present invention.
The invention is described below in further details and with reference to the accompanying drawings and examples.



BRIEF DESCRIPTION OF THE DRAWINGS
The preparation of derivatives of recombinant proteins according to present invention and some properties thereof are illustrated by FIGS. 1-7 .
FIG. 1 shows SDS-PAGE (15%) of dimeric IFN-alpha-2a during purification on SP-Sepharose Fast Flow in reducing conditions (Coomassie staining).
Lanes:
M—MW markers from the top: 116; 66,2; 45,0; 35,0; 25; 18,4; 14,4 KDa
1-4—Fractions of desorption of dIFN-2a in ion exchange chromatography on SP-Sepharose FF.
FIG. 2 illustrates SDS-PAGE (15%) of purified dimeric G-CSF-La under non-reducing conditions (silver staining).
1-6 lanes—1 μg, 0.5 μg, 0.25 μg, 0.12 μg, 0.06 μg of dimeric G-CSF-La respectively.
MW—molecular weight standards from the top: 250, 130, 100, 70, 55, 35, 25, 15, 10 kDa
FIG. 3 represents Western blot analysis of purified dimeric G-CSF, using monoclonal antibodies against monomeric G-CSF.
Lanes:
1, 5—1 μg monomeric G-CSF in reducing and non-reducing conditions respectively;
2, 6—1 μg dimeric G-CSF with linker La in reducing and non-reducing conditions respectively;
3, 7—1 μg dimeric G-CSF with linker L5 in reducing and non-reducing conditions respectively; 4, 8—1 μg dimeric G-CSF with linker L7 in reducing and non-reducing conditions respectively;
K—molecular weight standards from the bottom: 15, 25, 35, 40, 55, 70, 100 kDa.
FIG. 4 evidences purity evaluation of dimeric G-CSF-La by RP-HPLC, using C4 column ((Hi-Pore RP-304, 250×4.6 mm, 300 Å, “Bio-Rad”) and mobile phases: A (0.1% trifluoracetic acid (TFA) aqueous solution) and B (TFA-H 2 O-acetonitrile 0.1:9.9:90). Loaded amount of protein 50 μg. Detection wave length 215 nm.
FIG. 5 shows purity evaluation of dimeric G-CSF-La by SEC-HPLC, using column TSK-gel G2000 SWXL (7,8×300 mm “Tosoh Bioscience”). Mobile phase 50 mM sodium phosphate buffer, pH 7.2 with 150 mM NaCl. Loaded amount of protein 14 μg. Detection wave length 215 nm.
FIG. 6 shows concentration of dimeric G-CSF in rat blood serum after subcutaneous (s.c.) injection (500 μg/kg):
A—dGCSF-L5 and Neupogen™ control;
B—dGCSF-L5, dGCSF-L7, dGCSF-La, Neupogen™ control and storage buffer without target protein.
FIG. 7 demonstrates the influence of dimeric G-CSF in vivo on the number of neutrophils in rats after s.c. injection (500 μg/kg). Control probes of Neupogen™ and storage buffer without target protein.



DETAILED DESCRIPTION OF THE INVENTION
Recently (patent U.S. Pat. No. 5,580,853 “‘Modified polypeptides with increased biological activity”’ inventor—A. J. Sytkowski) it has been reported that chemical cross-linking of erythropoietin protein monomer unit into dimer with hetero-bifunctional cross-linking reagent allow to retain biological activity of obtained dimer comparable to wild type erythropoietin and that the erythropoietin dimer showed a significantly prolonged circulating time in vivo, relative to wild type erythropoietin. Chemical modification frequently generates contaminating products which must be separated from the correctly linked target product. Such modification process is expensive and requires additional purification process steps trying to obtain desired product with reproducible activity and quality characteristics.
The inventors of the present invention have discovered that the genetic fusion of therapeutic protein macromolecule into homo-multimer, particularly homo-dimer connecting monomer units via properly selected sequence and the length of linker molecule, allows to achieve biologically active dimer with enhanced in vivo circulation time. Protein macromolecule to be involved into proposed according to this invention procedure of homo-multimerization encompasses, but is not limited to any therapeutic protein, such as a colony stimulating factor, an interferon, a cytokine, a hormone, an interleukin, a growth factor, antibody fragments and other proteins.
Genetic fusion of therapeutic protein molecule into homo-multimer is achieved with the use of properly selected peptide linker moiety, introduced between monomer units of homo-multimer construction. Linkers suitable for selection are shown in Table 1. Linker molecules of the sequence (S-G 4 ) n -S, wherein n=2-7 and/or alpha-helical structured sequence of SGLEA-(EAAAK) m -ALEA-(EAAAK) m -ALEGS, wherein m=2-8, are essential linker molecules exemplified in this invention by producing biologically active homo-dimer of granulocyte colony-stimulating factor, which exhibits about two-three time longer circulation time relative to G-CSF monomer, such as Filgrastim (Neupogen™)





TABLE 1



Linkers used for dimeric proteins






Sequence


Number of

No:
Label
Linker sequence
amino acids



SEQ ID No: 1
L2
(SG 4 )-(SG 4 )-S
11



SEQ ID No: 2
L3
(SG 4 )-(SG 4 )-(SG 4 )-S
16



SEQ ID No: 3
L5
(SG 4 )-(SG 4 )-(SG 4 )-(SG 4 )-(SG 4 )-S
26



SEQ ID No: 4
L7
(SG 4 )-(SG 4 )-(SG 4 )-(SG 4 )-(SG 4 )-(SG 4 )-(SG 4 )-S
36



SEQ ID No: 5
Lα
SGLEA-(EAAAK) 4 -ALEA-(EAAAK) 4 -ALEGS
54






























The homo-multimers of therapeutic proteins may be expressed in host cells such as bacteria, e.g. E. coli BL21(DE3); yeast; higher eukaryotic cells, e.g. CHO, and others. More preferably, E. coli host cells are used for expression of homo-multimers of target therapeutic proteins. Bacteria are excellent producer of different recombinant proteins especially in case of the proteins without additional modifications of the amino acids. Cheap biosynthesis, sufficient yields of recombinant proteins, inability to be infected with mammalian viruses and other infective materials of mammalian origin (if media used is free from additives of mammalian origin) is important advantages of such producers. The strain E. coli BL21(DE3) in combination with effective vector pET21b used for illustrating of current invention are able to produce sufficient amounts of target multimeric protein, though recombinant proteins are frequently accumulated as insoluble aggregated and inactive particles. The purification and refolding step of multimeric proteins is critical stage of this invention to obtain biologically active and stable material for further steps.
E. coli produced homo-dimers of recombinant human G-CSF are accumulated within the cells as inclusion bodies (IB). With regards to this the procedures for isolation of IB, their purification by established washing procedures and further solubilization in a buffer system containing selected chaotropic agent has been elaborated.
In order to obtain acceptable yield of finally purified target homo-dimer of G-CSF the inclusion body proteins according to preferred embodiment of present invention were reduced during solubilization step in the presence of selected concentration of dithiothreithol (DTT). Proposed according to the present invention process for isolation and purification of homo-dimer of G-CSF comprises solubilization of pre-washed IB in a buffer containing 8 M urea, further dilution of IB solubilizate to defined range of protein concentration and urea concentration of 3 M, oxidative refoding of the protein by one type of established procedure using additives of Cu ions or a pair of reduced SH-agent, such as DTT or glutathione (GSH) and oxidizing agent, such as oxidized glutathione (GSSG) with the use of established concentration of each agent.
More specifically the method according to present invention is characterised in that renaturation of the target protein is performed by any of alternative procedures:
a) in the presence of CuSO4 by mixing at ambient temperature for 1-3 hours, preferably 2 hours, and stopping refolding by adding EDTA solution at pH 5.4±0.2 to final concentration of 5-15 mM, preferably 10 mM; or b) in the presence of a pair reduced(GSH)/oxidized(GSSG) glutathione taken in the concentration ratio 5-10:1 by mixing overnight or preferably for 24-96 hours at the temperature +2-+8° C.; or c) in the presence of 0.08-0.12 mM GSSG by mixing overnight or preferably for 24-96 hours at +2-+8° C.; or d) in the presence of GSSG in such manner that final molar ratio in solution of DTT/GSSG is 1: 1,5-10 by mixing overnight or preferably for 24-96 hours at +2-+8° C.

The refolded target protein is further subjected to chromatography purification step consisting of anion-exchange chromatography of urea-containing refolded protein solution over DEAE-Sepharose FF column. Elution of the target correctly folded homo-dimer of G-CSF has been elaborated with respect to recovery yield. Further, the combined eluate of fractions recovered from DEAE-Sepharose FF column is chromatographied over strong cation-exchanger as SP-Sepharose FF chromatography media. Finally purified homo-dimer of G-CSF is formulated as liquid formulation containing pre-defined concentration of buffer substance, tonicity adjusting agent and non-ionic detergent with its selected type and concentration range sufficient to avoid protein damage at the air/solution interface.
The structure of constructed and expressed dimeric proteins is confirmed using few independent methods such as sequencing of recombinant expression plasmid with subsequent translation of dGCSF gene into protein amino acid sequence and mass-spectrometry analysis of purified proteins (see Table 2).
Table 2
Sequence and molecular mass of dimeric proteins












Molecular
Full amino acid

Sequence No
Dimeric protein
mass, kDa
sequence of dimeric protein



SEQ ID No: 6
dIFN-2a_L3
39,63
MCDLPQTHSLGSRRTLMLLAQMRKISLFSC






LKDRHDFGFPQEEFGNQFQKAETIPVLHEM






IQQIFNLFSTKDSSAAWDETLLDKFYTELYQ






QLNDLEACVIQGVGVTETPLMKEDSILAVRK






YFQRITLYLKEKKYSPCAWEWRAEIMRSF






SLSTNLQESLRSKESGGGGSGGGGSGGG






GSCDLPQTHSLGSRRTLMLLAQMRKISLFS






CLKDRHDFGFPQEEFGNQFQKAETIPVLHE






MIQQIFNLFSTKDSSAAWDETLLDKFYTELY






QQLNDLEACVIQGVGVTETPLMKEDSILAV






RKYFQRITLYLKEKKYSPCAWEVVRAEIMR






SFSLSTNLQESLRSKE



SEQ ID No: 7
dGCSF-L2
38,17
MTPLGPASSLPQSFLLKCLEQVRKIQGDGA






ALQEKLCATYKLCHPEELVLLGHSLGIPWA






PLSSCPSQALQLAGCLSQLHSGLFLYQGLL






QALEGISPELGPTLDTLQLDVADFATTIWQQ






MEELGMAPALQPTQGAMPAFASAFQRRAG






GVLVASHLQSFLEVSYRVLRHLAQPSGGG






GSGGGGSTPLGPASSLPQSFLLKCLEQVR






KIQGDGAALQEKLCATYKLCHPEELVLLGH






SLGIPWAPLSSCPSQALQLAGCLSQLHSGL






FLYQGLLQALEGISPELGPTLDTLQLDVADF






ATTIWQQMEELGMAPALQPTQGAMPAFAS






AFQRRAGGVLVASHLQSFLEVSYRVLRHLA






QP



SEQ ID No: 8
dGCSF-L3
38,55
MTPLGPASSLPQSFLLKCLEQVRKIQGDGA






ALQEKLCATYKLCHPEELVLLGHSLGIPWA






PLSSCPSQALQLAGCLSQLHSGLFLYQGLL






QALEGISPELGPTLDTLQLDVADFATTIWQQ






MEELGMAPALQPTQGAMPAFASAFQRRAG






GVLVASHLQSFLEVSYRVLRHLAQPSGGG






GSGGGGSGGGGSTPLGPASSLPQSFLLKC






LEQVRKIQGDGAALQEKLCATYKLCHPEEL






VLLGHSLGIPWAPLSSCPSQALQLAGCLSQ






LHSGLFLYQGLLQALEGISPELGPTLDTLQL






DVADFATTIWQQMEELGMAPALQPTQGAM






PAFASAFQRRAGGVLVASHLQSFLEVSYRV






LRHLAQP



SEQ ID N:. 9
dGCSF-L5
39,12
MTPLGPASSLPQSFLLKCLEQVRKIQGDGA






ALQEKLCATYKLCHPEELVLLGHSLGIPWA






PLSSCPSQALQLAGCLSQLHSGLFLYQGLL






QALEGISPELGPTLDTLQLDVADFATTIWQQ






MEELGMAPALQPTQGAMPAFASAFQRRAG






GVLVASHLQSFLEVSYRVLRHLAQPSGGG






GSGGGGSGGGGSGGGGSGGGGSTPLGP






ASSLPQSFLLKCLEQVRKIQGDGAALQEKL






CATYKLCHPEELVLLGHSLGIPWAPLSSCP






SQALQLAGCLSQLHSGLFLYQGLLQALEGI






SPELGPTLDTLQLDVADFATTIWQQMEELG






MAPALQPTQGAMPAFASAFQRRAGGVLVA






SHLQSFLEVSYRVLRHLAQP



SEQ ID No:
dGCSF-L7
39,75
MTPLGPASSLPQSFLLKCLEQVRKIQGDGA



10


ALQEKLCATYKLCHPEELVLLGHSLGIPWA






PLSSCPSQALQLAGCLSQLHSGLFLYQGLL






QALEGISPELGPTLDTLQLDVADFATTIWQQ






MEELGMAPALQPTQGAMPAFASAFQRRAG






GVLVASHLQSFLEVSYRVLRHLAQPSGGG






GSGGGGSGGGGSGGGGSGGGGSGGGG






SGGGGSTPLGPASSLPQSFLLKCLEQVRKI






QGDGAALQEKLCATYKLCHPEELVLLGHSL






GIPWAPLSSCPSQALQLAGCLSQLHSGLFL






YQGLLQALEGISPELGPTLDTLQLDVADFAT






TIWQQMEELGMAPALQPTQGAMPAFASAF






QRRAGGVLVASHLQSFLEVSYRVLRHLAQ






P



SEQ ID No:
dGCSF-Lα
42,52
MTPLGPASSLPQSFLLKCLEQVRKIQGDGA



11


ALQEKLCATYKLCHPEELVLLGHSLGIPWA






PLSSCPSQALQLAGCLSQLHSGLFLYQGLL






QALEGISPELGPTLDTLQLDVADFATTIWQQ






MEELGMAPALQPTQGAMPAFASAFQRRAG






GVLVASHLQSFLEVSYRVLRHLAQPSGLEA






EAAAKEAAAKEAAAKEAAAKALEAEAAAKE






AAAKEAAAKEAAAKALEGSTPLGPASSLPQ






SFLLKCLEQVRKIQGDGAALQEKLCATYKL






CHPEELVLLGHSLGIPWAPLSSCPSQALQL






AGCLSQLHSGLFLYQGLLQALEGISPELGP






TLDTLQLDVADFATTIWQQMEELGMAPALQ






PTQGAMPAFASAFQRRAGGVLVASHLQSF






LEVSYRVLRHLAQP













































































































































































Molecular mass of dimeric protein is evaluated on SDS-PAGE in reducing conditions ( FIG. 1 and FIG. 2 ) and confirmed by mass-spectroscopy analysis. FIG. 2 also served as indication that there are no traces of —S—S— linked G-CSF dimer form. Correctness of purified dimeric protein structure and its identity is evidenced upon interaction with specific antibodies against monomeric G-CSF as has been demonstrated in Western blot analysis ( FIG. 3 ) where efficient response has been shown in the same manner for both proteins dimeric and monomeric G-CSF. Isolation and purification of the homo-multimers of therapeutic proteins produced according to the present invention is accomplished by the procedure established for purification of respective multimer form of therapeutic protein of interest (as in Example 3, below). Purity of isolated dimeric proteins is tested using reverse phase HPLC (RP-HPLC, FIG. 4 ) and size exclusion chromatography HPLC (SEC-HPLC, FIG. 5 ). Biologic activity of the dimeric proteins obtained is tested in vitro using proliferation test on murine cell line of myeloidic leukemia G-NFS-60 (as in Example 4, below) and in vivo test on the rats detecting number of neutrophils after injection of dimeric proteins (as in Example 6, FIG. 7 ). The concentration changes of dimeric proteins in the serum of rats' blood are also tested after injection and compared to the control monomeric protein (Example 5, FIG. 6 ).
Finally purified substance of homo-multimer protein produced according to the present invention is further formulated adding pharmaceutically acceptable carrier, diluent, excipient and/or auxiliary substances and used for therapeutic purposes in indications where monomer molecule of therapeutic protein is used (e.g. antiviral indications for IFN-alpha and neutropenia treatment indications for G-CSF).
Pharmaceutical composition of produced homo-multimer proteins according to the present invention could be formulated adding one or several components such as manitol, sorbitol, glucose, sacharose as carriers; acetate, phosphates, Tris, Hepes as pH regulating substances; mineral salts as substances for isotonic stabilization of composition; polisorbate 80 or 20, Pluronic type substances as surfactants; polyethilenglycol, polyvinylpyrrolidone, amino acids (methionine, arginine, histidine and others) as stabilization substances; EDTA, EGTA, IDA as chelating agents; kresol, benzyl alcohol, benzol derivatives as antimicrobial agents; glutathione, cysteine, acetylcysteine as SH agents; other auxiliary substances.
Embodiments of the Invention
Represented below is information on specific examples on preparation of new derivatives of recombinant proteins, including homo-multimers of GCSF, and properties thereof. These examples are presented for illustrative purpose and are not limiting the scope of the present invention.
EXAMPLE 1
Biosynthesis and Purification of Homo-Dimer of Recombinant Human Interfenon Alpha-2a (dIFN-2a) Protein
The bacterial strain producing dimeric IFN alpha-2a (dIFN-2a) E. coli BL21 (DE3) pET-21 IFN-L3-IFN was constructed using synthetic DNA fragment, coding the amino acid sequence of dimeric protein dIFN-alpha-2a-L3 (SEQ ID No:6, Table 2) after insertion into the expression plasmid pET-21b (Novagen). The DNA sequence of the plasmid was confirmed by direct sequencing. Bacterial strain, producing the dimeric IFN alpha-2a protein was obtained after transformation of E. coli BL21 (DE3) strain with the selected recombinant vector. The producer has been cultivated in 21 flasks. Medium was composed from 45,12 g M9 salts, 20 g yeast extract, 80 ml 20% glucose, 8 ml 1M MgSO 4 , 4 ml ampicillin 100 mg/ml. Expression of target protein was induced by IPTG (isopropyl-beta-D-thiogalactopiranoside). Final concentration of IPTG was 100 mM added when optical density (OD) reached 1.2 and cultivated additionally for 2-3 h. The biomass obtained have been frozen at −20° C. and used for further protein purification process.
9,4 g of frozen E. coli biomass is homogenized in 50 ml of 0.1 M Tris-HCl buffer, pH 8.5 containing 1 mM EDTA (ethylenediaminetetraacetic acid). Into obtained cell suspension 0.1% Triton X-100, 1% of 100 mM PMSF(phenylmethanesulfonyl fluoride) and 100 mM of β-mercaptoethanol is added. Homogenization mixture is mixed for 1 hour at ambient temperature, further is sonicated by ultrasonic in ice and finally centrifuged at 20.000 g for 15 min. Supernatant is removed and the collected pellet of inclusion bodies (IB) is twice washed by homogenization in 200 ml of washing buffer consisting of 1 M NaCl containing 0.1% of polysorbate 80. After each homogenization cycle the suspension of IB was centrifuged at 24,500 g for 15 min. Pre-washed IB are solubilized overnight in denaturation buffer of 50 mM Tris-HCl, pH 7.0 containing 6M guanidine hydrochloride (GuHCl) and centrifuged. Supernatant is diluted 20 times with the buffer: 100 mM Tris/HCl, pH 8.0, 0.5M L-arginine/HCl, 0,1mM EDTA, 1mM L-cysteine, 0.05 mM cystin. Renaturation is running for 45 h in 22° C. with slow mixing. Incubation buffer is dialysed in sodium acetate buffer pH 5,5 and loaded onto SP-Sepharose Fast Flow column. Desorption is performed with 0-0.5 M gradient of NaCl. Fractions of eluate containing target protein are combined and go to final formulation of pharmaceutical substance.
EXAMPLE 2
Biosynthesis of Homo-Dimer of Recombinant Human G-CSF (dG-CSF) Protein
The bacterial strain producing dimeric G-CSF (dG-CSF) E. coli BL21 (DE3) pET-21 GCSF/L5/GCSF was constructed using synthetic DNA fragment coding the amino acid sequence of dimeric protein dGCSF-L5 (see SEQ ID No:9, Table 2) after insertion into the expression plasmid pET-21b (Novagen). The DNA sequence of the plasmid was confirmed by direct sequencing. Bacterial strain producing the dimeric G-CSF protein was obtained after transformation of E. coli BL21 (DE3) strain with the selected recombinant vector. The producer have been cultivated in 21 flasks or in fermenter (Sartorius AG, 5L working volume). Medium was composed from 45,12 g M9 salts, 20 g yeast extract, 80 ml 20% glucose, 8 ml 1M MgSO 4 , 4 ml ampicillin 100mg/ml. Regulated parameters in fermenter were: temperature 37° C.; pH 6,8 (acid or base), aeration (pO 2 25-30%; 1.0-1.3 l/min, O 2 concentration was regulated by stirring intensity). Separately the optical density of culture was registered, which is indicating the amount of biomass obtained. Expression of target protein was induced by IPTG (isopropyl-beta-D-thiogalactopiranoside). Final concentration of IPTG was 100 mM added when OD reached 1.2 and cultivated additionaly for 2-3 h. The biomass obtained have been frozen at −20° C. and used for further protein purification process.
EXAMPLE 3
Isolation and Purification of Homo-Dimer of Recombinant Human G-CSF (dG-CSF Protein) in Which Monomer Units Are Genetically Connected Via Alpha-Helical Linker La
2.5 g of frozen E. coli biomass is homogenized in 50 ml of 0.1 M Tris-HCl buffer, pH 7.0 containing 5 mM EDTA. Into obtained cell suspension 0.1% lyzozyme, 0.1% Triton X-100, 1% of 100 mM PMSF and 0.7% of β-mercaptoethanol is added. Homogenization mixture is mixed for 30 min. at ambient temperature, further is sonicated by ultrasonic in ice bath and finally centrifuged at 24,500 g for 25 min. Supernatant is removed and the collected pellet of inclusion bodies (IB) is twice washed by homogenization in 50 ml of washing buffer, containing 1 M NaCl and 0.1% of polysorbate 80 and 50 ml of water for the third washing cycle. After each homogenization cycle the suspension of IB was centrifuged at 24,500 g for 25 min. Pre-washed IB are solubilized in denaturation buffer of 50 mM Tris-HCl, pH 8.0 containing 8 M urea, 1.0 mM EDTA and optionaly 100 mM of β-mercaptoethanol. 03.5 mM, or 2.0 mM, or 1.0 mM, or 0.1 mM, or 0 mM of DTT (dithiotreithol) is used.
Solubilization of IB is done by mixing overnight at +4° C. After that the solubilizate of IB is centrifuged at 24,500 g for 25 min. and diluted with 50 mM Tris-HCl buffer, pH 8.0 containing 3 M urea and 1 mM EDTA to the final protein concentration of 0.5 mg/ml (Bradford assay). Then IB refolding is performed in the presence of GSSG in a such manner that final molar ratio in solution of DTT/GSSG is equal to 1:6 and the refolding occurred by mixing for 90 h at +4° C.
After each refolding procedure the protein solution is centrifuged at 40.000 g for 25 min. to remove formed precipitate. Clarified protein solution is loaded onto chromatography column, containing 75 ml of anion-exchanger DEAE-Sepharose Fast Flow, equilibrated with 10 mM Tris-HCl buffer, pH 7.5 containing 20 mM NaCl or 10 mM Tris-HCl buffer, pH 7.5 at a linear flow-rate of 75 cm/h. The column is washed with the initial buffer of chromatography and the target protein is eluted by NaCl gradient from 20 mM to 80 mM over 1.4 column volume (CV) and further to 450 mM over 4 CV. Fractions of eluate containing target protein are pooled (assay by RP-HPLC) for further step of chromatography purification over SP-Sepharose FF column. For this purpose, the combined pool of fractions recovered from DEAE-Sepharose FF column is adjusted with 20 mM sodium acetate buffer pH 5,4 up to conductivity of 3.2 mS/cm and loaded onto column (XK 16/40 n Pharmacia′) containing 32 ml of SP-Sepharose Fast Flow equilibrated with 20 mM sodium acetate buffer pH 5.4 containing 20 mM NaCl. Protein solution is loaded into the column at a linear flow velocity of 75 cm/h. After column washing with the same buffer the target protein is eluted by increasing NaCl concentration from 20 mM to 450 mM in 20 mM sodium acetate, pH 5.4. Fractions of eluate containing target protein (assay by RP-HPLC) are combined and go to final formulation of pharmaceutical substance.
EXAMPLE 4
Determination of Biologic Activity of Homo-Dimer of Recombinant Human G-CSF (dG-CSF Protein), Using Proliferation Test of Sensitive Cell Line
Biologic activity of recombinant dG-CSF protein was detected, using murine cell line of myeloidic leukemia G-NFS-60 [Weinstein Y, lhle J N, Lavu S, Reddy E P, Truncation of the c-myb gene by retroviral integration in an interleukin 3-dependent myeloid leukemia cell line. Proc. Natl. Acad. Sci. USA, 1986, vol. 83, p. 5010-5014]. Active G-CSF protein initiates proliferation of this cell line after interaction with G-CSF receptor on the surface of the cell. For determination of biologic activity of dimeric G-CSF cells G-NFS-60 were cultivated in 96 well plates for 48-72 h, adding to cultivation media different amounts of (0.1-10 ng/ml) analysed GCSF protein sample. As a control the pharmaceutical samples of Neupogen™ (Filgrastim) with known biologic activity were used. Proliferation of the cells is evaluated by colorimetric method using MTS dye. The number of viable cells is dependent on the G-CSF biological activity. Data are shown in Table 3.





TABLE 3



Biologic activity of monomer G-CSF and homo-dimer of recombinant

human G-CSF (dG-CSF protein) samples







Biologic activity






No.
Preparation
IU/ml
IU/mg



1
dGCSF L3
3.7 · 10 6
2.3 · 10 7

2
dGCSF-L5
21.7 · 10 6
1.57 · 10 7

3
dGCSF-L7
10.8 · 10 6
2.16 · 10 7

4
dGCSF-Lα
7.05 · 10 6
3.06 · 10 7

5
Monomer G-CSF
60.0 · 10 6
10.0 · 10 7






























The results obtained indicate that the G-CSF protein structures invented and purified with the proposed methods provide the biologically active dimeric protein.
EXAMPLE 5
Pharmacokinetic Study of Homo-Dimer of Recombinant Human G-CSF (dG-CSF Protein) in Blood Serum
For pharmacokinetic study of recombinant proteins in blood serum experimental animals (Wistar rats) have been divided into the groups, 4-5 rats in each. Control and analysed proteins have been injected subcutaneously. Injection doses used were from 100 to 1100 μg/kg. Neupogen™ (Filgrastim) was used as control in the same concentrations calculated to the same rat weight. Blood serum was collected 3, 6, 12 and 18 h post injection. Collected blood samples were centrifuged and serum was sterilely taken, aliquated and frozen at −80° C. Blood serum samples have been tested according to manufacture instructions using G-CSF detection DuoSet kits of R&D company using ELISA method. Results ( FIG. 6 ) are indicating that concentration of dG-CSF in rat serum remains higher than control data during all 18 h of experiment except first three hours.
EXAMPLE 6
Pharmacodynamic Analysis of Biologic Effect of Homo-Dimmer of Recombinant Human G-CSF (dG-CSF Protein)
For pharmacokinetic analysis of recombinant proteins on the blood cells experimental animals (Wistar rats) have been divided into the groups, 4-5 rats in each. Control and analysed proteins have been injected subcutaneously. Injection doses used were from 100 to 1100 μg/kg. Neupogen™ (Filgrastim) was used as control in the same concentrations calculated to the same rat weight. Blood samples have been collected 24, 48, 72, 96 h post injection. Rat blood cells were analysed in blood analyser Hemavet 950. Biologic effect of dimeric protein was tested analysing the number of blood neutrophils before and after injection of protein samples. 20 μl of blood was taken from rat tail vein, mixed with 5 μl of 7.5% EDTA, incubated for 10 min. and analysed with Hemavet 950. Results are represented in FIG. 7 , showing that effect of dimeric protein (dG-CSF) on the blood cells (the number of neutrophils) is exceeding control data.
The examples performed allow to conclude that tested homo-dimer of G-CSF (dG-CSF) proteins have biologic activity in vitro and in vivo, that is similar to the effect of control monomeric GCSF protein. Constructed homo-dimeric G-CSF proteins showed some new valuable PK/PD features, such as prolonged positive influence on neutrophil number and increased circulation time of protein in rat blood in comparison with monomeric G-CSF protein. This characteristic feature of homo-dimer of G-CSF provide possibility for its use in clinical therapy instead of monomer of G-CSF, when the time of administration of G-CSF drug preparation, e.g. Neupogen shorter than two weeks is needed, since there long-acting pegylated form of G-CSF preparation, e.g. Neulasta is not acceptable. From the other side, the effective therapeutic dose of homo-dimer of G-CSF is expected to be much more lower than of Neulasta preparation.
Proposed according the present invention new approach for increasing in vivo circulation half-life time of recombinant proteins of therapeutic values distinguished in that genetic fusion of protein monomer unit into homo-multimers do not necessary leads to decrease of biological activity—the common problem for all known therapeutic proteins with extended circulation half-life. Moreover, properly selected linker between monomer units allows each monomer specifically interact with receptors and increase the biological activity. Increased biological activity is defined herein as a prolonged plasma half-life or higher potency. This may require reduction of the frequency of administration or reduction of the quantity of protein multimer to achieve an effective dose characteristic to monomer unit of the multimer construction.
Currently marketed second-generation therapeutic proteins produced via pegylation approach exhibit lower biological activity in comparison to non-modified proteins and consequently requires higher protein amount per dose than the first generation analogues.
For example, therapeutic dose of Neulasta (PEG-G-CSF) is 6 mg instead of 0.30-0.48 mg for Neupogen (G-CSF). Therefore, proposed new approach according to the present invention for increasing of in vivo circulation time of recombinant proteins of therapeutic values will decrease protein amount per therapeutic dose.
In general, the proposed method of the genetic fusion of therapeutic protein molecule into homo-multimer, particularly homo-dimer, structure by connecting monomer units via properly selected sequence of linker moiety allows to achieve biologically active dimer with enhanced in vivo circulation time. This method distinguished from known method of multimer production in that genetical fusion assures reproducibility of quality parameters of finally purified product in comparison with the procedure of chemical conjugation. Monomer units of selected therapeutic protein are fused into homo-dimer with the aid of linker sequence with assumption that the length and structure of the linker of choice is sufficient for each monomer unit interaction with specific receptor, antibody and so on.
On the other hand the method of present invention of homo-multimers production based on recombinant DNA technologies is more technologically attractive in comparison with chemical modification and allows to produce active pharmaceutical substance of interest with consistent quality. The genetic fusion of protein monomer units into homo-multimers will avoid increase of immunogenicity and toxicity effects of the fusions, since structure of monomer units remains unaffected.
This also enable to ensure that the extension of in vivo circulation half-life is achieved without any invasion on the native structure of the monomer unit with respect to replacement/addition or deletion of amino acid residue and further modification with polymer macromolecules like PEG. The proposed method may be expanded on practically all currently known therapeutic protein with emphasis on those representatives which naturally exist in vivo as dimers.