Patent Application
20120134960
The present invention belongs to the field of molecular biology related to the development of recombinant proteins. In particular, it is related to human interferon-alpha recombinant muteins having improved pharmacological, chemical, and physical properties; processes for obtaining said muteins and pharmaceutical formulations for use in mammals, particularly humans, in need of a therapeutic treatment with interferon. More particularly, this invention is related to human interferon-alpha muteins, containing at least one modified amino acid at a position pertaining to an alpha-helix-type secondary structure. Said modification generates at least one sequence of the Asn-Xaa-Ser/Thr type, introducing a N-glycosylation susceptible site by glycosylation of an asparagine residue in said sequence.
The group of cytokines known as interferons was characterized for the first time in 1957. They were designated as interferons (IFNs) because of their ability to interfere in viral replication, thereby conferring resistance to infection from infected cells to non-infected cells. Isaacs and Lindermann (Proc R Soc Lond Biol Sci, September 1957; 147(927): 258-67) were the first to demonstrate this characteristic in said cytokines.
IFNs are divided in two groups according to their structural and functional properties: type I interferons (IFN-alpha, IFN-beta, IFN-omega) and type II interferons (IFN-gamma). Each of these is expressed in a great variety of cell types at low concentrations. IFN-alpha is expressed mainly in B-lymphocytes and macrophages; IFN-beta is expressed in fibroblasts, and IFN-gamma in T-lymphocytes.
Recently, a new member has been proposed for the interferon family, known as type III interferon, comprised by interferons lambda 1, 2, and 3 (corresponding to Interleukin-29 (IL-29), IL-28a, and IL-28b, respectively) (Kotenko, S. V. 2003. IFN-lambda mediate antiviral protection through a distinct class II cytokine receptor complex. Nat. Immunol. 4:69-77; Sheppard, P. 2003. IL-28, IL-29 and their class II cytokine receptor IL-28R. Nat. Immunol. 4:63-68). These type III IFNs show a low degree of sequence homology to the α/β IFN family but they are activated by the same factors that induce expression of α/β IFNs and inhibit replication of several viruses including vesicular stomatitis virus (VSV) and encephalomyocarditis virus (EMVC). In addition, they may also inhibit replication of type B and C hepatitis viruses.
Interferon-alpha belongs to a multigenic family consisting of genes and pseudogenes, of which more than 20 variants have been identified. The first three-dimensional models of the human I IFNS was predicted in 1982 keeping in mind its amino acidic sequences. Recently, the tertiary structure of the IFN-alpha 2b was revealed by means of X-ray crystallography (Radhakrishnan et al. 1996. Zinc-mediated dimmer of human interferon-alpha 2b revealed by X-ray crystallography. Structure 4 (12): 1453-1463) that it is incorporated herein as part, by reference, in its entirety, into this application, in order to fully describe the state of the art to which this invention pertains, while the corresponding to IFN-alpha 2a in solution was analyzed by means of NMR spectroscopy (Klaus et al. 1997. The three-dimensional high resolution structure of human interferon-alpha 2a determined by heteronuclear NMR spectroscopy in solution. J. Mol. Biol. 274: 661-675) that is incorporated herein as part, by reference in its entirety into this application, in order to fully describe the state of the art to which this invention pertains.
These papers demonstrated that the IFN-alpha2 possesses a consistent globular structure in 5 alpha helices designated with the letters A (Ser11-Met21), B and B′ (Thr52-Ser68 and Lys70-Ala75), C (Glu78-Ile100), D (Leu110-Glu132) and E (Pro137-Leu157), which allows to classify this protein inside the group of the alpha-helical cytokines. The helices are linked by one long connection (AB loop) and 3 short segments (BC, CD and DE loops). Each helix is approximately straight, except for helix B that contains a pronounced bend of 70° centered on Thr69, allowing to divide the mentioned helix in the region B and B′ (
The structure of interferon-alpha 2b contains 2 disulphide bonds and an O-linked glycosydid chain, at position Thr 106. The post-translational modifications necessary for transferring a glycan to protein take place in the cell's endoplasmic reticulum immediately after translocation of the freshly synthesized polypeptide from the cellular cytoplasm. For a N-type glycosylation to occur, the presence of sequences of the Asn-Xaa-Ser/Thr type in the amino acid chain is required, where an N-glycosydic bond is formed between the amino group of the asparagine residue and an OH group of the carbohydrate.
The high-level expression of type I interferon in living organisms is induced by the presence of a viral infection. It may also be induced by a great variety of other non-viral agents, such as bacteria, mycoplasma and different polymers, for example, membrane lipopolysaccharides of Gram-negative bacteria and the presence of synthetic polymers (Merigan T: Induction of circulating interferon by synthetic anionic polymers of known composition. Nature 1967, 214: 416-417).
Type I-interferon also exhibits anti-proliferative and immunomodulating activity.
The signaling pathway of type I IFNs starts with the binding of said molecule to the cellular surface receptor target and mediates the activation of target genes in the nucleus through the Jak/STAT signaling pathway. There are other gene activation pathways resulting from binding to the IFN-receptor that do not depend on the above-mentioned signal transduction pathway (A Larner: Interferon signal transduction. Biotherapy, Jan. 1, 1996; 8 (3-4): 175-81).
Interferon alpha is used clinically in the treatment of chronic hepatitis B and C, acute viral encephalitis, cancers such as nasopharyngeal carcinoma, lymphoma, leukemia, and melanomas, among others.
One of the major drawbacks of the use of interferon in the treatment of diseases is its low stability when administered to a living organism, and therefore it is necessary to provide excess amounts of interferon to achieve suitable therapeutic levels. In the first place, this contributes to the high cost of interferon therapies, particularly of Interferon-alpha therapies, and in the second place, in certain patients these treatments have side-effects such as flu-like conditions, fever, fatigue, irritability, chills, headaches and muscle pain; upset stomach, loss of appetite, diarrhea, dizziness, etc. (Richard Grieve: Cost-effectiveness of interferon or peg-interferon with ribavirin for histologically mild chronic hepatitis C. Gut, June 2005; 10.1136/gut.2005.064774; Gary L. Davis: Interferon-alpha-2b alone or with ribavirin for the treatment of relapse of chronic hepatitis C. 1998. Volume 339 Number 21-1493).
Currently new strategies are being developed with the aim to increase in vivo stability of these bioactive proteins using enzymatic reactions (WO 98/13381, U.S. Pat. No. 6,620,916, U.S. Pat. No. 5,643,564, U.S. Pat. No. 4,184,917), chemical reactions (WO 96/21468, US 2004/0180054, U.S. Pat. No. 6,524,570, U.S. Pat. No. 4,179,337, U.S. Pat. No. 5,981,709, US 20060029573, U.S. Pat. No. 5,738,846), encapsulation of protein (WO2005074892), directed mutagenesis (US2004/0002474, WO92/01055, US 2005/0019871), etc. The purpose of these investigations is to achieve therapeutically effective doses not requiring the administration of large amounts of protein. Examples of these new strategies are disclosed in documents such as Patent WO98/13381 (Ajinomoto Kk (Jp); Takahara A N D., et al.), that reveals a method for modifying biologically active proteins and enhancing its in vivo activity, by the addition of branched ligands to glutamine residues present in the primary sequence of the protein. Patent documents U.S. Pat. No. 6,620,916 (Ajinomoto Co. Inc.; Takahara, et al.); U.S. Pat. No. 5,643,564 (Takeda Chemical Industries, Ltd; Hamaguchi, et al.), and U.S. Pat. No. 4,184,917 (Sandoz Ltd.; Dorner, et al.) describe methods to obtain interferon homologues with enhanced stability, by enzymatic modifications which alter their glycosylation profile. These new homologues act essentially in the liver and could improve effectiveness in this organ, but decrease the effect of interferon on other locations. Patent document WO 96/21468 (Amgen Inc; Collins David, et al.) reveals recombinant interferon-alpha analogues obtained by addition of glycosydic ligands with multiple lactoses conjugated to Arginine residues or to amino terminal residues of the natural interferon-alpha primary sequence. The product claimed in this document is directed to the liver and therefore would be acting at hepatic level and showing deficiencies in other organs such as, for example in kidney tumors. US Patent Application 2004/0180054 (Kim, Young-Min, et al.) reveals a conjugate comprising a protein (IFN-alpha), a non-peptidic polymer, and an immunoglobulin, thus increasing the in vivo stability of the protein. In Patent documents U.S. Pat. No. 4,179,337, U.S. Pat. No. 5,981,709, US 20060029573, U.S. Pat. No. 5,738,846, U.S. Pat. No. 6,524,570; the inventors disclose processes, compositions, and uses of interferon-alpha conjugated to polyethyleneglycol for increasing product stability. These methods comprise purification steps that are necessary for removing reaction products before carrying out the minimum necessary purification steps for obtaining a high purity product. In addition, the proposed process shows the difficulty of directing a chemical reaction to a specific site of the protein, which leads to heterogeneous products, entailing the possibility of obtaining inactive products.
In Patent Application WO 2005074892 (Shenzhen Neptunus Interlong B I; Wang et al.) the inventors describe a cream containing interferon liposomes used for the treatment of skin diseases caused by viral infections (for example, herpes zoster), warts, genital ulcers, etc. This kind of preparation has the disadvantage of its limited application for the topical treatment of diseases.
US Patent Application 2004/0002474 (Maxygen, Inc.; Heinrichs, Volker, et al.), claims a method to obtain interferon-alpha homologues by modifications in the gene or amino acid sequence. In Patent document WO 92/01055 (Boehringer Ingelheim Int; Adolf Guenther, et al.) the inventors disclose a method for obtaining interferon-alpha homologues having new O-glycosylation sites. The product obtained by these strategies shows no improvements in pharmacokinetic properties as compared to pegylated interferon.
US Patent Application 2005/0019871 (Lee, Eun Jung, et al.) reveals a method for obtaining interferon-alpha isoforms, in which the nucleotide sequence has been modified in order to introduce Asn-Xaa-Ser/Thr motifs in sites that are not involved in forming secondary structures, and proposes sites located exclusively outside the alpha-helix structures. Although recombinant human interferon-α2b isoforms with new N-glycosylation sites were obtained, there is no sign of improvement in stability and/or release over time profiles, as compared to known commercially available isoforms such as interferon pegylated with 12000 Dalton Polyethyleneglycol (PEG 12000). This application only describes isoforms with only two new N-glycosylation sites in molecular regions lacking alpha-helix structures, which would represent areas of greater susceptibility to the action of N-glycanase.
The present invention solves the problems described in the prior art and provides a preparation comprising a recombinant human interferon-alpha mutein, having an N-glycosylation profile which is different from those known in the prior art. Said mutein shows improved release over time and stability profiles, and an in vivo biological activity comparable to that of interferon pegylated with a 12000 Dalton Polyethyleneglycol (PEG 12000). In addition said mutein is obtained by a process comprising few steps and simple purification, yielding a high purity final product.
It is a main object of the present invention to provide a recombinant human interferon-alpha mutein, preferably of the 2b type, containing at least one glycosylation site, preferably N-glycosylation, at a position of its amino acid sequence that forms part of an alpha-helix-type secondary structure. In a preferred embodiment of the present invention, said recombinant human interferon-alpha mutein 2b further exhibits at least one N-glycosylation site at a position located outside an alpha-helix structure. Said N-glycosylation sites, preferably comprising a consensus sequence of the Asn-Xaa-Ser/Thr-type, at a position of its amino acid sequence that is part of an alpha-helix-type secondary structure, are positions selected from the group consisting of: Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156. While said N-glycosylation sites located outside alpha-helix-type secondary structures occupy positions selected from the group consisting of the following positions in natural human interferon-alpha 2b, Pro4, Thr6, Arg23, Leu26, Asn45, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161.
It is another object of the present invention to provide a gene encoding a recombinant human interferon-alpha mutein, preferably of type 2b, comprising at least one site-directed mutation that generates one N-glycosylation site, preferably a consensus sequence of the Asn-Xaa-Ser/Thr type, involved in an alpha-helix structure. Said site-directed mutation is carried out in amino acids selected from the group consisting of Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156 amino acids. In a preferred embodiment said gene further exhibits at least one site-directed mutation that generates a consensus sequence of the Asn-Xaa-Ser/Thr type, where the Asn can be bound to an oligosaccharide through an N-glycosydic bond and is located outside an alpha-helix-type secondary structure. Said N-glycosylation site located outside an alpha-helix-type secondary structure is located in amino acids selected from the group consisting of Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161.
It is another object of the present invention to provide a method for generating new sites capable of being N-glycosylated using a site-directed mutagenesis procedure in the gene that encodes said recombinant human interferon-alpha mutein of the invention.
It is another object of the present invention to provide a method for obtaining an eukaryotic cell line, for producing said recombinant human interferon-alpha mutein of the invention by transformation or transfection of a cell line containing said gene encoding said recombinant human interferon-alpha mutein of the invention, inserted in a suitable expression vector. Preferably, the eukaryotic cell line is the CHO.K1 cell line.
It is another object of the present invention to provide a method for producing the recombinant human interferon mutein of the invention, said method comprising the steps of: a) culturing said transformed or transfected eukaryotic cell line with an expression vector containing the gene encoding the human interferon mutein of the invention, and b) isolating the expressed and secreted recombinant human interferon-alpha mutein from the culture medium.
It is another object of the present invention to provide a process for purifying said recombinant human interferon-alpha mutein of the invention by immunoaffinity chromatography.
It is another object of the present invention to provide a method for producing the human interferon mutein of the invention from prokaryotic cells comprising the steps of: a) transforming or transfecting a prokaryotic cell with a suitable expression vector containing the gene encoding the recombinant human interferon-alpha mutein of the invention; b) selecting a clone expressing the polypeptide of the recombinant human interferon-alpha mutein of the invention; c) culturing said clone, d) purifying e) glycosylating “in vitro” a polypeptide of the human interferon-alpha mutein expressed by the clone of step c); and f) purifying the human interferon-alpha mutein of the invention.
It is another object of the present invention to provide a pharmaceutical composition comprising at least the recombinant human interferon-alpha mutein of the invention for the treatment of diseases such as melanomas, chronic hepatitis C, acute and chronic hepatitis B, acute and chronic non-A, non-B hepatitis, Kaposi's sarcoma, multiple sclerosis, genital warts, leukemia, viral infections, among others.
It is another object of the present invention to provide a method for the treatment of melanomas, chronic hepatitis C, acute and chronic hepatitis B, acute and chronic non-A, non-B hepatitis, Kaposi's sarcoma, multiple sclerosis, genital warts, leukemia, viral infections, comprising the administration to a mammal, preferably a human, in need thereof, of an effective amount of a pharmaceutical composition comprising at least the recombinant human interferon-alpha mutein of the invention and wherein said composition further comprises pharmacologically acceptable carriers.
It is another object of the present invention to provide the use of the recombinant human interferon-alpha mutein of the invention for manufacturing a medicament for the treatment of chronic hepatitis C, using a therapeutic protocol in which the usual therapeutic doses of interferon are reduced.
a: shows the amino acid sequence of human type I interferon-alpha 2b, human interferon-alpha 1, human interferon-beta and murine interferon-beta. The highlighted amino acids in the consensus sequence correspond to the sequences conserved in the human alpha interferons, while the residues highlighted in the rest of the sequences are conserved in all type I interferons. The alpha helix-type secondary structure corresponding to the molecules of human interferon alpha 2b and murine interferon beta are indicated as segments in the upper and lower part of the respective sequences (adapted from Radhakrishnan et al. 1996. Zinc-mediated dimer of human interferon-alpha2b revealed by X-ray crystallography. Structure 4 (12): 1453-1463).
b: nucleotidic sequence of the interferon alpha 2b. The thin lines indicate the codons corresponding to the first amino acid of the protein sequence (Cys), the last amino acid of this sequence (Glu) and the STOP codon.
Some terms used in the present specification are defined below to facilitate understanding of the invention:
“Natural human interferon-alpha 2b” (hIFN-α2b), refers to a cytokine as it is found in nature, without having been subjected to any kind of artificial modification.
“Amino acid substitution”, refers to the change of one amino acid in the primary sequence of hIFN-α2b for another amino acid.
“Recombinant human interferon-alpha mutein of the invention”, refers to recombinant molecules of human interferon-alpha, preferably alpha 2b, containing at least one glycosylation site, preferably N-glycosylation, at a position of its amino acid sequence forming part of an alpha-helix-type secondary structure.
“Human interferon-alpha” includes analogues, mutants, isoforms, and fragments of natural interferon-alpha.
“N-glycosylation site”, refers to an Asn-Xaa-Ser/Thr tripeptide, where X may be any residue except a proline residue. The “position” of the “N-glycosylation site” is indicated by the position occupied by an amino acid residue in the amino acid sequence of a natural human interferon-alpha 2b that will be replaced by an Asn or it is the asparagine of said consensus sequence. Said Asn residue, in said consensus sequence, may be subjected to an N-type enzymatic glycosylation.
Glycosylation of certain eukaryotic proteins takes place at certain positions of the polypeptide backbone, and commonly there are two types of glycosylation. O-type glycosylation involves binding of an oligosaccharide to an —OH group of a serine or threonine residue, and N-type glycosylation, which involves binding of an oligosaccharide to an —NH group of an Asparagine residue. Particularly, N-glycosylation takes place in the consensus sequence, Asn-X-Ser/Thr, where X may be any amino acid different from Proline. All the oligosaccharides bound to a protein through an N-type binding have a pentasaccharide nucleus in common comprised by three mannose residues and two N-acetylglucosamine residues. Any sugars bound to this pentasaccharide nucleus may acquire a great variety of oligosaccharide patterns. The presence or absence of said oligosaccharides affects the physical properties of proteins and may be critical in their function, stability, secretion, and location in the cell.
The present invention is related to the production of a human interferon-alpha mutein having at least one amino acid substitution in the sequence of the human natural interferon-alpha at a position forming part of an alpha-helix secondary structure, resulting in the consensus sequence Asn-Xaa-Ser/Thr, where the Asn residue is capable of being N-glycosylated. For example, the recombinant human interferon-alpha mutein of the invention, containing at least one N-glycosylation site at a position forming part of an alpha-helix secondary structure, is obtained by an amino acid substitution at position Lys70 with an Asn residue (Lys70Asn). This generates the consensus sequence, Asn-Xaa-Ser/Thr, where the asparagine residue is part of an alpha-helix secondary structure and is susceptible to N-glycosylation. Another example may be the substitution of the amino acid residue Leu95 with a Serine or Threonine residue, thus generating a consensus sequence, Asn-Xaa-Ser/Thr, wherein the asparagine residue at position Asn93 is susceptible to N-glycosylation. It has been demonstrated, that the presence of at least one N-glycosylation site in alpha-helix-type structures confers a greater stability to interferon and therefore a longer half-life in blood.
Said N-glycosylation sites, generated by substitution of an amino acid forming part of an alpha-helix-type secondary structure are selected from the group consisting of amino acid positions Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156.
One preferred embodiment of the present invention is a recombinant human interferon-alpha mutein having at least one amino acid substitution at a position forming part of an alpha-helix secondary structure and in addition at least one substitution in an amino acid at a position located outside the alpha-helix structure so that Asn-Xaa-Ser/Thr consensus sequences are obtained, where the Asn residue is susceptible to N-glycosylation. Said N-glycosylation sites at positions located outside alpha-helix structures are selected from the group consisting of amino acid positions Pro4, Thr6, Arg23, Leu26, Asn45, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161.
In a preferred embodiment, the recombinant human interferon-alpha mutein of the invention contains in its amino acid sequence an N-glycosylation site at a position forming part of alpha-helix structures, wherein the position of the glycosylation site is selected from the group consisting of amino acid positions Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156. Even more preferably, the position is selected from the group comprised by amino acid positions Lys70, Asn93, Glu113. Even more preferably, the position is Lys70.
In a further preferred embodiment of the present invention, the recombinant human interferon-alpha mutein of the invention has two N-glycosylation sites, where said sites are selected from the group consisting of amino acid positions Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156, Pro4, Thr6, Arg23, Leu26, Asn45, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161. Even more preferably, said mutein contains a glycosylation site at position Lys70 and further an N-glycosylation site at a position selected from the group consisting of positions Pro4, Arg23, and Asp77.
Preferably, the recombinant human interferon-alpha mutein of the invention contains three N-glycosylation sites selected from the group consisting of amino acid positions Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156, Pro4, Thr6, Arg23, Asn45, Leu26, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161. Preferably, the recombinant human interferon-alpha mutein of the invention having three N-glycosylation sites, contains one N-glycosylation site at position Lys70 and two N-glycosylation sites selected from the group consisting of positions Pro4, Arg23, and Asp77.
More preferably, the recombinant human interferon-alpha mutein of the invention is a mutein containing four N-glycosylation sites selected from the group consisting of amino acid positions Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156, Pro4, Thr6, Arg23, Leu26, Asn45, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161. Even more preferably, said sites are Pro4, Arg23, Lys70, and Asp77.
In still another preferred embodiment, the recombinant human interferon-alpha mutein of the invention contains 5 N-glycosylation sites selected from the group of amino acid positions, Leu9, Arg12, Asn65, Leu66, Phe67, Lys70, Asp71, Phe84, Asn93, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Asn156, Pro4, Thr6, Arg23, Leu26, Asn45, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161. Even more preferably, said sites are selected from the group consisting of positions Pro4, Arg23, Lys70, Asp77, Asn93, Glu113.
In order to obtain the recombinant human interferon-alpha mutein of the invention containing sites susceptible to N-glycosylation, two different strategies were followed:
1) locating all serines and/or threonines of the hIFN-α2b amino acid sequence and changing the amino acid occupying the first position of the consensus sequence by an asparagine,
2) as an alternative strategy, locating all asparagines present in the hIFN-α2b sequence and substituting the amino acid at the third position by a serine or threonine.
The modifications carried out in the natural amino acid sequence of human interferon-alpha for obtaining the recombinant mutein of the invention are a result of a genetic modification of the gene encoding natural human interferon-alpha 2b. Further, said genetic modifications are introduced in such a way that they generate an Asn-Xaa-Ser/Thr consensus sequence in the amino acid sequence of the human interferon-alpha, wherein the Asn residue is susceptible to N-glycosylation.
The gene encoding the recombinant human interferon-alpha mutein of the invention, claimed hereunder, comprises at least one genetic modification in a codon of the gene encoding human natural interferon-alpha. Said modification generates the Asn-Xaa-Ser/Thr consensus sequence at a position that is a part of an alpha-helix secondary structure, thereby generating an N-glycosylation site at a position forming part of an alpha-helix-type secondary structure. Preferably, said modification of the nucleotide sequence is carried out in at least one position selected from the group consisting of amino acid positions Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158.
In one of the preferred embodiments of the present invention, said gene encoding the recombinant human interferon-alpha mutein of the invention, further shows at least one genetic modification in a codon of the gene encoding the human natural interferon-alpha, such that said modification generates the Asn-Xaa-Ser/Thr consensus sequence at a position located outside an alpha-helix secondary structure. This generates an N-glycosylation site at a position located outside an alpha-helix-type secondary structure. Said modifications in positions located outside alpha-helix-type secondary structures are carried out at positions selected from the group consisting of positions Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161.
In a preferred embodiment of the present invention, said gene has only one modification in its nucleotide sequence which takes place in the codon encoding amino acids selected from the group consisting of Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158. More preferably, it takes place in the codon encoding the amino acid at position Lys70.
In an additional preferred embodiment, the gene encoding the recombinant human interferon-alpha mutein of the present invention has two modifications in its nucleotide sequence such that two N-glycosylation sites are generated. Said modifications involve codons encoding amino acids selected from the group consisting of Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158, Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Leu161. More preferably, the modifications in its nucleotide sequence are in the codon encoding Lys70 and in the codon encoding one of the amino acids selected from the group consisting of Pro4, Arg23, and Asp77.
In another preferred embodiment of this invention, the gene encoding the recombinant mutein of the present invention has three modifications in its nucleotide sequence such that three N-glycosylation sites are generated. Said modifications involve codons encoding amino acids selected from the group consisting of Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158, Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Leu161. More preferably, modifications in its nucleotide sequence are in the codon encoding Lys70 and in two codons encoding amino acids selected from the group consisting of Pro4, Arg23, and Asp77.
In a further more preferred embodiment of this invention, the gene encoding the recombinant mutein of the present invention has four modifications in its nucleotide sequence such that four N-glycosylation sites are generated. Said modifications involve the codons encoding amino acids selected from the group consisting of Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158, Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Leu161. More preferably, the modifications in its nucleotide sequence involve codons encoding Pro4, Arg23, Lys70, and Asp77.
In still another preferred embodiment, the gene encoding the recombinant human interferon-alpha mutein of the invention has five modifications in its nucleotide sequence in codons encoding amino acids selected from the group consisting of Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158, Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Leu161. Even more preferably, in amino acids selected from the group consisting of Pro4, Arg23, Lys70, Asp77, Leu95, and Glu113.
In addition, the present invention provides a method for generating said sites susceptible to N-glycosylation. Said method comprises the generation of point mutations in the nucleotide sequence of the gene encoding the human natural interferon-alpha, by means of a site-directed mutagenesis technique in said gene. The method comprises the following steps:
a) cloning a gene encoding natural human interferon-alpha 2b in a suitable plasmid.
b) generating mutations required for producing the human interferon-alpha mutein of the invention using a site-directed mutagenesis technique, and c) cloning the genetically modified gene from step b, into a suitable expression vector.
The expression vector is selected from the group of vectors capable of carrying the gene of the invention and further containing the necessary elements for expressing the gene of interest in eukaryotic cells. Said vector may be expression vector pCl-neo.
The site-directed mutagenesis technique of the invention involves the use of oligonucleotides specifically designed to that end. This technique comprises two stages. In the first stage, two PCR reactions are carried out separately using oligonucleotides that hybridize to the terminal ends of the fragment cloned into a pCl-neo vector (oligonucleotides designated IFNalphaF and IFNalphaR, Table I), and oligonucleotides carrying a point mutation of Table II (mut a and mut b) which hybridize to the internal region of the gene where the mutation is to be introduced. A reaction mixture is obtained in tube a using a reverse external oligonucleotide (IFNalphaR) and the direct oligonucleotide mut a. Another reaction mixture is obtained in tube b with a direct external oligonucleotide and the reverse oligonucleotide mut b. PCR products from both reactions are purified by agarose gel electrophoresis and used as a template for the second stage. This second stage comprises a second PCR reaction using direct and reverse external oligonucleotides. The first three cycles are carried out without the addition of primers to allow hybridization and elongation of the complete product (fill in) and finally these are added for the amplification.
To obtain more than one N-glycosylation site within the recombinant human interferon-alpha mutein of the invention, said muteins are constructed sequentially as follows: first, a mutein with one N-glycosylation site is generated, using a site-directed mutagenesis technique, and then said mutein is used as a starting template for generating a new N-glycosylation site.
The method for obtaining a derivate eukaryotic cell line which produces the recombinant human interferon-alpha mutein of the invention, more particularly a cell line CHO.K1, transfected or transformed with an expression plasmid pCl-neo containing the gene encoding the recombinant human interferon-alpha mutein of the invention, involves the following steps:
A process, also an object of the present invention, for obtaining the recombinant human interferon-alpha mutein of the invention from said CHO.K1 cell line, comprises the steps of:
a) culturing the recombinant human interferon-alpha mutein-producing eukaryotic cell of the invention, and
b) isolating the recombinant human interferon-alpha mutein of the invention expressed and secreted into the culture medium.
A method for purifying the recombinant human interferon-alpha mutein of the invention, also an object of the invention, involves obtaining specific monoclonal antibodies for said mutein, adsorbing said monoclonal antibodies on a suitable chromatographic column and purifying the mutein of the invention by immunoaffinity chromatography.
Interferon-alpha may be expressed in bacteria, yeasts or insect cells according to procedures well known in the art. In all these cases, the recombinant interferon either is not glycosylated or it has a degree of glycosylation that is lower and different from the glycosylated interferon produced in animal cells. In all these cases, sugar chains may be glycosylated or remodelled “in vitro”. For this purpose, there exist numerous protocols described in detail in the following patents or patent applications: WO03031464, WO9425615, WO9216640, US20030040037, US20030003529, US20020137134, US20020019342, US20030124645, US20020160460, US20020142370, US20020119516.
One of the preferred embodiments of the present invention provides, a method for producing the human interferon mutein of the comprising the steps of: a) transforming or transfecting a prokaryotic cell with a suitable prokaryotic expression vector containing the gene encoding the recombinant human interferon-alpha mutein of the invention; b) selecting a clone expressing the polypeptide of the recombinant human interferon-alpha mutein of the invention; c) culturing said clone in a suitable culture medium, d) purifying the product, e) glycosylating “in vitro” the human interferon-alpha mutein polypeptide expressed by the clone of step c); and f) purifying the human interferon-alpha mutein of the invention.
Glycosylation profiles of various recombinant hIFN-α2b muteins were analyzed: muteins with one glycosylation site located outside an alpha-helix secondary structure; the recombinant human interferon-alpha mutein of the invention; natural hIFN-α2b; and a non-glycosylated isoform; as shown in
The recombinant human interferon-alpha mutein of the invention in a preferred embodiment comprising four N-glycosylation sites at positions Pro4, Arg23, Lys70, and Asp77, rhIFN-α2bM4/M23/M70/M77, exhibits a homogeneous glycosylation profile with a high glycosylation value, as can be seen in
By creating these new N-glycosylation sites it has become possible to increase the in vivo stability of cytokine as compared to any previously known isoform of rhIFN-α2b. In a preferred aspect of the invention, the creation of the recombinant human interferon-alpha mutein of the invention with four N-glycosylation sites (rhIFN-α2bM4/M23/M70/M77) displays a release over time profile equivalent to the commercially available molecule of interferon-PEG 12000, resulting in a product with a high degree of N-glycosylation and a more homogeneous glycosylation profile. In addition, notably the product thus obtained does not involve chemical reactions or the addition of synthetic polymers.
The resulting biological activity versus time values show that the recombinant human interferon-alpha mutein of the invention, in particular that corresponding to a mutein with a new N-glycosylation site at position Lys70, shows a 1.63-fold increase as compared to a mutein with a new N-glycosylation site lacking an alpha-helix structure (rhIFN-α2bM23). This clearly indicates the importance of selecting N-glycosylation sites within structures of the alpha-helix type.
The recombinant human interferon-alpha mutein of the invention exhibits improved pharmacokinetic, physical, and chemical properties with respect to previously known isoforms obtained my means of cell cultures.
A further object of the present invention is a pharmaceutical formulation containing the recombinant human interferon-alpha mutein of the invention, for administration and treatment of diseases such as melanomas, chronic hepatitis C, acute and chronic hepatitis B, acute and chronic non-A, non-B hepatitis, Kaposi's sarcoma, multiple sclerosis, genital warts, leukemia, viral infections, among others. It also comprises the preparation of a powder, gel, cream, lyophilizate, tablets, or a solution to be administered by a route selected from the group consisting of subcutaneous, parenteral, oral, sublingual, intranasal, or topic administration. The formulation, in a preferred solution preparation, comprises concentrations from 0.02 mg/ml to 3 mg/ml of protein mass in solution. Said solution formulations which contain the interferon of the present invention, also contain a buffer, a stabilizer, a cryoprotectant, and a solvent. Useful buffers are those maintaining pH values from 4.5 to 7.5, preferably from 6.5 to 7.0, more preferably pH 6.8. The buffers that may be used are selected from citrate/citric acid, acetate/acetic acid; dibasic/monobasic phosphate, preferably dibasic/monobasic phosphate at a molar concentration comprising from 0.005 to 0.1 molar. The preferred stabilizers used for the invention are selected from poly(oxy-1,2-ethanedyl) derivatives, and among these, more preferably poly(oxy-1,2-ethanedyl) mono-9-octadecenoate, Polysorbate 80, which may be used in a range of concentrations from 0.01 to 1 mg/ml. Cryoprotectants useful for the invention are selected from carbohydrates such as saccharose or mannitol, surfactants such as glycerol, dimethylsulfoxide or Tween. Preferably, the cryoprotectant is a carbohydrate, more preferably saccharose in a range of concentrations from 20 to 100 mg/ml.
A further object of the present invention comprises the use of the recombinant human interferon-alpha mutein of the invention for manufacturing a medicament to be used in a therapeutic protocol including reduced therapeutic doses of IFN alpha 2b. It may be administered as a monotherapy or in the form of a combination therapy with ribavirin; as a monotherapy subcutaneously at a dose from 0.1 micrograms/kg to 2.0 micrograms/kg bodyweight/week. More preferably at a dose from 0.5 micrograms/kg to 1.0 micrograms/kg bodyweight/week. In the case of a combination therapy, a dose from 0.5 to 3.0 micrograms/kg bodyweight/week of the mutein of the invention is administered in combination with ribavirin capsules. More preferably, at doses from 1.0 to 2.0 micrograms/kg bodyweight/week of the mutein of the invention in combination with ribavirin capsules.
The following examples are included to assist further understanding of the present invention, but they should not be construed as a limitation of scope thereof. On the contrary, it should be clearly understood that those of skill in the art might envision other embodiments, modifications, and equivalents without departing from the spirit of the present invention and/or the scope of the appended claims.
The hIFN-α2b gene was obtained by a PCR amplification reaction, using as a template genomic DNA from human peripheric blood leucocytes (Sambrook et al., 1989). For the PCR reaction, specific oligonucleotides hybridizing by both ends to the coding sequence of hIFN-α2b were used, designated IFNalphaF and IFNalphaR, as shown in Table I.
A fragment of 673 pb was cloned into the commercial plasmid pGEM® T Easy Vector (Promega) following the protocol described by the manufacturer. The sequencing of this construct, designated pGEM-rhIFN-α2bwt, confirmed the identity of the cloned fragment, showing a 100% homology to the sequence published in GenBank for hIFN-α2b.
To select the amino acids to be mutated in order to artificially create new N-glycosilation sites, it was decided to carry out the least possible number of changes in the natural hIFN-α2b amino acid sequence. In this way, and keeping in mind the consensus sequence being sought (Asn-Xaa-Ser/Thr), two strategies were developed:
1-Identifying all serines and/or threonines in the natural hIFN-α2b amino acid sequence, the third position of the consensus sequence, and replacing the amino acid occupying the first position by asparagine.
2-Identifying all asparagines present in the natural hIFN-α2b sequence (first position of the consensus sequence) and mutating the amino acid in the third position by a serine or threonine.
Once those amino acids were identified, the resulting glycosylation sites of the corresponding mutations were screened for two basic aspects: high probability of glycosylation and conservation of biological activity of the protein.
The first aspect was evaluated in order to generate a high degree of occupation of the chosen sites. To that end, the following analysis was carried out:
a-Solvent Accessible Surface Area (ASA %) of the amino acid located in the first position of the consensus sequence (Asn or another amino acid mutated by Asn). This study provided information concerning the degree of exposed surface of the amino acid present in the protein structure that would bind the oligosaccharide chain. This parameter was calculated using the ASAview computer program (Ahmed, S., Gromiha, M., Fawereh, H., Sorci, A., BMC Bioinformatics, 2004, 5:51; http://www.netasa.org/asaview/).
b-Probability of glycosylation on the basis of the information from known glycoprotein databases. This parameter is determined using the NetNGlyc 1.0 Server program (www.cbs.dtu.dk/services/NetNGlyc/) created by the Center for Biological Sequence Analysis (CBS, Technical University of Denmark).
In addition, the selected mutations were screened for their capacity for avoiding loss of biological activity of the protein. This aspect was studied taking into account two features:
a-Proximity of the candidate glycosylation site to the molecular region involved in receptor binding. If this parameter is not taken into account, there might exist a high probability of generating glycoproteins having hydrocarbon chains interfering with the binding to the receptor, whereby biologically inactive forms of hIFN-α2b are obtained.
b-Solvent Accessible Surface Area (ASA %): the binding of oligosaccharides to hidden or internal residues of the protein might generate alterations of its tertiary structure, thereby decreasing biological activity.
Possible sites for a site-directed mutagenesis to obtain consensus sequence Asn-Xaa-Ser/Thr are: Leu9, Arg12, Leu66, Phe67, Lys70, Asp71, Phe84, Leu95, Glu113, Arg125, Met148, Ser150, Ser152, Leu153, Gln158 in secondary alpha-helix structures and Pro4, Thr6, Arg23, Leu26, Phe47, Ala50, Asp77, Gly104, Thr106, Lys134, Gln158, Leu161 located outside alpha-helix-type secondary structures.
Among possible sites for a site-directed mutagenesis to obtain consensus sequence Asn-Xaa-Ser/Thr, amino acids Pro4, Arg23, Lys70, and Asp77 were selected (shown in
The site-directed mutagenesis procedure for introducing N-glycosylation sites in the hIFN-α2b gene was performed using overlapping extension PCR, comprising basically 2 consecutive PCR reactions. The description of the technique is summarized in
The oligonucleotides used for the selected point mutations are described in Table II. In this way, nucleotide sequences corresponding to four mutated variants of hIFN-α2b were obtained, where a single N-glycosylation site was incorporated, designated rhIFN-α2bM4, rhIFN-α2bM23, rhIFN-α2bM70, and rhIFN-α2bM77. In these variants, the codons corresponding to amino acids Pro4, Arg23, Lys70, and Asp77 were substituted by a codon codifying for one Asn, respectively.
The resulting DNA fragments were digested with the EcoRI and XbaI restriction enzymes, and cloned into the pCl-neo expression vector, tus obtaining the constructs designated pCl-neo-rhIFN-α2bM4, pCl-neo-rhIFN-α2bM23, pCl-neo-rhIFN-α2bM70, and pCl-neo-rhIFN-α2bM77.
In order to obtain a mutant of rhIFN-α2b with four N-glycosylation sites, designated pCl-neo-rhIFN-α2bM4/23/70/77, an N-glycosylation site was added sequentially using the site-directed mutagenesis technique described above. In this way, a mutant with two N-glycosylation sites was constructed, using a molecule containing one N-glycosylation site as a template. Then a variant with three N-glycosylation sites was constructed using the molecule containing two N-glycosylation sites as a template. Finally, the mutant with four N-glycosylation sites was obtained, by a mutagenesis reaction with the molecule containing three N-glycosylation sites. In each step, each variant in the expression vector pCl-neo was cloned.
All DNA constructs containing the different N-glycosylated variants of rhIFN-α2b were sequenced, which allowed to confirm the success of mutations.
In order to evaluate gene integrity of each human interferon-alpha variant, its expression, level of cytokine glycosylation and specific activity of each variant, transfection of CHO.K1 cells was carried out by a lipofection technique using the plasmids pCl-neo-rhIFN-α2bM4, pCl neo-rhIFN-α2bM23, pCl-neo-rhIFN-α2bM70, pCl-neo-rhIFN-α2bM77 and pCl-neo-rhIFN-α2bM4/M23/M70/M77.
To this end, 3.105 cells.ml−1 CHO.K1 per well were seeded on a 6-well plate using a mammalian cell (MC) culture medium prepared from a 1:1 (V:V) mixture of D-MEM and Ham's F12 media supplemented with 2.441 g/l NaHCO3, anhidrous D(+) glucose 6.6 mM, 1 mM sodium pyruvate, 7.8 mM glutamine, 0.13 mM tryptophan, 0.3 mM aspartic acid, 0.76 mM serine, and 50 μg/ml gentamycin sulphate. The medium was further supplemented with 5% (V/V) bovine foetal serum (BFS). The next day, culture supernatants were removed and the cell layer was washed using MC medium. At the same time, different mixtures containing 10 μg of cathionic lipid (LipofectAMINE® 2000, Invitrogen) and 6 μg of the corresponding plasmid DNA diluted in MC medium were prepared. These preparations were incubated for 30 minutes at room temperature before addition to the cell culture. Then, the washing solution was removed from each well and the different liposome-containing solutions were added and further incubated for 4 hours. Once incubated, the culture supernatant was removed and culture medium supplemented with 5% (V/V) SFB was added. After 72 hours of incubation, the culture supernatant was harvested in order to evaluate the expression of glycosylated variants of rhIFN-α2b.
Expression and glycosylation level of the different isoforms were screened using ELISA sandwich and Western Blot techniques, respectively.
Flat bottom 96-well polystyrene plates were coated with 100 μl of a solution of 1 μg.ml−1 mAb (100 ng per well) diluted in a 50 mM sodium carbonate/bicarbonate solution, pH 9.6. Plates were incubated for 1 hour at 37° C. and overnight at 4° C. After washing 6 times with phosphate saline solution (PBS), 0.05% (V/V) Tween 20, plates were blocked with 200 μl PBS with the addition of 1% (W/V) bovine serum albumin (BSA). Then, 100 μl of 1:2 dilutions of non-glycosylated rhIFN-α2b (standard protein) from 25 ng.ml−1 to 0.39 ng.ml−1 were added using PBS, 0.1% (W/V) BSA, 0.05% (V/V) Tween 20 (dilution solution), followed by incubation for 1 hour at 37° C. After washing, 100 μl of a solution of anti-rabbit rhIFN-α2b polyclonal antibodies diluted 1:2.000 with dilution solution were added. Plates were incubated for 1 hour at 37° C. Finally, 100 μl of peroxidase-conjugated goat anti-rabbit immunoglobulin antibodies diluted 1:1.000 with dilution solution were added, followed by incubation for 1 hour at 37° C. After washing the plates, they were developed using as a substrate 0.12 volumes of H2O2 diluted in 50 mM phosphate citrate solution, pH 5.3, with the addition of O-phenylenediamine at a concentration of 3 mg.ml−1. 100 μl of development solution were added to each well and plates were incubated in the dark at room temperature. After incubating the plates for 15 minutes, the color generated by the solution in each well was measured by absorbance at 450 nm using a microtiter plate reader. The results of the transient transfection experiments showed expression of recombinant rhIFN-α2b for all examined variants at 72-hours post-transfection.
In a first step, a polyacrylamide gel electrophoresis procedure was carried out in the presence of sodium dodecylsulphate reagent (SDS-PAGE) and a sulphide bond reducing agent following substantially the method described by Laemmli (1970). To this end, samples were treated with a solution of 0.05 M Tris-HCl, 2% (W/V) SDS, 10% (V/V) glycerol, 5% (V/V) β-mercaptoethanol, 0.05% (W/V) bromophenol blue, pH 6.8. Samples were incubated at 100° C. for 3 minutes and seeded directly on a stacking gel, having a 5% (W/V) concentration of monomer/branching (acrylamide/bisacrylamide). The separation gel was polymerized with acrylamide/bisacrylamide having a concentration of 15% (P/V). Runs were performed at a constant voltage (200 mV) until the running front reached within 0.5 cm of the bottom of the separation gel. After electrophoresis, proteins were transferred to a nitrocellulose membrane (Bio-Rad). A standard protein transfer protocol was used employing 25 mM Tris, 192 mM glycine, 20% (V/V) methanol, pH 8.3, as transfer solution. Transfer was carried out at a constant current intensity of 180 mA for 1 hour. After transfer was over, the success of the transfer was evaluated detecting the presence of proteins using a solution of 0.25% (W/V) Ponceau Red in 15% (V/V) glacial acetic acid and 40% (V/V) methanol. Then, the dye was eluted by successive washes using Tris-Buffered Saline (TBS) and the membrane was blocked with a solution of TBS, 1% (W/V) BSA. After three successive washes using TBS, the membrane was incubated in a solution of anti-rabbit rhIFN-α2b polyclonal antibodies diluted 1:1.000 with dilution solution (0.1% (P/V) TBS, BSA). After the corresponding washes, the membrane was incubated in a 1:2,000 solution of goat anti-rabbit immunoglobulin antibodies conjugated with peroxydase. All incubations were carried out with stirring for 1 hour at room temperature. Finally, the membrane was developed by a chemoluminiscent reaction using the ECL Plus Western Blotting Reagent commercial kit (GE Healthcare). Luminescence emission was detected by exposure of the membrane to photographic films for a variable time period. Film development was carried out manually employing a conventional fixation and photographic developing method.
All variants containing a single site susceptible to N-glycosylation exhibited a mixture of IFNα muteins having different molecular mass, corresponding to non-glycosylated, O-glycosylated, and N,O-glycosylated fractions, in varying proportions according to the examined mutant. In the particular case of rhIFN-α2bM70 mutein, expression of muteins of higher molecular mass as comparted to the remaining muteins incorporating a single site susceptible to N-glycosylation was observed. Said isoforms would correspond to molecules exhibiting a higher content of glycosidic structures. For rhIFN-α2bM4/M23/M70/M77 mutein, the assay showed a variable glycosylation level with disappearance of the O-glycosylated isoform and a reduced proportion of those corresponding to a low degree of occupation. In addition, a greater concentration of isoforms with a high level of occupation having a molecular mass ranging from 21 and 45 kDa was observed.
In all cases (mutants carrying a single site or 4 sites susceptible to N-glycosylation) the presence of oligosaccharides linked by N-type bonds was confirmed by a specific N-deglycosylation treatment using the PNGase F enzyme (Asparagine amidase, Biolabs). In this manner, all muteins showed matching protein bands corresponding to the O-glycosylated variant, thus confirming the presence of hydrocarbon chains linked through an N-glycosidic bond to the thus incorporated sites.
The different hIFN-α2b molecules, mutated individually at amino acid positions Arg23, Leu47, Lys70, and Leu95 so as to obtain glycosylation sites at positions Arg23, Asn45, Lys70, and Asn93, were tested for stability in the presence of the PNGase F enzyme responsible for the release of oligosaccharides bound through an N-glycosidic bond. To this end, 350 μl of culture supernatant for each of the above-mentioned mutants were treated with 250 U of PGNase F (Biolabs) and incubated at 37° C. The deglycosylation reaction was inhibited at different time intervals (30, 60, 90, 120, 150, 180, and 300 minutes) by taking 45 μl aliquotes from said mixture and incubating them with 15 μl of a 0.05 M Tris-HCl solution, 2% (W/V) SDS, 10% (V/V) glycerol, 5% (V/V) β-mercaptoethanol, 0.05% (W/V) bromophenol blue, pH 6.8, for 5 minutes at 100° C. Once the deglycosylation reaction was completed, all samples were analyzed by a reducing SDS-PAGE technique and then a Western blot analysis substantially following the steps as indicated in example 6.
The percentage of N,O-glycosylated, O-glycosylated, and non-glycosylated isoforms in the samples was estimated by densitometry and then their proportion was calculated using the commercial software ImageMaster TotalLab V1.11 (GE Healthcare, Sweden).
When analyzing the molecular structure of hIFN-α2b, it may be observed that amino acid position Arg23 is located at the beginning of a connecting loop of two α-helix structures and amino acid position Phe47 is located in the central region of another connecting loop, which confers greater susceptibility to the action of a deglycosylating enzyme. On the other hand, positions Arg70 and Leu95 form part of an α-helix. This type of structure might confer a greater resistance to both sites and as a consequence a superior stability to the action of N-glycanases.
Biological activity of the different hIFN-α2b variants was determined by measuring the antiviral effect of the same on cultures of MDBK cells (Madin-Darby Bovine Kidney, ATCC CCL-22) infected with vesicular stomatitis virus (VSV), Indiana strain (ATCC VR-158). To this end, flat bottom 96-well plates were seeded with 100 μl of a cellular suspension containing 250,000 cell.ml−1 using MEM culture medium supplemented with 2 mM L-glutamine, 2.2 mg.ml−1 sodium bicarbonate, 50 μg.ml−1 gentamycin, and 10% (V/V) bovine foetal serum (growth medium).
Plates were incubated overnight at 37° C. and 5% CO2, saturated humid atmosphere. The next day, culture supernatant was removed and 100 μl of consecutive 2-fold dilutions of non-glycosylated rhIFN-α2b (standard) from 20 U.ml−1 to 0.156 U.ml−1 were added and the corresponding test samples at appropriate dilutions. All dilutions were carried out in a culture medium identical to the growth medium with a concentration of 2% (V/V) SFB (assay medium). Plates were incubated in an oven for 6 hours. After this, supernatant was removed and plates were washed using assay medium. Finally, 100 μl of VSV virus suspension prepared in assay medium were added. A viral dilution capable of generating 100% cytopathic action after 20-24 hours was used as working dilution. Plates were incubated overnight 37° C. and 5% CO2, saturated humid atmosphere. The next day, when a cytopathic action of about 100% in control wells in absence of IFN was observed, culture supernatant was removed by inversion of the plates and development thereof was carried out using 50 μl of 0.75% (W/V) crystal violet solution in 40% (V/V) methanol. Plates were incubated at 37° C. for 15 minutes. The dye was removed and plates were washed with distilled water until absence of dye was verified in the washes. After blotting the plates, 250 μl of 20% (V/V) acetic acid solution were added. Plates were homogeneized until each well showed a uniform color, and finally color absorbance was determined by spectrophotometric measuring at λ=540 nm.
The specific biological activity parameter (ABE) was determined for each molecule (Table III). rhIFN-α2bM4, rhIFN-α2bM23, and rhIFN-α2bM77 mutants showed variable levels of specific activity within a range from 160 to 290 U.ng−1. In this manner, incorporation of mutations in the protein sequence and/or the presence of N-type hydrocarbon chains did not alter significantly the activity of said variants as compared to the specific biological activity of non-glycosylated rhIFN-α2b (200 U−1). In the case of rhIFN-α2bM70 mutein a 48% decrease of the above-mentioned parameter was evident. However, considering that muteins IFN-α2bM4, IFN-α2bM23 and IFN-α2bM77 conserved their specific biological activity, and a higher level of glycosylation of mutein rhIFN-α2bM70, the corresponding mutein carrying the 4 N-glycosylation susceptible sites indicated previously was obtained. When assessing its specific biological activity (26 U.ng−1), we observed an 87% decrease as compared to the non-glycosylated molecule.
CHO.K1 cells were transiently transfected using different constructs essentially following the stages as described in Example 4 for obtaining N-glycosylated rhIFN-α2b-producing stable cell lines. Then, the cultures were subjected to selection pressure using the antibiotic Neomycin (200 μg.ml−1) from 72 to 168 hours post-transfection. After two weeks of culture in the presence of the antibiotic, cytokine expression levels were quantified by a sandwich ELISA assay and producing lines (CHO pCl-neo-rhIFN-α2bM4, CHO pCl-neo-rhIFN-α2bM23, CHO pCl-neo-rhIFN-α2bM70, CHO pCl-neo-rhIFN-α2bM77, and CHO pCl-neo-rhIFN-α2bM4/23/70/77) were kept in liquid nitrogen. Then, the cell lines were cloned using limiting dilution methods. Table IV shows details of selected clones and their specific productivity in the stationary phase. The clones of each cell line showing higher productivity are shown in bold. They were used later in the production phase of each mutated rhIFN-α2b variant.
M42F8
45.43
M231D7
106.93
M701E3
42.55
M771C6
76.29
3D2F5
46.50
In a first step, the specific productivity of rhIFN-α2b during different cell culture phases was determined, indicating the clones showing greater specific productivity of the cytokine during the stationary phase.
Secondly, different concentrations of SFB were assessed as culture medium supplements to reduce protein content in supernatants and facilitate further purification of rhIFN-α2b variants. To this end, clones were cultured using culture medium (MC) supplemented with 5% (V/V) SFB and, when they reached a stationary phase, the conditioned supernatant was replaced by fresh culture medium containing different serum concentrations: 5%, 1%, 0.5%, and 0.1% (V/V). After 24 hours of culture, a drastic drop of specific productivity was observed when changing SFB concentration from 5% to 1% (V/V). However, specific productivity using from 1% to 0.5% (V/V) SFB concentrations was similar, and therefore the latter serum concentration was used for carrying out the production. By this, a 10-fold reduction of protein content was achieved in the samples to be purified.
Finally, an experiment for determining the number of culture medium changes and optimum time-interval between them was carried out in order to obtain the greatest recombinant protein mass as possible. A total of eight harvests were carried out, and their corresponding changes of culture medium, every 24 hours and four harvests every 48 hours in stationary phase cultures. It was found that total accumulated mass for rhIFN-α2b was similar in both experiments, and a higher concentration of cytokine in supernatants harvested every 48 hours was obtained.
In all the above-mentioned experiments, the quality of the proteins thus produced was evaluated and it was shown that, irrespective of the assay conditions, the specific biological activity and isoform molecular mass profiles for the different molecules were conserved.
The different rhIFN-α2b variants were obtained using 500 cm2 culture flasks. To this end, approximately 2.105 cell.ml−1 were seeded using MC culture medium supplemented with 5% (V/V) SFB. When the culture reached the stationary phase of growth, sequential changes every 48 and 72 hours were performed using MC culture medium supplemented with 0.5% (V/V) SFB. Culture supernatants were centrifuged at 3,000 r.p.m for 10 minutes and stored at −20° C. for further purification of the corresponding variants.
The different mutated variants of rhIFN-α2b were purified by an immunoaffinity chromatography method, using the mAb selected as capturing antibody as a ligand in a sandwich ELISA assay.
In the first place, 10 female BALB/c mice of 2 months of age were used for the in vivo production of mAb in ascitic liquid. To this end, 0.5 ml of 2,6,10,14-tetramethyldecanoic acid (Pristane®, Sigma) were inoculated intraperitonially in each mouse and after 10 days 2.106 hybridoma cells per animal resuspended in 0.5 ml of PBS were inoculated by the same route. The ascitic liquid was collected by intraperitoneal puncture using a 1.2 mm gauge needle, from the first week after inoculation of hybridomas. The concentration of murine immunoglobulins in the ascitic fluid was determined using a sandwich ELISA assay.
Purification of mAb was carried out by affinity chromatography using a Sepharose High Trap protein A 5 ml column (GE Healthcare), following the method described by Harlow and Lane (1998). Pure immunoglobulin concentration was calculated by spectrophotometric readings at 280 nm using a percent extinction coefficient) (E1cm1% of 12.5. Multiplying 10×(E1cm1%)−1 by optical density readings is equal to mAb concentration expressed in mg.ml−.
The affinity ligand was coupled to a Sepharose 4B matrix activated with cyanogen bromide following a standard protocol (GE Healthcare). The percentage of mAb coupling to the resin was of 98.3% and theorical matrix capacity was of 186 μg of non-glycosylated rhIFN-α2b per ml of gel.
The following purification protocol was used for rhIFN-α2b muteins with one N-glycosylation site and with four N-glycosylation sites. Culture supernatants containing cytokines were worked up by addition of Triton X-100 until a concentration of 0.3% (V/V) was achieved. The worked-up samples were loaded onto a column previously adjusted using a solution of 0.1 M Tris, 0.3% (V/V) (pH 7.5) Triton X-100 at a rate of 0.5 ml.min−1. Then, it was washed with 5 column volumes using each of the following solutions:
1-solution of 0.5 M NaCl, 0.2% (V/V) Triton X-100 in 0.025 M Tris (pH 7.5).
2-solution of 0.15 M NaCl in water (pH 5.0).
Elution of rhIFN-α2b was assessed using three different solutions:
1-solution of 0.15 M NaCl in 0.2 M acetic acid (pH 3.0)
2-solution of 0.1 M glycine (pH 2.5)
3-solution of 0.1 M glycine (pH 2.0).
2 ml fractions were collected. Each fraction was neutralized by addition of a solution of 1 M Tris (pH 9.0). The matrix was stored in 0.02% (W/V) azide solution in 0.025 M Tris (pH 7.5).
The presence of rhIFN-α2b was evaluated in all the fractions by a sandwich ELISA assay, achieving a higher recovery percent (ranging from 95% to 100%) by using a 0.1 M glycine solution (pH 2.0).
Purification efficiency was studied by SDS-PAGE analyses under reductive conditions with additional silver staining and Western blot assays (
Pharmacokinetic properties of variants having a N-glycosylation site located outside α-helix regions (rhIFN-α2bM23) and forming part of an α-helix (rhIFN-α2bM70) were evaluated using a subcutaneous inoculation route. To that end, female Wistar rats, of two months of age, weighing 200 g on average, were inoculated with the indicated IFN variants using a single dose of 5×105 U.Kg−1/body weight of the animal. Then, blood samples were collected by puncture of the retroorbital vein using heparinized capillary tubes at different post-injection times and the presence of the cytokine was determined quantifying its volumetric biological activity. Data thus obtained were used to plot biological activity versus time (
The area under the biological activity curves (AUC), calculated as a function of time corresponding to the glycosylated rhIFN-α2bM70 mutein showed an 1.63-fold increase as compared to the rhIFN-α2bM23 variant.
13-a)-Intravenous Inoculation Route
In order to evaluate the influence of glycosidic moieties on the pharmacokinetic parameters of glycosylated rhIFN-α2b variants, the rhIFN-α2bM77 variant (representing molecules having a single N-glycosylation site located outside an alpha-helix structure) and the rhIFN-α2bM4/23/70/77 mutein were selected. Studies were carried out by comparison to the rhIFN-α2b molecule produced in bacteria (non-glycosylated) and to versions covalently conjugated to a molecule of 12 kDa polyethylenglycol (PEG) (rhIFN-α2b-PEG 12 kDa; Schering Plough). To that end, female Wistar rats, of two months of age, having an average weight of 200 g, were inoculated with a single dose of 52.6 pmol of each IFN variant per animal using, in the first place, an intravenous administration route (tail vein). In this assay, batches of 4 animals per each variant were used. Then, blood samples were collected by puncture of the retroorbital vein using heparinized capillary tubes at different post-injection times. They were used to determine the presence of the cytokine quantifying its volumetric biological activity (AB) by the above-mentioned biological activity in vitro assay. Data thus obtained were used to calculate the percent of remnant biological activity in each sample as compared to biological activity units of IFN administered to each rat and said percentage was plotted versus time from inoculation (
The behavior of the different molecules was appropriately adjusted to a bicompartimental model, according to the following equation:
C=A.e
−t/t1
+B.e
−t/t2,
wherein C is the fraction of circulating remnant drug at time t, A and t1 are initial phase parameters reflecting protein distribution to extravascular body fluids, whereas B and t2 are parameters characterizing the terminal phase of drug removal.
Thus, a regression adjustment was carried out for each curve, and pharmacokinetic parameters as indicated in Table V were determined.
As may be seen from this Table, half-life times of the initial phase (proportional to t1) of non-glycosylated rhIFN-α2b and rhIFN-α2bM77 molecules are similar to each other, but lower as compared to values calculated for the rhIFN-α2bM4/23/70/77 and rhIFN-α2b-PEG variants. In addition, half-life times corresponding to the removal phase was higher for the rhIFN-α2bM4/23/70/77 muteins.
Taking into account non-glycosylated and glycosylated IFN variants, a reduction of plasmatic clearance was observed as hydrocarbon content increased. In the particular case of the rhIFN-α2bM4/23/70/77 mutein, a 9-fold reduction of said parameter was visualized as compared to the non-modified rhIFN-α2b molecule. In addition, no significant differences were observed when calculating plasmatic clearance corresponding to rhIFN-α2bM4/23/70/77 and rhIFN-α2b-PEG molecules.
13 b)-Subcutaneous Inoculation Route
In addition, pharmacokinetic properties of the mutein having four N-glycosylation sites (IFN-α2bM4/23/70/77) were evaluated using a subcutaneous inoculation route. To that end, female Wistar rats, of two months of age, having an average weight of 200 g, were inoculated with the indicated IFN mutein using a single dose of 5×105 U.Kg−1 of body weight. At the same time, animals were inoculated with identical doses of non-glycosylated rhIFN-α2b or rhIFN-α2b-PEG 12 kDa. Then, blood samples were collected by puncture of the retroorbital vein using heparinized capillary tubes at different post-injection times. They were used to determine the presence of the cytokine by quantifying its volumetric biological activity by the above-mentioned in vitro assay of biological activity. The values thus obtained were used to plot biological activity in each sample as a function of time (
The area under the biological activity curve (AUC) calculated as a function of time corresponding to the N-glycosylated rhIFN-α2bM4/23/70/77 variant showed a 37-fold increase as compared to the non-glycosylated cytokine variant. At the same time, glycosylated cytokine reached a maximum biological activity concentration from the first hour of post-injection, remaining approximately constant for 10 hours. Said maximum concentration was higher than that achieved by the non-glycosylated variant, evidenced as a peak concentration after 30 minutes of post-inoculation. Finally, biological activity levels of the latter variant became non-detectable after 4 hours of post-injection.
| Number | Date | Country | Kind |
|---|---|---|---|
| 060102627 | Jun 2006 | AR | national |
This application is a U.S. National Phase Application of PCT International Application No. PCT/ES2007/070117, filed Jun. 15, 2007.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/ES07/70117 | 6/15/2007 | WO | 00 | 8/26/2009 |
