Patent Grant
7897743
This invention relates to the field of molecular immunology, generally, and to vectors useful for expression of proteins, especially antibodies, such as fully human, humanized, and chimeric antibodies, as well as fusion proteins which incorporate the antibody and a protein or protein fragment, in eukaryotic cells, mammalian cells in particular. The resulting antibodies and fusion proteins are also a feature of the invention.
One serious problem with using murine antibodies for therapeutic applications in humans is that they quickly raise a human anti-mouse response (HAMA) which reduces the efficacy of the antibody in patients, and prevents continued administration thereof. Parallel issues arise with the administration of antibodies from other, non-human species. One approach to overcoming this problem is to generate so-called “chimeric” antibodies. These can comprise murine variable regions, and human constant regions (Boulianne et al. (1984) Nature 312(5995): 643-646.; incorporated by reference herein in its entirety). Although chimeric antibodies contain murine sequences and can elicit an anti-mouse response in humans (LoBuglio et al. (1989) Proc. Natl. Acad. Sci. USA 86(11): 4220-4224; incorporated by reference herein in its entirety), trials with chimeric antibodies in the area of hematological disease (e.g., Non-Hodgkin-Lymphoma; Witzig et al. (1999) J. Clin. Oncol. 17(12): 3793-3803; incorporated by reference herein in its entirety) or autoimmune disease (e.g., rheumatoid arthritis, chronic inflammatory bowel disease; Van den Bosch; et al, Lancet 356(9244):1821-2 (2000), incorporated by reference herein in its entirety) have led to FDA approval and demonstrate that these molecules have significant clinical potential and efficacy.
Recent studies have indicated that granulocyte-macrophage colony stimulating growth factor (GM-CSF) plays a role in the development of rheumatoid arthritis (RA) (Cook, et al., Arthritis Res. 2001, 3:293-298, incorporated by reference herein in its entirety) and possibly other inflammatory diseases and conditions. Therefore, it would be of interest to develop a drug which would block GM-CSF and its effect on cells. The present invention provides a chimeric antibody, targeting the GM-CSF molecule, which has blocking capacity.
The increased use of chimeric antibodies in therapeutic applications has created the need for expression vectors that effectively and efficiently produce high yields of functional chimeric antibodies in eukaryotic cells, such as mammalian cells, which are preferred for production. The present invention provides novel expression vectors, transformed host cells and methods for producing chimeric antibodies in mammalian cells, as well as the antibodies themselves and fusion proteins containing them.
The present invention provides expression vectors which are useful in the expression of proteins, such as antibodies, especially fully human, humanized or chimerized antibodies, and fusion proteins containing these. Both light chains and heavy chains can be expressed. The expression vectors of the present invention comprise a human elongation factor 1 α (EF1α) promoter/enhancer sequence, an internal ribosome entry site (IRES) sequence (U.S. Pat. No. 4,937,190; incorporated herein in its entirety), a nucleotide sequence that confers neomycin resistance to a cell containing the expression vector, and a nucleotide sequence under control of a simian virus 40 promoter (SV40) that confers ampicillin resistance to a cell containing the expression vector. In a preferred embodiment, the EF1α promoter/enhancer sequence is upstream and adjacent to a nucleotide sequence encoding a chimeric light chain.
The expression vector of the present invention may contain a nucleotide sequence encoding any immunoglobulin light chain. In a preferred embodiment the light chain variable region is of murine origin, and the light chain constant region is either human kappa or human lambda. In a more preferred embodiment, the chimeric light chain variable region is derived from a murine antibody that binds to GM-CSF, CD-30, or G250 and in especially preferred embodiments, to the human forms of these molecules.
The present invention also provides a further expression vector useful in the expression of proteins, such as antibodies, especially fully human, humanized or chimeric antibodies, and fusion proteins containing these. This second embodiment differs from the first in that instead of the neomycin resistance sequence, described supra, it comprises a nucleotide sequence which encodes dihydrofolate reductase or “dhfr,” which generates resistance against the well known selection marker methotrexate. Such an expression vector may contain nucleotide sequences encoding any antibody or portion thereof, such as heavy or light chains of fully human, humanized or chimerized antibodies. In a preferred embodiment, a heavy chain is expressed, where the variable region is of murine origin, and the heavy chain constant region is human IgG1. In a more preferred embodiment, the chimeric heavy chain variable region is derived from a murine antibody that binds CD-30, GM-CSF or G250, preferably the human forms of these.
In another embodiment, the present invention provides host cells transformed or transfected with any one of the expression vectors of the present invention. In a preferred embodiment, a host cell, preferably a eukaryotic cell, more preferably a mammalian cell, is transformed or transfected with an expression vector comprising a chimeric immunoglobulin light chain and an expression vector comprising a chimeric immunoglobulin heavy chain. The present invention contemplates prokaryotic and eukaryotic cells, such as mammalian cells, insect cells, bacterial or fungal cells. In a preferred embodiment, the host cell is a human or Chinese Hamster Ovary (“CHO”) cell.
The present invention also provides methods for the recombinant production of a chimeric immunoglobulin light or heavy chain comprising the step of culturing a transformed or transfected host cell of the present invention. In one embodiment, the methods of the present invention further comprise the isolation of the chimeric immunoglobulin light or heavy chain.
The present invention also provides methods for the recombinant production of a fully human, humanized or chimeric immunoglobulin comprising culturing a host cell that has been transformed or transfected with an expression vector comprising a chimeric immunoglobulin light chain and an expression vector comprising a chimeric immunoglobulin heavy chain, or an expression vector encodes both chains. In one embodiment, the methods of the present invention further comprise the self-assembly of the chimeric heavy and light chain immunoglobulins and isolation of the chimeric immunoglobulin. Methods for accomplishing this are well known in the art.
The present invention also provides the chimeric immunoglobulin light chain, heavy chain or assembled chimeric immunoglobulin produced by the methods of the present invention. In another embodiment, the present invention provides compositions comprising the chimeric immunoglobulin light chain, heavy chain or assembled chimeric immunoglobulin of the present invention and a pharmaceutically acceptable carrier.
As used herein “chimerized” refers to an immunoglobulin such as an antibody,
wherein the heavy and light chains of the variable regions are not of human origin and wherein the constant regions of the heavy and light chains are of human origin.
“Humanized” refers to an immunoglobulin such as an antibody, wherein the amino acids directly involved in antigen binding, the so-called complementary determining regions (CDR), of the heavy and light chains are not of human origin, while the rest of the immunoglobulin molecule, the so-called framework regions of the variable heavy and light chains, and the constant regions of the heavy and light chains are of human origin.
“Fully human” refers to an immunoglobulin, such as an antibody, where the whole molecule is of human origin or consists of an amino acid sequence identical to a human form of the antibody.
“Immunoglobulin” or “antibody” refers to any member of a group of glycoproteins occurring in higher mammals that are major components of the immune system. As used herein, “immunoglobulins” and “antibodies” comprise four polypeptide chains—two identical light chains and two identical heavy chains that are linked together by disulfide bonds. An immunoglobulin molecule includes antigen binding domains, which each include the light chains and the end-terminal portion of the heavy chain, and the Fc region, which is necessary for a variety of functions, such as complement fixation. There are five classes of immunoglobulins wherein the primary structure of the heavy chain, in the Fc region, determines the immunoglobulin class. Specifically, the alpha, delta, epsilon, gamma, and mu chains correspond to IgA, IgD, IgE, IgG and IgM, respectively. As used herein “immunoglobulin” or “antibody” includes all subclasses of alpha, delta, epsilon, gamma, and mu and also refers to any natural (e.g., IgA and IgM) or synthetic multimers of the four-chain immunoglobulin structure.
“Antigen-binding fragment”, “antigen-binding domain” and “Fab fragment” all refer to the about 45 kDa fragment obtained by papain digestion of an immunoglobulin molecule and consists of one intact light chain linked by a disulfide bond to the N-terminal portion of the contiguous heavy chain. As used herein, “F(ab)2 fragment” refers to the about 90 kDa protein produced by pepsin hydrolysis of an immunoglobulin molecule. It consists of the N-terminal pepsin cleavage product and contains both antigen binding fragments of a divalent immunoglobulin, such as IgD, IgE, and IgG. Neither the “antigen-binding fragment” nor “F(ab)2 fragment” contain the about 50 kDa Fc fragment produced by papain digestion of an immunoglobulin molecule that contains the C-terminal halves of the immunoglobulin heavy chains, which are linked by two disulfide bonds, and contain sites necessary for compliment fixation.
“Epitope” refers to an immunological determinant of an antigen that serves as an antibody-binding site. Epitopes can be structural or conformational.
“Hybridoma” refers to the product of a cell-fusion between a cultured neoplastic lymphocyte and a normal, primed B- or T-lymphocyte, which expresses the specific immune potential of the parent cell.
“Heavy chain” refers to the longer & heavier of the two types of polypeptide chain in immunoglobulin molecules that contain the antigenic determinants that differentiate the various Ig classes, e.g., IgA, IgD, IgE, IgG, IgM, and the domains necessary for complement fixation, placental transfer, mucosal secretion, and interaction with Fc receptors.
“Light chain” refers to the shorter & lighter of the two types of polypeptide chain in an Ig molecule of any class. Light chains, like heavy chains, comprise variable and constant regions.
“Heavy chain variable region” refers to the amino-terminal domain of the heavy chain that is involved in antigen binding and combines with the light chain variable region to form the antigen-binding domain of the immunoglobulin.
“Heavy chain constant region” refers to one of the three heavy chain domains that are carboxy-terminal portions of the heavy chain.
“Light chain variable region” refers to the amino-terminal domain of the light chain and is involved in antigen binding and combines with the heavy chain to form the antigen-binding region.
“Light chain constant region” refers to the one constant domain of each light chain. The light chain constant region consists of either kappa or lambda chains.
“Murine anti-human-GM-CSF 19/2 antibody” refers to a murine monoclonal antibody that is specific for human GM-CSF. This antibody is well known and it has been studied in detail. See Dempsey, et al, Hybridoma 9:545-58 (1990); Nice, et al, Growth Factors 3:159-169 (1990), both incorporated by reference.
“Effective amount” refers to an amount necessary to produce a desired effect.
“Antibody” refers to any glycoprotein of the immunoglobulin family that non-covalently, specifically, and reversibly binds a corresponding antigen.
“Monoclonal antibody” refers to an immunoglobulin produced by a single clone of antibody-producing cells. Unlike polyclonal antiserum, monoclonal antibodies are monospecific (e.g., specific for a single epitope of a single antigen).
“Granulocytes” include neutrophils, eosinophils, and basophils.
“GM-CSF” refers to a family of glycoprotein growth factors that control the production, differentiation, and function of granulocytes and monocytes-macrophages. Exemplary, but by no means the only form of such molecules, can be seen in U.S. Pat. No. 5,602,007, incorporated by reference.
“Inflammatory condition” refers to immune reactions that are either specific or non-specific. For example, a specific reaction is an immune reaction to an antigen. Examples of specific reactions include antibody responses to antigens, such as viruses and allergens, including delayed-type hypersensitivity, including psoriasis, asthma, delayed type hypersensitivity, inflammatory bowel disease, multiple sclerosis, viral pneumonia, bacterial pneumonia, and the like. A non-specific reaction is an inflammatory response that is mediated by leukocytes such as macrophages, eosinophils and neutrophils. Examples of non-specific reactions include the immediate swelling after a bee sting, and the collection of polymorphonuclear (PMN) leukocytes at sites of bacterial infection. Other “inflammatory conditions” within the scope of this invention include, e.g., autoimmune disorders such as psoriasis, rheumatoid arthritis, lupus, post-ischemic leukocyte mediated tissue damage (reperfusion injury), frost-bite injury or shock, acute leukocyte-mediated lung injury (acute respiratory distress syndrome or ARDS), asthma, traumatic shock, septic shock, nephritis, acute and chronic inflammation, and platelet-mediated pathologies such as ateriosclerosis and inappropriate blood clotting.
“Pharmaceutically acceptable carrier” refers to any carrier, solvent, diluent, vehicle, excipient, adjuvant, additive, preservative, and the like, including any combination thereof, that is routinely used in the art.
Physiological saline solution, for example, is a preferred carrier, but other pharmaceutically acceptable carriers are also contemplated by the present invention. The primary solvent in such a carrier may be either aqueous or non-aqueous. The carrier may contain other pharmaceutically acceptable excipients for modifying or maintaining pH, osmolarity, viscosity, clarity, color, sterility, stability, rate of dissolution, and/or odor. Similarly, the carrier may contain still other pharmaceutically acceptable excipients for modifying or maintaining the stability, rate of dissolution, release, or absorption or penetration across the blood-brain barrier.
The fully human, humanized or chimerized antibodies of the present invention may be administered orally, topically, parenterally, rectally or by inhalation spray in dosage unit formulations that contain conventional non-toxic pharmaceutically acceptable carriers, adjuvants and vehicles. As used herein, “parenterally” refers to subcutaneous, intravenous, intramuscular, intrasternal, intrathecal, and intracerebral injection, including infusion techniques.
The fully human, humanized or chimerized antibodies may be administered parenterally in a sterile medium. The antibodies, depending on the vehicle and concentration used, may be suspended or dissolved in the vehicle. Advantageously, adjuvants such as local anesthetics, preservatives and buffering agents can be dissolved in the vehicle. The most preferred routes of administration of the pharmaceutical compositions of the invention are subcutaneous, intramuscular, intrathecal or intracerebral administration. Other embodiments of the present invention encompass administration of the composition in combination with one or more agents that are usually and customarily used to formulate dosages for parenteral administration in either unit dose or multi-dose form, or for direct infusion.
Active ingredient may be combined with the carrier materials in amounts necessary to produce single dosage forms. The amount of the active ingredient will vary, depending upon the type of antibody used, the host treated, the particular mode of administration, and the condition from which the subject suffers. Preferably, the amount of fully human, humanized or chimerized anti-GM-CSF immunoglobulin, for example, is a therapeutically effective amount which is sufficient to decrease an inflammatory response or ameliorate the symptoms of an inflammatory condition. It will be understood by those skilled in the art, however, that specific dosage levels for specific patients will depend upon a variety of factors, including the activity of the specific immunoglobulins utilized, the age, body weight, general health, sex, diet, time of administration, route of administration, and rate of excretion, drug combination and the severity of the particular disease undergoing therapy. Administration of the fully human, humanized or chimerized immunoglobulins of the present invention may require either one or multiple dosings.
Regardless of the manner of administration, however, the specific dose is calculated according to approximate body weight or body surface area of the patient. Further refinement of the dosing calculations necessary to optimize dosing for each of the contemplated formulations is routinely conducted by those of ordinary skill in the art without undue experimentation, especially in view of the dosage information and assays disclosed herein.
Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and the following illustrative examples, make and utilize the compounds of the present invention and practice the claimed methods. The following working examples therefore, specifically point out preferred embodiments of the present invention, and are not to be construed as limiting in any way the remainder of the disclosure.
Total RNA from the hybridoma producing murine 19/2 antibody was obtained by standard RNA isolation techniques (Chomczynski et al. (1987) Anal. Biochem. 162: 156-159; incorporated by reference herein in its entirety). First strand cDNA was prepared using a commercially available, first strand cDNA synthesis kit and priming with d(T)18 for both the heavy and light chains (Renner et al. (1998) Biotechniques 24(5): 720-722; incorporated by reference herein in its entirety). The resulting cDNA was subjected to PCR using combinations of primers for the heavy and light chains. The nucleotide sequences of the 5′ primers for the heavy and light chains are shown in Tables 1 and 2 respectively. The 3′ primers are shown in Table 3. The light chain primer hybridized within the mouse kappa constant region not far from the V-C junction. The heavy chain 3′ primer hybridised within the CH-1 constant region of mouse heavy chain subgroup 1 not far from the V-CH1 junction.
Using the cloning strategy described, supra, PCR products for VH and VL of murine 19/2 were cloned using a commercially available product, and art recognized techniques. For the murine 19/2 VL region, PCR products were obtained using the mouse kappa constant region primer and primers MKV2 and MKV7 (SEQ ID NOS: 14 & 19). For the mouse 19/2 VH region, PCR products were obtained using the mouse gamma 1 constant region primer and primers MHV2, MHV5 and MHV7 (SEQ ID NOS: 2, 5 and 7). Extensive DNA sequencing of the cloned V-region inserts revealed two different light chain sequences and 2 different heavy chain sequences. Pseudogenes for heavy and light chain were amplified and were eliminated by standard sequence analyses. A novel immunoglobulin-coding sequence was determined for both the heavy and light chains. This is set forth at SEQ ID NOS: 27, 28, 29 & 30, which present the cDNA and amino acid sequences for the murine 19/2 heavy chain variable region (27 & 28), and the light chain variable region (29 & 30).
When comparing the DNA sequence of the leader sequence for 19/2 heavy chain obtained with the primers described supra, with the database, it appeared that the 19/2 HC leader sequence is short (17 amino acids) and unique vis a vis public data bases. Specifically, amino acids 2, 3 and 5 were E, L & M, as compared to S, W & F in the data bases. As compared to the database, hydrophilic amino acids in the N-terminal region were separated by neutral or basic ones, respectively; however, since the influence of these changes on the secretory capability of the leader sequence is unclear, this sequence was unaltered in further experiments.
The chimeric 19/2 antibody was designed to have the mouse 19/2 VL and VH regions linked to human kappa and gamma-1 constant regions, respectively. PCR primers were used to modify the 5′- and 3′-sequences flanking the cDNA sequences coding for the mouse 19/2 VL and VH regions. PCR primers specific for 19/2 light chain V-region were designed using the sequence of the 19/2 light chain V-region gene obtained. These adapted mouse 19/2 variable regions were then subcloned into mammalian cell expression vectors already containing the human kappa (pREN-Neo vector) or the gamma-1 (pREN-DHFR vector) constant regions. The vectors employ parts of the human elongation factor 1α (EF1α) promoter/enhancer sequence to efficiently transcribe the light and heavy chains. The vectors also contain an IRES sequence following the multiple cloning site to allow for stringent, bicistronic expression and control of the individual selection marker in CHO cells. This pair of vectors was used in all of the recombinant work described herein, i.e., to manufacture all chimeric antibodies. The expression vectors were designed to have the variable regions inserted as PmeI-BamHI DNA fragments. PCR primers were designed to introduce these restrictions sites at the 5′- (PmeI) and 3′- (BamHI) ends of the cDNAs coding for the V-regions. In addition, the PCR primers were designed to introduce a standard Kozak sequence (Kozak (1987) Nucleic Acids Res. 15(20): 8125-8148, incorporated by reference herein in its entirety) at the 5′-ends of both the light and heavy chain cDNAs to allow efficient translation, and to introduce splice donor sites at the 3′-ends of both the light and heavy chain cDNAs for the variable regions to be spliced to the constant regions. The PCR primers used for the construction of the chimeric 19/2 light and heavy chains were as follows: catgtttaaacgccfccaccatgggcttcaagatggagtca (5′ end, light chain variable region, SEQ ID NO: 31); agaggatccactcacgtttcagttccacttggtcccag (3′ end, SEQ ID NO: 32); catgtttaaacgccgccaccatggagctgatcatgctcttcct (primer for the 5′ end of the heavy chain variable region, SEQ ID NO: 33); and agaggatccactcacctgaggagactctgagagtggt (primer for the 3′ end of the heavy chain variable region, SEQ ID NO: 34). The DNA and amino acid sequences of the mouse 19/2 VL and VH regions were adapted for use from the construction of chimeric 19/2 light and heavy chains. The entire DNA sequences of mouse 19/2 light and heavy chains cloned into the eukaryotic expression vectors pREN-Neo and pREN-DHFR, respectively, are set forth as SEQ ID NO: 35 & 36, with the resulting light and heavy chains resulting in chimerized molecules. Specifically, in SEQ ID NO: 35, nucleotides 1357-1756 encode the murine, light chain sequence, with nucleotides 1763-2206 encoding the human kappa region. Within this sequence (1763-2206), a 120 base pair region constituting an intron and splice acceptor site begins at nucleotide 1886. Within SEQ ID NO: 36, nucleotides 1357-1770 encode the murine 9/2 heavy chain constant sequence with a splice donor site. Nucleotides 1777-2833 encode the human IgG1 constant region. Within this sequence, there is a 60 base pair intron region and splice acceptor site which precedes the coding region.
The objective of the experiments described herein was to create stable cell lines expressing chimeric 19/2 (c19/2) anti-human GM-CSF monoclonal antibodies (mAb) in CHO (Chinese hamster ovary) DG44 cells and to test the secreted antibody for its binding properties. To do this, the DHFR negative CHO cell line DG044 was used. See Morris et al. (1990) Gene 94(2): 289-294; incorporated by reference herein in its entirety). The CHO cells were cultured in RPMI, supplemented with 10% FCS and Hypoxanthine-Thymidine. DNA for transfection was purified from E. coli cells using a commercially available product, and the instructions provided therein. All DNA preparations were examined by restriction enzyme digestion. Sequences of chimeric 19/2 mAb variable regions in their respective vectors were confirmed using an ABI PRISM 310 or LICOR Sequencer.
Vectors encoding heavy and light chains of chimeric 19/2 mAbs were co-transfected simultaneously into CHO DG44 cells growing at log phase, using electroporation (270V, 975 uF). Cells were plated in 10 cm dishes and cultured with standard medium. Twenty-four hours later, medium was harvested and replaced by fresh RPMI medium supplemented with 10% dialyzed FCS and 500 ug/mL geneticin. After the initial phase of cell killing was over (7-10 days), GMP-grade methotrexate was added at a concentration of 5 nM and gradually increased to 100 nM over the following weeks. Out-growing colonies were picked and screened for antibody production.
CHO DG44 cells were centrifuged in an Eppendorf microcentrifuge, briefly, at full speed, washed once with PBS, and pelleted once again. Genomic DNA was prepared by ethanol precipitation after SDS lysis and Proteinase K treatment of the cell pellets.
A mixture containing one of the primer pairs described supra, dNTPs, buffer, and Pfu polymerase was used to amplify either the heavy or light chain variable region using genomic DNA as a template using methods well known in the art. The resulting PCR products were digested with the appropriate restriction enzyme and analysed by agarose gel electrophoresis to confirm their identity.
The primer pairs for the light chain were:
For the light chain and SEQ ID NO: 37 plus
for the heavy chain.
The undigested heavy chain PCR product had a predicted size of 1200 base pairs, while the light chain PCR product had a predicted size of 800 base pairs. Identity was verified by restriction enzyme digest with BamHI.
CHO cell lines were transfected with the corresponding plasmids. Geneticin resistant cells were obtained and these cells were further selected for resistance to methotrexate. Single colonies were picked after amplification and transferred into 24-well plates. Culture supernatant was tested for chimeric IgG 3-4 days later by standard Dot Blot assays.
Any positive colonies were sub-cloned and cultured to achieve sufficient antibody production. The chimeric 19/2 antibody was purified from the supernatant on protein G columns and tested for its specific binding with recombinant GM-CSF by Western Blot (
Finally, the identity of producer cell lines were confirmed using PCR amplification of both their heavy and light chain variable regions. The DNA sequence of the heavy chain variable region PCR products for chimeric 19/2 mAb transfected cells was confirmed.
In order to optimize cell growth and antibody production, the CHODG44/pREN c19/2 cell line was first cultured in commercially available IMDM containing 10% FCS, at 37° C., in a 10% CO2 atmosphere. The cells were then weaned into serum free medium, and cultured in a custom made medium, i.e., IMDM SFII, with the following additives, at 37° C., in a 10% CO2 atmosphere.
Culture supernatants were harvested asceptically, and then clarified by centrifugation. The antibodies were then purified by affinity chromatography on a 5 ml protein. A sepharose fast flow column that had been pre-equilibriated in 50 mM Tris-HCL, pH8, was used. The column was washed, 20 times, with this buffer, and any bound antibody was eluted using 50 mM sodium citrate, pH 3.0, and the eluate was then neutralized, immediately, using 1M Tris-HCl, pH8. Antibodies were concentrated with a centrifugal filter, and dialyzed overnight at 4° C. in PBS. The yield was about 4-5 mg/liter. The purity of the antibodies was examined via SDS-PAGE, under both reducing and non-reducing conditions, using a 4-20% gradient on the SDS-PAGE.
Purified antibodies migrated as a single band under non-reducing conditions, and separated into the heavy and light chains, as expected, under reducing conditions.
The antibodies were also analyzed via size exclusion chromatography, (0.5 mg/ml), on a precalibrated HPLC column. Running buffer (5% n-propanol/PBS (0.5 M phosphate, 0/25 M NaCl, pH 7.4)) was used, at a flow rate of 0.2 ml/min at a temperature of 22° C., which is ambient column temperature.
The analysis demonstrated the integrity of the antibodies, which had calculated molecular weights of 179 kilodaltons.
The experiments described in this example were designed to determine the binding activity of the antibodies.
Biosensor analyses were carried out using a commercially available, BIAcore 2000, and a carboxymethyldetran coated sensor chip. The chip was derivatized with 1000, 300, or 100 RVs of recombinant human GM-CSF, on channels 1, 2, and 3 of the machine using standard amine coupling chemistry with channel 4 retained as the control blank channel.
Samples of the chimeric antibody were diluted in HBS buffer (10 mM HEPES, pH7.4, 150 mM NaCl, 3.4 mM di-NA-EDTA, 0.005% Tween-20), and aliquots were injected over the sensor chip at a flow rate of 1 μl/min. After injection, dissociation was monitored by allowing HBS buffer to flow over the chip surface for 5 minutes. Any bound antibody was then eluted, and the chip surface was regenerated, between samples, via injecting 40 μl of 100 mM HCl, pH 2.7, at a rate of 5 μl/min. In order to carry out kinetic analyses of the binding of the chimeric antibody, varying concentrations, ranging from 1-10 nM, were injected over the chip surface, and both apparent association (“Ka”) and dissociation (“Kd”) rate constants were calculated, using a Langmuir 1:1 binding model, with global and local fitting for calculation of Rmax, using B1 Aevaluation V3.1 software.
The results indicated that the chimeric antibody had slightly higher affinity for rhGM-CSF than the murine antibody. The calculated Ka for the chimeric antibody was 5.1×105M−1s−1 using 100 RU of GM-CSF. No dissociation was observed, regardless of analyte concentration, precluding Kd determination and indicating very high affinity.
Global fitting of Rmax, using the software referred to, gave an off rate of Kd=1.9×10−5s−1 and a high affinity for the chimeric antibody of 2.69×1010M−1.
These experiments were designed to determine both the binding activity of the antibodies, and if they cross-reacted with each other.
Nunc plates were coated with recombinant human GM-CSF (1 μg/ml), in carbonate buffer (pH 9.6, 0.05 M), 50 μl/well, and were incubated at 4° C., overnight, and were then blocked with 3% FCS/PBS at room temperature, for one hour.
Half-log, serially diluted triplicate 100 μl samples of either murine or chimeric antibody (10 μg/ml) were added to each well, to yield final concentrations of from 1.0 ng/ml to 10 μg/ml. Following incubation for 1 hour at room temperature, either goat antimouse IgG or antihuman IgG, labelled with horseradish peroxidase (10 ul/well Fc specific; 1:1000 dilution in 1% FCS/PBS) were used to detect bound antibody. After extensive washings, the bound antibodies were visualized by the addition of ABTS substrate (100 μl/well).
Optical density was read at 415 nm in a microplate reader.
The same protocol for binding antibody to the solid phase was used to determine if the antibodies competed with each other. As in the experiments, supra, half-log, serially diluted 100 μl samples, in triplicate, of 10 μg/ml of the murine or chimeric antibody were combined with 20 μg/ml of competing antibody, and then 100 ml of the mixture was added to the coated ELISA plates. Incubation was as above, and anti-murine or anti-human IgG labelled with horseradish peroxidase was used, also as described supra.
The results indicated that the antibodies did compete for binding for recombinant human GM-CSF. A shift in the binding curve was effected by addition of the excess, competing antibody. This indicated binding to, and competition for, a common epitope.
These experiments were designed to test the neutralizing activity of the anti-GM-CSF antibodies. Two human GM-CSF dependent cell lines, i.e., TF-1 and AML-193 were used. Growth curves were established, in the presence or absence of 0.5 ng/ml of recombinant human GM-CSF, and viable cell numbers were determined, via Trypan Blue exclusion, on day 0, 1, 2, 3, 5 and 7.
In a first bioassay, recombinant human GM-CSF, in amounts ranging from 0.0003 ng/ml up to 10 μg/ml, was mixed with anti-human GM-CSF antibodies, at a final concentration of 30 μg/ml, in 96 well, microtitre plates. Either TF-1 or AML-193 cells were added (103 cells/well), and plates were incubated at 37° C. for 7 days.
After this incubation period, the DNA proliferation marker MTS was added, at 20 μl/well. Dye incorporation was measured after 2 hours, by measuring light absorbance at A490 nm.
Increased MTS dye incorporation was observed as the amount of rhGM-CSF in the medium increased. Total growth inhibition of both cell types was observed with the chimeric antibody when rhGM-CSF concentration was 0.1 ng/ml or less, and there was marked inhibition of cell growth at 0.3-10 ng/ml rhGM-CSF.
In contrast, while the murine antibody had a similar effect on AML-193 cells, it was less effective on TF-1 cells. These results are seen in
In a second bioassay TF-1 and AML-193 cells were grown in the presence of 0.5 ng/mL rhGM-CSF and increasing amounts of murine or chimeric 19/2 mAbs (0.003-100 μg/mL) were added to the culture media and the neutralizing activity assessed after 7 days culture. Results are shown in
Additional experiments were carried out to produce a chimeric, HRS-3 antibody. The murine form of this antibody is described by Hombach, et al, Int. J. Cancer 55:830-836 (1993), incorporated by reference. The murine antibody binds to CD-30 molecules.
The protocols set forth for production of chimeric, anti GM-CSF antibody set forth supra were used. Since the antibodies were different, and sequences were known, however, different primers were used. These primers serve to introduce splice sites into the cDNA sequences encoding the murine heavy chain and light chain variable regions, and are set forth at SEQ ID NOS: 44, 45, 46 & 47, with SEQ ID NOS: 44 & 45 the nucleotide and amino acid sequences of the heavy chain, and 46 & 47 comparable sequences for the light chain
The primers were:
Following amplification, the murine heavy and light chain variable regions were cloned into the pREN Neo and pREN-DHFR sequences, which are set forth at SEQ ID NOS: 48 & 49, respectively. The cloning was possible because the amplification introduced PmeI and BamHI restriction sites into SEQ ID NO: 44, at nucleotides 1-7, and the final 6 nucleotides. Comparable sites are found at nucleotides 1340-1348, and 1357-1362 of SEQ ID NO: 48. Similarly, PmeI and BamHI restriction sites were introduced at nucleotides 1-8, and the last 6 nucleotides of SEQ ID NO: 47, such that this nucleotide sequence could be cloned into SEQ ID NO: 49, at positions 1337-1344, and 1349-1354.
The chimeric HRS-3 antibody was designed to have murine HRS-3 VL and VH regions linked to human kappa and gamma-1 constant regions, respectively. PCR primers were used to modify the 5′- and 3′-sequences flanking the cDNA sequences coding for the murine HRS-3 VL and VH regions. Modification included the insertion of a NcoI site at the 5′ primer end and a splice donor site followed by a BamHI restriction site at the 3′-end of both the light and heavy chain cDNAs for the variable regions to be spliced to the constant regions. These adapted mouse HRS-3 variable regions were then subcloned through the NcoI/BamHI restriction sites into a prokaryotic vector harboring a 5′PmeI site followed by a 5′ Kozak sequence and by a human antibody leader sequence. Sequences were cut from the prokaryotic vector by PmeI/BamHI digest and subcloned into mammalian cell expression vectors already containing the human kappa (pREN-Neo vector) or gamma-1 (pREN-DHFR vector) constant regions, described supra.
Once the constructs were established, they were transfected into DG044 cells, as described supra.
Positive colonies were sub-cloned, cultured to achieve sufficient antibody production, after which the antibodies were purified, on protein G columns via the Fc fragment.
The purified antibodies were analyzed via SDS-PAGE, following Laemmli, Nature 227:680-5 (1970), as modified by Renner, et al, Eur. J. Immunol. 25:2027-35 (1995), incorporated by reference. Samples from different stages of purification were diluted, in either reducing or non-reducing buffer, and were separated on 10-12% polyacrylamide gel via electrophoreses followed by standard Coomassic staining.
The results were in accordance with production of a complete, chimeric antibody, as evidenced by the banding patterns found in both reducing and non-reducing solutions.
The binding capacity of the chimeric HRS-3 antibody was determined via flow cytometry, in accordance with Renner, et al, supra. In brief, 1×106 cells of a target tumor line which expressed CD-30 were washed, twice, in PBS, and then incubated with varying concentration of antibody, at 4° C., for 30 minutes. The cells were then washed, and incubated with a secondary antibody, which was directed to the light chain, conjugated to either FITC or PE.
The results indicated that there was weak binding from cell culture supernatant purified from transfected CHO cells, and string binding with purified antibody. No binding was found when CD-30 negative tumor cells were used.
The antibody dependent cellular toxicity (ADCC), and the complement dependent toxicity of the chimeric HRS-3 antibody were determined using a europium released assay, as described by Hombach, et al, supra, and Renner, et al, supra.
In brief, for the ADCC assay, peripheral blood lymphocytes were isolated from tow healthy donors, and used at an effector:target ratio of 10:1, with 10,000 europium labelled, CD-30 antigen positive L540CY tumor cells. Antibody was added at varying concentrations (10 μl, 0.1 and 0.01 μg/ml), as was a control of 0 μg/ml. The effect was compared to the murine antibody, a bispecific murine anti-CD16/CD30 antibody, and an irrelevant, chimeric IgG1 antibody. A CD30 negative line was also used. Maximum lysis was measured after 0.025% Triton was added, and all assays were carried out in triplicate.
The results indicated that the chimeric antibody performed better in the ADCC than the murine antibody.
In the CDC assays, 10,000 europium labelled cells (100 μg) (L540Y), were incubated, with 50, 5, 0.5, or 0.05 μg/ml antibody in a 500 volume. Freshly isolated complement (50 μl) was added, and the mixture was incubated for 2 hours, at 37° C. The murine antibody was also tested, as was an anti CD-16 antibody and a chimeric anti IgG antibody, which served as controls, as did a CD-30 negative cell.
As in the ADCC assay the chimeric antibody was superior in terms of percent lysis to all other antibodies tested.
This example details the production of a fusion protein of a chimeric, G250 specific antibody, and tumor necrosis factor (“TNF” hereafter).
G250 is an antigen also now as “carbonic anhydrase 9,” or “CA9,” or “MN.” The G250 antigen and the corresponding antibody was described as being associated with renal cancer carcinoma by Oosterwijk, et al, PCT/US88/01511. The G250 antibody has also been the subject of several clinical trials (Oosterwijk, et al., Int. J. Cancer 1986: Oct. 15, 38(4):489-494; Divgi, et al., Clin. Cancer Res. 1998: Nov 4(11):2729-739.
Zavada, et al, have issued a series of patents in which the G250 antigen is referred to as “MN” or “MN/CAIX.” See, e.g., U.S. Pat. Nos. 6,051,226; 6,027,887; 5,995,075, and 5,981,711, all of which are incorporated by reference. These parents provide details on the antigen, and describe various tumors in which it is found, including cervical cancer, bladder cancer, mammary carcinoma, uterine, cervical, ovarian, and endometrial cancer.
Recently, Ivanov, et al, Am. Journal of Pathology 158(3):905-919 (2001), conducted investigations of CA9 and CA12 on tumor cells, and cell lines.
cDNA sequences for the light and heavy variable regions of a murine G250 specific antibody are known, and these include the endogenous antibody leader sequence.
PCR primers were used to modify both the 5′ and 3′ regions, in order to introduce restriction sites necessary for the introduction of the coding sequences to the vectors employed, which were SEQ ID NOS: 48 & 49, supra. The cDNA sequence which encodes the murine G250 heavy chain variable region is set forth at SEQ ID NO: 50, with the amino acid sequence at SEQ ID NO: 51 and the light chain variable region, at SEQ ID NO: 52, with amino acid sequence at SEQ ID NO: 53. The first 8 nucleotides in each of SEQ ID NOS 50 & 52 represent a PmeI restriction site. The first 19 amino acids encoded by the nucleotide sequence represent the leader region, and the first 24 the leader sequence for the light chain. The last 6 nucleotides in each of SEQ ID NOS: 50 & 52 are a BamHI restriction site. The same protocol as was used for the HRS-3 chimera was used to splice these variable regions into SEQ ID NOS: 46 & 47.
To secure the cDNA encoding human TNF, a human leukocyte cDNA library was used. The peripheral blood lymphocytes were stimulated with PMA, and the cDNA for TNF was amplified, using standard methods. Restriction sites were introduced in the cDNA sequence, so that the cDNA for TNF was positioned right after the hinge region of the G250 heavy chain. A (Gly) Ser coding sequence linked the two. SEQ ID NOS: 54 & 55 set forth the nucleotide and amino acid sequences of a TNF fragment, and SEQ ID NO: 56, a construct wherein the human gamma-1 heavy chain is followed by the TNF coding sequence, right after the IgG1 hinge region.
Within SEQ ID NO: 56, nucleotides 1419-1754 encode a partial, human IgG1 constant region, containing the CH1 and hinge domain, preceded by a 60 base pair intron region and splice acceptor site. The linker, i.e., (Gly)4Ser is encoded by nucleotides 1755-1769. The coding sequence for the human TNF fragment is set forth at nucleotides 1776-2296.
The resulting constructs were transfected into host cells, as described supra, and expressed. Note that SEQ ID NO: 56 contains a variant of the heavy chain vector noted supra, as it contains the human CH1 and hinge regions, followed by the TNF encoding sequence.
Cells were transfected and cultured as described supra for the HRS-3 chimera, and amplification was carried out using the primers of SEQ ID NOS: 40-43, described supra. The predicted size of the amplification product was 1100 base pairs, and this was in fact confirmed.
Positive colonies were then sub-cloned and cultured, as described supra. The chimeric G250-TNF fusion proteins were purified using anion exchanged chromatography on DEAE columns, using 5 ml samples, and increased salt concentrations in the elution buffer (NaCl, 0→0.5 M) (pH 8). The purity of the fusion proteins was determined, on SDS-PAGE, under reducing conditions. Two bands, of 45 and 28 kDa, respectively, appeared, consistent with the production of a chimeric fusion protein.
The purity of the chimeric fusion protein was confirmed in a sandwich ELISA. In brief, plates were coated with 1:6000 dilutions of affinity purified, goat anti-human IgG serum, and incubated overnight. They were then blocked with 2% gelatin. Either cell culture supernatant, or purified antibody was added, at varying concentrations, and then contacted with biotinylated goat anti-human TNFα specific serum, at 0.1 μg/ml, followed by visualization with a standard streptavidin peroxidase reagent.
The ELISA confirmed the purity of the antibody.
FACS was carried out, as described supra for the chimeric HRS-3 antibodies, this time using the fusion protein, and G250 positive tumor cells. Two different purification runs were tested, with chimeric G250 antibody as a positive control, and an irrelevant chimeric IgG1 antibody as a negative control.
The results indicated that the chimeric fusion protein bound as well as the chimeric antibody did. No binding was detected when G250 negative cells were used.
These experiments were designed to determine if the fusion proteins retained the ability of TNF to mediate cell death.
This was accomplished using an MTT assay as described by Renner, et al, Eur. J. Immunol. 25:2027-2035 (1995), incorporated by reference, and TNF sensitive (“WEHI-R”) cells. The WEHI cells were seeded at a density of 10,000 cells/well. Then, after 18 hours, sterile samples of the fusion protein, recombinant TNF, chimeric G250 antibody, or a negative control (plain medium), were added, at concentrations of 1.0×105, 1.0×102, 1, 1.0×10−2, 1.0×10−4, and 1.0×10−5 ng/ml, and the culture was incubated for additional period of from 48-72 hours. Any viable cells were detected, via standard methods, including Annexin V staining, and flow cytometry. To do this, 1×106 WEHI cells were incubated, overnight, with varying antibody concentrations, and dye positive cells were counted. The effect of antibody loaded tumor cells in WEHI killing was determined by pre-staining with commercially available PKH-26GL dye.
The chimeric fusion proteins were found to be as effective as recombinant TNF in killing cells.
It is known that TNF stimulates H2O2 release by human leukocytes. The chimeric fusion proteins were tested for this property.
Granulocytes were isolated from blood samples via standard methods, and were resuspended in reaction buffer (KRPG=145 mM NaCl, 5 mM Na2HPO4, 4.8 mM KCl, 0.5 mM CaCl2, 1.2 mM MgSO4, 0.2 mM glucose, pH 7.35). This mix was added plates that had been precoated with fibronectin (1 μg/ml, 2 hours, 37° C.) to permit granulocyte adherence. Following this, 100 μl of a dye solution (10 ml KRPG+50 μl A6550+10 μl horseradish-peroxidase) were added and incubated for 15 minutes at 37° C. Granulocytes were added, at 30,000 cells per well, and then either buffer (KRPG), PMA (5 ng/ml), the chimeric fusion protein (1 μg/ml) plus recombinant human IFN-γ (100 μ/ml), or the fusion protein plus the recombinant IFN-γ (at the indicated concentrations), were added. H2O2 release was measured for 3 hours, using standard methods.
The PMA served as a positive control. The chimeric fusion protein induced H2O2 release significantly higher than antibody alone, and the H2O2 release increases even more when IFN-γ was added.
This application is a divisional of application Ser. No. 10/365,123, filed Feb. 12, 2003, now U.S. No. 7,381,801 which claims priority of application Ser. No. 60/355,838, filed Feb. 13, 2002, and incorporated by reference in their entirety.
| Number | Name | Date | Kind |
|---|---|---|---|
| 7381801 | Renner et al. | Jun 2008 | B2 |
| Number | Date | Country | |
|---|---|---|---|
| 20100167395 A1 | Jul 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60355838 | Feb 2002 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10365123 | Feb 2003 | US |
| Child | 12012028 | US |
