WO1984003106A1

WO1984003106A1 - Production and characterization of hybridoma antibodies directed specifically against common determinant(s) present among closely related, but distinct proteins

Info

Publication number: WO1984003106A1
Application number: PCT/US1984/000161
Authority: WO
Inventors: Kurt Frimann Berg
Original assignee: WADLEY INSTITUTES OF MOLECULAR MEDICINE; WADLEY INST OF MOLECULAR MEDIC
Current assignee: WADLEY INSTITUTES OF MOLECULAR MEDICINE; WADLEY INST OF MOLECULAR MEDIC
Priority date: 1983-02-04
Filing date: 1984-02-03
Publication date: 1984-08-16
Anticipated expiration: 1986-08-03
Also published as: CA1306961C; EP0139676B1; EP0139676A1; JPS60500864A; US4902618A; EP0139676A4

Abstract

Production, detection of, and characterization of monoclonal antibodies/hybridoma antibodies (and the corresponding hybridoma cells including derived hybrid cells) which are able to recognize proteins (or part(s) thereof) belonging antigenically to the Human Leukocyte Interferon System (HuIFN-alpha) - be the said proteins produced natively (for example by white blood cells) or produced by recombinant techniques (genetic engineering) or produced by peptide derivation of the said proteins or part(s) thereof) - exhibiting the same common antigenic determinant(s) as found to be common in the native HuIFN-alpha system. Such antibodies (or the relevant part(s) thereof) can be used as "universal" antibodies for purification and characterization of all the above mentioned proteins belonging to the HuIFN-alpha system.

Description

PRODUCTION AND CHARACTERIZATION OF HYBRIDOMA

ANTIBODIES DIRECTED SPECIFICALLY AGAINST

COMMON DETERMINANT(S) PRESENT AMONG

CLOSELY RELATED, BUT DISTINCT PROTEINS

TECHNICAL FIELD

The invention is related to description, production, and characterization of new specific hybridoma/monoclonal antibodies recognizing common antigenic determinant( s) among the closely related proteins belonging to the HuIFN-a system - be they natively produced (in vivo or in vitro) or produced by recombinant DNA-techniques (genetic engeneering) or produced by peptide syntheses.

BACKGROUND ART

In erferons are now widely recognized as being an important group of biological proteins rendering cells or whole organisms resistant to virus attack (the so-called antiviral activity, cf. Stewart

(1980)). Furthermore, interferons have proved to play a significant role in the regulation of the immune system in vivo and in vitro together with numerous other so-called non-viral activities, such as re-transformation of transformed cells, inhibition of growth of tumor cells, etc., cf. for example Vilceck et al. (1980). For these reasons the interferon system has during the last decade been intensively investigated and numerous findings have been reported (cf. Stewart (1980), Berg (1982)).

One of the main difficulties observed during the beginning of the interferon research was the difficulties involved in the purification and characterization of the interferon proteins. it was believed that only one protein species existed in the human system, but later, it was shown by several investigators, e.g. Berg et al. (1975), Stewart (1980), Berg (1982), that there indeed exist three main types of interferon in the human system: Human Leukocyte Interferon (HuIFN-α), Human Fibroblast Interferon (HuIFN-β ) , and Human Immune Interferon (HulFN-γ) .

HulFN-α has now been completely purified. Originally it was thought to consist of one single entity only, but later on, as the purification progressed, it turned out that HuIFN-α consists of at least 6-7 different interferon protein species, and finally it was shown that at least 13 species exist

OMH ET *_*_. ^{W p}o (Berg et al. (1980), (1982a, b) . These species were purified simultaneously by combining traditional techniques (such as gelfiltration and the like) with antibody affinity chromatography using antibodies from rabbits. The pure specie was next, in turn, used as immunogen whereby "monospecific" rabbit antibodies against the interferon proteins were produced by using only a single interferon specie (e.g.: 20.000 specie) as immunogen. Thus, the monospecific, polyclonal rabbit antibodies were able to purify all species of HulFN-α simultanously when used in an antibody affinity column. This approach was followed by others employing immunoglobulins produced from hybridomas, (e.g.: Secher et al. (1980)). However, there was a serious drawback to this method: only part of the interferon proteins was bound to the hybridoma column (cf. Berg et al (1982, b) . Usually, half of the interferon activity was found in the wash, and the other half was bound to the column (and could be eluted specifically).

Therefore, the impression was, generally speaking, that HuIFN-α might not contain a common antigenic determinant, at least not as measured by means of the mouse hybridoma system.

SUMMARY OF THE INVENTION

The present invention addresses itself to this problem. In the following it will be shown that it, now, indeed is possible to produce hybridoma-produced 5 antibodies ( immunoglobulins) which are capable of binding all species of HuIFN-α (up to 12 different species) - or part thereof. The invention is new and surprising compared to the previous knowledge. The implications of the present inventions are imanent:

-_j^O all proteins belonging to HuIFN-α, all the bacterial derived HuIFN-_α species-obtained by genetic engineering techniques) will bind this immunoglobulin (which is characterized in that it binds all species of HuIFN-α as shown in an antibody affinity

15 chromatography) , whereby a universal approach for purifying all species of HuIFN-α - be it native ones or bacterially (plasmid) derived - have been invented for the first time. It is also within the scope of this invention to use other hybridoma systems (apart

20 from the murine system) yielding hybridoma antibodies (or the essential parts thereof) of the specificity as defined above. For example, it is forseen that also the rat system will be used in the future.

The said immunoglobulins might also be produced

25 by the now well known genetic engineering techniques, for example by means of the hybridoma cells producing the said immunoglobulins by first isolating the proper m-RNA (the specificity of which can be studied/measured/isolated by very well known standard

30 technique as for example illustrated in Methods in Enzymology, vol. 68 (1981), Recombinant Techniques. The purified m-RNA, which can be tested for the correct specificity in oocytes (by means of a semi- solid binding assay of the translated product - vide Materials and Methods in the present application) can then be transformed to c-DNA by means of reverse RNA- transscriptase and after making a plasmid library by incorporating the said c-DNA into relevant plasmids, the correct cloned can be found by screening taking into account that a probe can be made out from the correct m-RNA (identified/isolated/described as shown above), cf. Weissmann (1982). It is also within the scope of the said invention to produce- only parts of. immunoglobulins showing the desired specificity.

The said immunoglobulins will be very useful in different immunological assays employing, for example ELISA-(enzyme-immuno) or RIA-(radio-immuno) techniques or the like, in which the presence of a monoclonal antibody molecule (or parts thereof), with the above specificity is essential. Thus, the said immunoglobulins may be very useful in immuno assays for detecting interferon proteins (or parts thereof) by means of conventional techniques (see example Methods in Enzymology, vol. 70, part A, 1980).

It will also be possible to detect (quantity/quality) the presence of interferon receptors (which, in advance, have received interferon proteins, or parts thereof) or the like since these entities also can be recognized by the above antibodies.

BEST MODES OF CARRYING OUT THE INVENTION

It is contemplated that one of the reasons for being successful in obtaining the said hybridoma antibodies showing the remarkable binding properties can be found in the method for preparing the antigen- preparations consisting of highly purified interferon proteins stabilized with SDS (cf. Berg et al. (1980), (1982a)). The presence of SDS is thought to be an important factor in the presentation of the proteins to the immunocompetent cells in that the proteins in question (e.g.: interferon species) are exposing all their antigenic determinants in the presence of SDS (the proteins can also be pretreated with SDS before being confronted with the immunocompetent cells) , due to the denaturating power of SDS, whereby otherwise hidden determinants are fully exposed. It is also within the scope of this invention to treat proteins, in general, with SDS before (or simultaneously) the immunization is carried out the purpose of this procedure being to obtain more hidden antigenic determinants which otherwise are not recognized. By employing this "SDS-approach" during the course of immunization it is to be expected that the hybridoma antibodies against the most frequent determinants (i.e.: that or those found to be most common among all the proteins belonging to a protein mixture composed by related, but nevertheless, distinct proteins) will be found more often when using such a denaturing^' approach during the course of preparing the relevant immunogens (comprizing proteins, e.g.: pure species of HulFN-α). This general principle applies also to other hybridoma systems apart from the murine system.

-TORE

OMPI The above findings correlate very well with the well known homology which has been shown to exist on the gene level, for the genes coding for HulFN-α , cf. eissmann et al . (1982) . Since several regions are known to be constant (cf. Weissmann et al . (1982)) it is contemplated that this hybridoma, which is defined by the mere ability to recognize all species of HulFN-α , probably reconizes a specific determinant located in such a region, which is common for all the interferon species belonging to the HuIFN-α species. It is also contemplated that the said antibody also might recognize MuIFN-α having the same determinant as involved in the said specificity of the said antibodies. It is not known, at present, how widely distributed this determinant might be in the interferon-system (work is in progress) . For immunization, isolation of a spleen lymphocyte mixture fusion and hybridization, growing of hybrids and selection, subcloning of hybrides producing the desired antibody and isolations of these hybrids or the corresponding antibodies, the reader is referred to numerous previously published descriptions, as f.x., Kohler et al . (1975) , (1976) , Lovborg (1982) , Methods in Enzymology, Vol. 70 (1980) , Secher et al (1980) , Cellular Immunology,

Vols. 1-3 (1978), Kennett, McKern & Bechtol (1980).

Most of the plasmacytoma cell lines used to make hybridomas and some hybridomas producing antibodies are available from public cell distributing centers, such as for example The Salk Institute, Cell

Distribution Center, P. O. Box 1809, San Diego, Ca. 92112, USA, or other similar places. The present cell line, usd in this invention, was obtained from

O PI The Human Genetic Mutant Cell Repository in New Jersey, wide the section called: cells. The cells are generally grown in Dulbecco's Modified Eagle's Medium (DMEM) with high glucose (4, 5 g pr liter) or 5 in RPMI 1640. Addition of glucose to RPMI 1640 up to 4, 5 g pr liter does improve growth at higher densities. Cells are maintained in stationary suspension culture at a concentration of 10 -10 cells pr. ml. Optimum condition of a culture to be

-_]_0 used for fusion is high viability (i.e.: greater than 90%) with 3-8 x lo cells pr. ml. For more experimental details the reader is referred to f.x. Lovborg (1982) and Kennett et al (1980).

Test for Mycolplasma was always performed since

-_jc it was shown by others that the presence of mycoplasma very easily can hinder a successful hybridization. All myeloma cells were always checked 1-3 days before used. Only if the test turned out to be absolutely negative were the cells accepted for

20 fusion.

The institution of the mycoplasma test into the hybridization experiments greatly improved the numbers of positive hybridomas obtained in the fusion experiments, thus confirming similar observations

₂₅ performed by several groups (cf. Lovborg (1982)).

EXPERIMENTAL SECTION Interferon titrations Were performed as previously described using for

30 example VERO-cells and VSV as chellenge virus (see f.x. Berg et al. (1980)). All units are expressed into international units (69/19 B, MRC, UK). If very low amounts of interferon were to be detected the so- called bovine system, comprising BMDK-cells and VSV, - was used since it turned out to be extremely reliable and very sensitive. As low as 0.1 units of interferon activity could easily be detected. Therefore, this system was widely used in the semi- neutralization of hybridoma-supernatants.

Preparation of the immunogen Crude human lekocyte interferon was purified by gelfiltration followed by an antibody affinity chromatography yielding almost pure interferon g proteins with a specific activity of 10 units pr. mg. protein. Recovery was about 75%. The total eluate from a typical experiment contained 5 x 10 units which was divided into 100.000 units containing aϋquotes (in 0.1 ml eluting buffer containing 0,1% SDS) and kept at - 20°C until used (stable for at least 5 months) - cf. Berg et al. (1980). Sometimes, less purified preparations of leukocyte interferon was also used (inc. SDS).

Immunization

The Balb/c mice-preferentially female mice - were immunized as follows: The first injection (100.000 units) was mixed with Freund's Adjuvans (F.A.) and given intra peritoneally (I.P.). The following 5-8 injections were given as weekly injections but consisted only of 30.000 units and no F.A. was included. After 5-8 injections the mice developed antibodies to HuIFN-α (detected by the "traditional" neutralization test (wide neutralization tests) in the range of 200 - 2000 neutralizing units/ml - cf. Berg (1982). This

OMPI estimate very often yielded considerably lower figures compared to the semi-solid netralization method (wide below) which turned out to be far more sensitive. Tow - five days before the fusion the mice received 100.000 units in total as a booster; half as I.P. and the other half as subcutaneous (S.^'C.) injections without F.A.

NEUTRALIZATION OF ANTIBODIES TO HuIFN-α The simple ("traditional") method for neutralization of interferon consists of mixing equal volume of interferon proteins containing a predetermined amount of interferon units with the antibody sample in question using several dilution step, see f.x. Berg (1982) . This method, however, will only recognize those antibodies reacting with the antigenic determinants close to the biological determinants on the interferon molecule(s) . Thus, as pointed out by Berg (1982) , it is quite possible that a- great deal of antibodies will escape identification by using the "traditional" neutralization method only.

To overcome this problem several ELISA-tests were employed taking into account that f.x. sheep antibodies to HulFN-α (produced out from partially purified HuIFN-α by ethanolic precipitation, cf. Stewart (1980)) bind all species of HuIFN-α. Thus, immunoplates (NUNC-DENMARK) were firstly coated with igG-solutions from a sheep antiinterferon serum followed by pure interferon proteins and after a suitable incubation the samples from the hybridomas were added and finally rabbit antimouse IgG- galatosidase conjugates were added, as detailed in details in f.x. Methods in Enzymology, vol. 70, part A, 1980) . This method turned out, however, to yield too many false positive results in the sense that ELISA-positive supernatants did not turn up as positives in the semi-solid neutralization test (wide below) .

SEMI-SOLID NEUTRALIZATION TEST

This method was developed in order to test ELISA-positive samples. The test takes advantage of the very simple fact that a "possible" antibody molecule directed against HulFN-α will bind to the interferon molecule. Therefore, if one first immobilizes the "possible" antibody (which can be done by another antibody directed against the said "possible" antibody) before adding an exact, but small amount of interferon, preferably as low as 0,1 units in total (wide the section of titration of interferon) a very small drop in the interferon titer (content) could easily be detected using the "bovine" system. This system proved superior to the ELISA- system. The following method was used:

1. Coat with rabbitanti mouse immunoglobulin (DAKO-PATT, Denmark) using 300 pg/ml in PBS, inc. 0,05% sodium azide (S.A.) at 37°C for 1 hour.

Transfer to 4°C for at least 24 hours (or several weeks) .

2. Wash with PBS incl. 0,05% S.A. incl Tween 80(0,05%) - the so-called: washing buffer - three times.

3. Saturate unbound sites of the immunoplates with 2% egg albumin in PBS (SIGMA) at room temperature for 2 hours.

4. Wash 3 times with washing buffer.

5. Add 100 μ 1 supernatent from -the hybrodoma culture to be tested to the well. Incubate for 24 hours at room temperature (incl. 0,05% S.A.). 6. Wash 2 times with washing buffer.

7. Wash 2 times with washing buffer - (without any^' S.A.) .

8. Add 100 μ of interferon solution (f.x. as native interferons) containing 0,1 unit in total (wide the section concerning interfetitration) for 2 hours at room temperature.

9. Transfer the mixture (100 μl) to MDBK- cells, in microwells, which in advance have been grown to confluency using the same micro-trays as used for the interferon titrations (NUNC-Denmark) and incubate for 20 hours at 37°C (5% C0₂ ⁾ .

10. Add VSV at a predetermined concentration yielding a distinct CPE after 24 hours in control cultured cells not receiving any interferon. 11. Read the microtrays as usual (wide the section: interferon titration and references cited herein) , taking into account that if the cells are destructed in one well the supernatent from the original hybridoma clone (culture) did contain antibodies which were able to bind (remove) the . HuIFN-α: thus, this result signifies a positive clone.

' CLONING AND FURTHER GROWTH OF POSITIVE HYBRIDOMAS positive clones, identified as described above, were divided into several subclones by means of the "limited dilution technique" using conditioned medium and feeder layers (peritoneal macrophages) from Balb/c mice) such as described in details by Lovborg (1982). The subcloned hybridoma were all checked regularly for binding activity and if positive persistently the cells were allowed to grow further before injected (I.P.) into balb^'c mice which had received an immunostimulator in advance (f.x. Pristane, 4-6 days in advance). After 5-15 days (depending on the amount of injected cloned celles) new hybridoma cells of the correct specificity can be harvested together with large volumes of ascites fluids containing the desired antibody in an extreme high concentration (about more than 95% of all the immuniglobulins present are of the desired specificity) . The hybridoma cells are separated from the ascites fluid subsequent to a centrifugation and can now, in turn, be used for another injection in new mice (Balb/c mice) whereby more ascites fluids can be obtained together with more hybridomas, and so forth, taking into account, that the hybridoma cells harvested from one mice can be used for production of 20 new mice each yielding approximately the same number of hybridoma cells as the first. The hybridoma cells may also be grown in vitro in roller flasks using the relevant conditions such as already has been described in detail by Lovborg (1982) and in principle the same cells and antibodies will be obtained as described above using the mouse-system ( in vivo) .

ISOLATION OF ANTI IFN- IMMUNOGLOBULINS

Since it is well known that the antibodies produced in ascites fluids are highly succeptible to proteolytic degradation from the proteolytic enzymes

O PI present in the ascites fluid the harvested ascites fluids were first cooled to +4°C and centrifuged (to remove hybridoma cells and other debris) before treated with ammonium sulphate whereby the immunoglobulins were presipitated. This isolation was performed immediately- after harvesting the ascites fluids. The ammonium sulphate precip. immunoglobulins were kept at -20°C. Before titration a small aliquote was dissolved in f.x. 5 ml of medium, dialyzed versus 1 liter of PBS, before being checked in the semi-solid neutralization-test. The immunoglobulins were further purified by traditional methods such as gelfiltration, ionexchange chromatography, and the like.

CELLS

The mouse myeloma cell line used ^*in the present invention is characterized as follows: Mouse Myeloma, line P3 x 63 NSl, HPRT-deficient, produces Kappa, non-immunoglobulin secreting. Medium contains 4.500 mg glucose pr. liter.

The above line was obtained from the Human Genetic Mutant Cell Repository (under cataloque nr.: GM 3573), Copewood and Davies Street, Camden, New Jersey, 08103, USA. (NIH Publication no. 81-20011, Revised Oct. 1981).

MEDIUM

The myeloma cells were grown in the same media composition as described by numerous other investigators, cf. Lovborg (1982), using RPMI 1640, 10% calf serum (new born) , containing penicillin, streptomycin, gentamycin together with Napyruvate etc. as described by for example Lovborg (1982). STORAGE OF HYBRIDOMA CELLS

The ascites fluids, which include the hybridoma cells of the desired specificity, are harvested from several mice (wide section: cloning and growing of cells) by centrifugation and washed 1 time with RPMI

1640 and fuged. (the remainder of the ascites fluid is processed as described elsewhere, wide the section: isolation of immunoglobulins) . About 40 x g 10 cells are resuspended in 5 ml of fresh Eagle's medium (incl. 20% calf serum) to which is added an equal part of 19-20% DMSC in RPMI 1640 (Dimethyl Sulfoxide, Merck & Co . ) in a dropwise manner (10 drops pr. minn.) mixing the suspension thereby avoiding a local heat (which might be detrimental to the hybridoma cells) . After mixing the said suspension, comprising the hybridoma cells, are divided into 1 ml ampules which are sealed and kept at 4°C for 24 hours, after which, they are transferred to a N₂~ container whereby the ampules are cooled 1°C pr min (according to the instructions from the manufacturer producing the so-called "Linde- container" ) . The cells in the ampules are finally stored in liquid nitrogen, cf. Lovborg (1982) .

THAWING OF THE CELLS

One ampule is thawed by immersion into a 37°C warm water bath. After centrifugation the cells are resuspended in new medium and counted and adjusted to

4 , 5 contain 10 -10 cells pr el. The medium is changed after 24 hours or possibly earlier, in case the cells already have adhered. The further growing is detailed described by Lovborg (1982). Alternatively, the thawn cells (from the ampoule) is directly injected into a Balb /c mouse (I. .) whereby new cells and ascites fluid (containing the desired antibody) can be harvested in 3-7 days (depending on the amounts of cells injected) .

MYCOPLASMA TEST primary human amnion cells (obtained from the deprtment of gynaecology) are seeded on small cover slips (of glass) and after attachment the cover slips are ready for use (the cells should not reach each other as confluency tends to lower the sensitivity). The sample, consisting of the myeloma cells, is added to the surface of the cover slip (about 10-15 drops of the sample) by means of a

Pasteur pipette and the cover slip (containing the sample) is carefully wrapped into a plastic film to avoid evaporation. If the original sample, consisting of myeloma cells, were infected by mycoplasma, the same mycoplasma would also infect the amnion cells. Normally, the latter are free from mycoplasma infections when taken fresh. (cf. later -control cultures). On the following day the cultures are stained by means of "Hoechst Stain" for 2 hours, according to the staining procedure given by the manufacturer (Hoechst, W. Germany) : 1-2 grains of the stain is dissolved into 5 ml buffer at pH 5,5. Only a few drops should be used in the actual staining since the pH(S,5) is below physiological range (7,2). The stain is incubated for 2 hours with the cells, after which, the cells are fixed with methanol and dried. If the amnion cells have been infected with the mycoplasma (stemming from the original sample) small fluorescencing particles will clearly be seen in a fluorescencemicroscope as an indication for mycoplasma infection. If no fluorescing particles are found no infection has occurred. A control culture (i.e.: only amnion cells) is also included. (the latter should be negative) . When the myeloma cells were found to contain mycoplasmas they were discarded and fresh, mycoplasma free, myeloma cells were taken from the cell freezer (liquid nitrogen container).

EXAMPLE 1.

One female Balb/c mouse was immunized as described in the section. Immunization (wide also the section: Preparation of immunogens) using 100.OOo units of HulFN-^α (stabilized with SDS) incl. F.A. followed by the usual weekly shots comprising 30- 40.000 units (I.P.) and finally, three days before fusion, the mouse received a booster of 100.00 units as previously described. During the period of immunization the "traditional" neutralization titer (wide the section: Neutralization) were obtained:

TABLE 1 NEUTRALIZATION TITERS OBSERVED IN THE IMMUNIZED MOUSE

Week neutralization units/ml serum

1-3 0

3-5 50-80

7 1500 _After the fusion had been performed by means of the spleen cells from the immunized animal^'and the NSl cells (wide section: Cells) the presence of 62 living clones were established by microscopic inspections (in Costar Trays, NUNC, Denmark. Cf. Lovborg

(1982)). All supernatants (62) were checked by the semi-solid neutralization method (wide the said section) and only two clones were found to be positive (partially) . The two clones were further grown and subcloned and during this procedure (wide section: Cloning and further growth of positive hybridomas) one positive clone was lost due to a mistake during the cloning procedure, the other (LO- 22) was further propagated in Balb/c mice, as previously described by yielding 5,5 ml of ascites fluid together with a large amount of hybridoma cells (after four weeks subsequent to the injection (I.P.) of the primally, subcloned hybridoma cells estimated to 1000 in total. After isolation the suspended cells were further injected into 10 new Balb/c mice as previously described, and after 7 days 65 ml of ascites fluid was harvested together with a large number of hybridoma cells. Some of these were frozen, serving as a stock culture, other were transplanted into new mice. All the cell-free ascites fluids were precipitated and isolated as described in the section: Isolation of anti-HulFN-α immunoglobulins. (several grams were isolated at this stage) . Semi-solid neutralization assays were simultaneously performed during the above experiments. The first two positive clones, however, did only neutralize the native human leukocyte

OMPI interferon (from virus induced buffy coats) partially, although the supernatants from the primary hybridoma cultures could be diluted more than 100- fold without reaching an end-point. The same tendency, at a much higher level, was seen when the

LO-22 was grown as an ascites tumor: at a dilution of

7 more than lo there was still a partial neutralization/binding observed. (wide the semi- solid neutralization method). At the early stage of these experiments, the above mentioned binding results obtained in the semi-solid binding assay lead to the impression that the LO-22 could only recognize part of the species of HuIFN-α, which was in line with the results from other groups (cf. Secher et al . (1980), Allan et al . (1980) , Berg (1982)) and which was generally accepted as the scientific concensus.

In order to obtain more detailed and precise information about the binding capabilities of the LO-22 hybridoma antibody it was decided to carry out an antibody affinity chromatography taking into account that only half or even less of the applied leukocyte interferon would be expected to bind (cf. the partial neutralization observed on p. 17) .

The immunoglobulins from 3 ml of ascites fluid from the original mouse (which had received only about 1000 cells I.P.) was isolated (wide the section: Isolation of anti IFN- α immunoglobulins) and coupled to CNBr-Activated Sepharose (cf. Berg (1977), Berg et al. (1980)) as previously described. The equilibrated column was loaded with 90 ml of crude, native human leukocyte interferon and after a thorough wash the column was eluted by means of low pH. The following, surprising and unexpected results

O PJ were obtained, wide table 2: 96% of the interferon was removed from the original interferon preparation (input). Based on the results obtained in the semi- solid neutralization tests (wide table 1) the almost opposite result would have been anticipated (i.e.: only a portion was expected to be caught by the LO-22 column (f.x. 30%)).

TABLE 2 ANTIBODY AFFINITY CHROMAΪOGRAPHY WITH LO-22 COLUMN

volume IFN-units IFN-units recovery ml human syst. bovine syst. %

Input 90 4,lxl0⁶ 4,8xl0⁶ 100

Wash 100 2xl0⁵ 9,5xl0⁴ 4

Eluate 16 2,8xl0⁶ 6,5xl0⁶ 70

In order, further to analyze the binding properties of the LO-22 column, the wash (consisting of 2,0x10 human interferon units, cf . table 2) was concentrated to about 4 ml (by means of sucrose and dialysis tubing, cf. Berg et al (1980)) before being loaded back on the LO-22 column. The second wash was titrated in both human and bovine systems and the following results were obtained: less than 100 units (bovine) were detected, whereas, about 50.000 units (human) were detected. Furthermore, the second wash (consisting of about 50.000 human units) was able to neutralize anti HulFN-β completely by means of the traditional neutralization test (wide the section neutralization) . The purified eluate from table 2 (consisting of 2,8x10 units (human) was not able to neutralize the said anti-HuIFN-β in an analogous neutralization test, at all. Thus, the 50.000 units (human) seen in the second wash consists of another species - namely HuIFN-β . The above results are in complete agreement with for example Vilcek et al. (1977) and Berg (1982).

The eluate from LO-22 column (table 2) was analyzed on analytical thin slab SDS-PAGE (using gradient gels, 12%-24%, 25 cm long, cf. Berg, et al. (1980), (1982a), (1982b)) and an identical pattern of HulFN-α species were seen when comparing with the previously published reports. Thus all species of HulFN-_α were seen (no "bovine" spcie was present in the particular batch used in this experiment, cf. Berg et al (1982a), (1982b)). The SDS-PAGE was stained and the individual bands were cut out and titrated and eluted, as previously described by Berg et al. (1980), (1982a), (1982b), and the following

results were obtained ( table 3 )

TABLE 3 BIOLOGICAL PROFILE OF AN LO-22 ELUATE OBTAINED BY SDS-P5GE

interferon units interferon units

Slice No. (WISH cells) (BDVK cells) human units bovine units

1 0 0

2 15.000 140.000

3 20.000 45.000

4 5.000 15.000

5 5.000 15.000

6 15.000 45.000

7 3.000 15.000

8 1.300 7.000

9 500 4.000

10 15.000 13.000

11 300 5.000

12 300 13.000

13 7.000 15.000

All 12 species were found to be located in the usual-* molecular we ight range of 16 .000 -25.000 which is in complete agreement with previously described reports concerning the location and numbers of species present in a HulFN- α preparation ( Berg et al . ( 1982a) , ( 1982b) ) . Probably, the LO-22 column is also able to recognize not previously detected species of HuIFN-α wi th a higher molecular we ight since fractions containing interferon activity were also found around 30.000 ( fract ion 14 ; 15 ; and 16 yielding 600 human units/1000 bovine units; 1400/1800, and 60/180) . These fractions are presently being more closely investigated.

The above results (from the antibody affinity chromatography) are, somehow, in contradiction to the results obtained by the semi-solid neutralization method (table 1) indicating only a partial binding of the species of HuIFN-α to LO-22 immunoglobulin. At present, the reason for this contradiction is thought to be based on the quality of the rabbit antimouse immunoglobulin used in the coating of the immunoplates (wide the section: semi-solid neturalization method) .

The immunoglobulins produced from the LO-22 hybridoma cells were characterized as belonging to the IgG-class by means of Ouch erloni-tests in which several fractions of the ascites fluids (and supernatants from cultured LO-22 cells) were checked against rabbit antimouse immunoglobulins ( DAKO-PATT -Denmark; Behring Werke - . Germany) or goat antimouse IgG F(ab)2 (Cappel) or goat antilgM (heavy chain) (Cappel) . Several controls were included (the NSl-cells alone yielded, as expected, no positive reaction against for example IgG) . A heavy precipitate was seen especially between the hybridoma antibodies, LO-22, and goat antimouse IgG F(ab)₂ antiserum (Cappel) .

COMMENTS TO EXMPLE 1. The above experiments which are detailed described have proved the following:

1. That the LO-22 hybridoma antibodies/column recognizes (bind) all species of HuIFN-α having the specific characterizations belonging to the HulFN-α system. (cf. Berg (1982), Berg et al.-(1980), (1982a) , (1982b)) .

2. That Human fibroblast interferon, HuIFN-β , which is known to be present in small amounts in a crude human leokocyte interferon preparation (about 1-2% of the total interferon activity has been shown to belong the HuIFN-β specie - cf. Bert et al. (1975)) is not recognized by the LO-22 hybridoma antibodies/column.

INDUSTRIAL APPLICABILITY

The monoclonal antibodies/hybridoitia (and the corresponding hybridoma cells including derived hybrid cells) which are able to recognize protein (or part(s) thereof) belonging antigenically to the Human Leukocyte Interferon System (HuIFN-α) can be used as "universal" antibodies for purification and characterization of all the above mentioned proteins belonging to the HulFN-α system.

REFERENCES

Allen, G. & Fantes, K.H. (1980): A family of structural genes for human lymphoblastoid (leukocyte- type) interferon. Nature, 287, 158-166.

Berg, K. , Ogburn, C.A. , Paucker, K. , Mogensen, K.E. , Cantell, K. (1975): Affinity chromatography of human leukocyte and diploid cell interferons on sepharose- bound antibodies. J. Immunol., 114, 640-644.

Berg, K. , (1977) : Sequential antibody affinity chromatography of human leukocyte interferon. Scand. J. Immunol. _6_, 77-86.

Berg, K. & Heron, I. (1980): The complete purification of human leukocyte interferon. scan. J. Immunol., 11, 489-502.

Berg, K. (1982): Purification and Characterization of Murine and Human Interferons. A Review of the Literature of the 1970' s. Acta. Pathol, Microbiol., et Immunol. Scand., Section C, Suppl . no. 279. pp.l- 136.

Berg, K. & Heron, I. (1982a): Human leukocyte interferon comprises a continuum of 13 interferon species, in: Lymohokines, (eds. Khan & Hill), Academic press, N.Y., p.397-416.

Berg, K. , Secher, D. & Heron, I. (1982b): Purification and characterization of the HuIFN-α species, in: The Interferon System, A review to 1982

O PI - Part I. (Eds. Baron, Dianzani , Stanton) , Texas Reports on Biology and Medicine, Vol. 41, Tx. , p.225- 234.

Kennett, R.H., McKern, T.J., Bechtol, K.B. (1980): Monoclonal Antibodies, Hybridomas: A New Dimension in Biological Anaylses, incl. appendix. Phenum press, N.Y. pp. 1-421.

Kohler et al. (1975): Nature, 256, 495-497.

Kohler et al . (1976): Eur. J. Immunol., _6_, 511-519.

Lovborg, U. (1982): Production and Maintenance, Trowbridge, London pp.1-65.

Rubinstein, M. (1982): Pure species of HulFN-α. Phil. Trans. R. Soc. Lond. B. 299.

Secher, D. & Burke, D.C. (1980) : A monoclonal antibody for large-scale purification of human leukocyte interferon. Nature, 285 , 446-450.

Vilcek, J., Havell, E.A. , Yamazaki, S. (1977): Antigenic, phisicochemical and biological characterization of human inteferons. Ann. N.Y. Acad. Sciences, 284 , 703-710.

Vilcek, J. , Gresser, I., Merigan, T.C. (1980): Regulatory Functions of interferons. Annals of New York Academy of Sciences, volume 350, pp.1-641.

Claims

1. Mouse hybridoma/monoclonal antibodies, or part(s) thereof exhibiting the relevant binding properties, which will bind or recognize common antigenic determinant(s) present in the family (group) of distinct, but closely related proteins (or part(s) thereof) belonging to the human leukocyte interferon system - HuIFN-^α.

2. Rat hybridoma antibodies/monoclonal antibodies as claimed in Claim 1.

3. A method for preparing the immunogen- preparation for obtaining a suitable antigen- presentation (during the course of immunization) comprising treating the proteins (or part thereof) - belonging to the HuIFN-α system and possessing the common antigenic determinant(s) - with 0.1% - 5% SDS (Sodium Dodecyl Sulphate).

4. A method for producing hybridoma/hybrid cells yielding antibodies as claimed in Claim 1 and 2 by, a) immunizing a mouse with denatured species of HuIFN-α

(or parts thereof) as claimed in Claim 3; b) isolating the lymphocytes from the immunized mouse; <■■") performing a fusion between the B-lymphocytes and mouse-B- myeloma cells; d) transferring the mixture of the formed hybrid cells -and non-hybrid cells to a selection-medium and performing the selection between the hybrids and non-hybrids yielding hybrid-cells; e) performing a cloning and subcloning with respect to hybrids yielding a positive (partial) test in a semi-solid binding assay test system; and f) performing the cloning also by injecting small numbers of hybrid cells into mice (I.P.) thereby obtaining the same hybrid cells as above (e).

5. A method for producing rat hybridoma/hybrid cells as claimed in any preceding claims.

6. A method for producing the antibodies exhibiting the specificity as claimed in Claim 1 by genetic engineering techniques by means of the specific m-RNA which is isolated from the hybrid cells as claimed in any of the preceding claims. The specificity of the m-RNA is measured/isolated by known standard methods using the oocyte-system or the wheat-germe system. The m-RNA is transformed into c- DNA by means of reverse transscriptase and a plasmid library is constructed by known methods. The correct clone(s) can be detected by means of a probe made from the isolated/identified m-RNA and the positive clone(s) can then be inserted in relevant known expression-plasmids yielding the desired antibodies, or parts thereof, exhibiting the characteristic binding properties as claimed in Claim 1.

7. The use of the antibodies as claimed in Claim 1, 2, and 6 for purifying the proteins (or parts thereof) belonging to the HuIFN-^α system.

8. The use of the antibodies as claimed in Claims 1, 2, and 6 for identification/detection

(qualitatively/quantitatively) by means of known immunological/immunochemical methods involving the use of enzyme - or radioactive-tracers.

9. A method for identifying the common determinants) as claimed in Claim 1 by means of the said immunoglobulins using labelled antibodies.

10. The common determinant s) belonging to the HuIFN-α system as defined by the antibodies in Claim 1, 2, 6 and 9.