GB2123835A - Interferon E - Google Patents

Abstract

Disclosed is a new material having antiviral activity designated interferon epsilon. The material may be produced, for example, by exposing primary, diploid human eipthelial cells to a virus or certain nucleic acids and then incubating the cells under conditions in which the new interferon is produced and is secreted into the culture medium. The material produced by this process when assayed with cell cultures of human and bovine origin exhibits a unique profile of activity. The material also displays no significant cross reactivity with antisera prepared against either interferon alpha, interferon beta, or interferon gamma in neutralization assays.

Description

SPECIFICATION Interferon E Background of the invention This invention relates to a novel composition of matter (hereinafter called interferon epsilon or IFN-E) useful, for example, in human and other cell cultures as an anti-viral and/or antiprofileration agent, to processes for producing the material, and processes for treating human epithelial cells so as to resist viral infection.

Interferons are materials which have antiviral properties. They are produced by certain types of cells which have been stimulated by exposure to virus, certain nucleic acids, or antigen/mitogen complexes. Interferons are extremely potent drugs which show great promise as clinical antiviral and anti-tumor agents.

There are currently three known types of human interferon: interfon alpha (IFN-), produced from human leukocytes or lymphoblastoid cells; interferon beta (IFN-P), produced from fibroblasts; and interferon gamma (lFN-p), produced from human T-lymphocytes. All three are secreted by the respective cells after the cells are stimulated by a virus, certain nucleic acids or antigen/mitogen complexes. Recent evidence indicates that these interferons may comprise a mixture of structurally related proteins, as opposed to a single protein. Human interferons may be differentiated from each other by their differing levels of activity in human and bovine cell cultures. An assay for measuring the activity may be conducted as disclosed in The Interferon System, William E.Stewart, Springer Verlag Co., New York, 1978, which is a modification of the method of Ho and Enders reported in The Proceedings of the National Academy of Sciences, Volume 45, pages 385-389, 1959. IFN-a has a level of antiviral activity on bovine kidney cell cultures of 1 to 5 times its activity on human fibroblast cell cultures. IFN-p has a level of antiviral activity on bovine kidney cell cultures of 1/100 to 2/100 of its activity on human fibroblast cell cultures. IFN-y has an activity on bovine cell cultures of less than 1/100 of its activity on human fibroblast cultures.

In the most productive method or producing IFN-a from human leukocytes, approximately one unit of interferon is produced per 50 cells. IFN-y is produced from human T-lymphocytes at a yield of about 1 unit per 1000 cells. IFN-P produced from human fibroblasts has an optimum yield of about 1 unit per 1 5 cells.

An antiviral unit of interferon is that concentration which protects one-half the cells in a human fibroblast culture from challenge with Vesicular Stomatitis Virus at a standard concentration (one pfu/cell).

Antiviral activity has been detected in media from other cells of human and of animal origin.

Interference in the multiplication of influenza virus was first detected by Issacs and Lindenmann (1957) in cultures of chicken chorioallantoic membrane. Another example is the Hela cell line which is a transformed cancer cell of epithelial (cervical) origin as reported by G. Gey et al. in Cancer Research, Vol. 12, pp. 264-265, 1 952. Interferon activity has been detected in various transformed or neoplastic cell lines originally derived from human epithelial tissue. See Production of Intefferon by Human Tumor Cell Lines, Jameson et al, Archives of Virology 62, 209-219 (1979). Prior art human interferon preparations have been observed to have antiviral activity in human cell cultures.

Summary of the invention A new type of antiviral material designated interferon E has now been discovered and produced.

This new material which, like the other interferons, appears to be composed of at least two operationally related proteins, may be produced, for example, by subjecting normal human epithelial cells to an interferon induction technique and then incubating the cells under conditions to promote production of the new interferon. lFN-, in contrast to the alpha, beta, and gamma prior art substances, has a level of antiviral activity in bovine kidney cell cultures of 1/4 to about 1/2 of its activity in human cultures. IFN-E shows no significant cross reactivity with antisera prepared against IFN-a, lFN-p, or IFN-y in neutralization assays. It is stable at pH 2. Its production by epithelial cells is eliminated by the introduction of RNA synthesis inhibitors or protein synthesis inhibitors in the culture.The antiviral activity of preparations containing IFN-E is destroyed by proteases.

IFN-E can be produced from all types of primary human diploid epithelial cells. Cultures of human epidermal, conjunctival, vaginal, and esophageal cells have all been used with success. Nonlimiting examples of other epithelial tissue types which may be used include nasal, pharyngeal, bronchial, pleural, and intestinal. Advantageously, it has been discovered that primary epithelial cells may be induced to produce interferon- using the same techniques as used in the prior art for production of other types of interferon from leukocytes, lymphoblasts, and fibroblasts. In a preferred induction technique employing Newcastle Disease Virus (Bankowski strain) or Sendai Virus, 1 00-500 epithelial cells are required to produce a unit of IFN- .

As noted above, IFN-E is produced by primary, diploid epithelial cells of human origin. The phrase "primary, diploid human epithelial" cells, or cell cultures, as used herein, refers to cells sampled from healthy human tissue or a culture of such cells produced, for example, in accordance with the procedure disclosed in U.S. Patent No. 4,01 6,036 to Green et al., the disclosure of which is incorporated herein by reference. These types of cell cultures are to be distinguished from non-human cells, human cells of other than epithelial origin, and transformed or neoplastic epithelial cells.

The new interferon may also be produced in naturally or artificially transformed eukaryotic cells containing the portion of human epithelial nucleic acid responsible for the production of IFN-E and in cells modified by recombinant DNA techniques.

It has also been discovered that an interferon produced from a given type of human cell, e.g., epitheliai cells, has greater antiviral activity on that particular cell type than on other human cell types.

Thus, interferon epsilon has increased antiviral and anti-tumor potency in the treatment of epithelial tissue afflicted with a tumor or viral infection, as compared to its efficacy with respect to other human cell types.

Accordingly, it is an object of the invention to provide a new type of antiviral substance useful, for example, in protecting human cells from virus attack. Another object of the invention is to provide a method of producing an interferon of the type indigenous to normal, healthy, human epithelial tissue.

Yet another object is to provide a method of treatment of viral infection or tumor growth in a specific tissue type by exposure to an interferon derived from the same tissue type. Another object is to provide a type of interferon which is more active on human epithelial cells than on other types of human cells, e.g., human fibroblast cells. These and other objects and features of the invention will be apparent from the following description and from the appended claims.

Description Broadly, to produce IFN-E, normal human epithelial cells may be cultured in vitro (for example, in accordance with the method of Green et al.) in monolayer, within microcapsules in accordance with the method disclosed by Jarvis et al in U.S. Application S.N. 243,586 filed March 14, 1 981, or by other methods. The cells are then subjected to an IFN induction technique by exposure to certain viruses or nucleic acids. For example, the growth medium can be replaced with a medium containing a known viral or nucleic acid type IFN inducer. After a short incubation, typically on the order of an hour, the inducer is removed, and the cells are placed in normal, fresh growth or maintenance medium and incubated, e.g., for 24 hours.Exposure of the cells to the inducer triggers expression of the cells' DNA code for epsilon interferon production. During the subsequent incubation, the cell will synthesize and secrete IFN-E. At this point the medium is harvested and if assayed by, for example, the method of Ho and Enders (Proceedings of the National Academy of Science, Volume 45, page 385-389, 1959), will be found to contain IFN-E antiviral activity.

The product which results from this procedure does not show significant cross-reactivity with antibody to IFN-a, IFN-P, or IFN-y in neutralization experiments. It is stable at pH 2, and is proteinacous in character. Immunsabsorption experiments indicate that anti IFN-P antisera reacts with a portion of the interferon product of induced epithelial cells. The new interferon also exhibits a profile of activity in mammalian cells which is different from known types of interferon.

These chemical characteristics mark IFN-E as a unique substance.

Currently, IFN-E has been produced successfully from normal epithelial cells derived from human epidermic, conjunctiva, vagina, and esophagus, employing "normal induction" (Field et al., Proceedings of the National Academy of Sciences, Vol. 58, p. 1004-1009, 1967) and "superinduction" (Havell and Vilcek, AntimicrobialAgents in Chemotherapy, Vol. 2, pp. 476-484, 1 972) with the nucleic acid, poly l.C (Miles Laboratories), with Newcastle Disease Virus, (Baron and Issacs, British Medical., Vol.

56, pp. 1 8-20, 1962) and with Sendai virus (parainfluenza-1, Gresser, Proceedings of the Society of Experimental and Biological Medicine, Vol. 108, pp. 303-307, 1961). The results of assays indicate that each of the induction techniques and each of the different types of cell cultures tested produce IFN-E. Production levels vary with the type of inducer employed and with the various modifications of the induction techniques. Newcastle Disease Virus works well. Other viruses which may be used are set forth in the following table.

Table I Virus References Influenza-A Gresser and Dull (1964); Andrews (1961) Parainfluenza-3 Chany (1960) Measles Petralli, Merigan, and Wilbur (1965a); DeMaeyer and Enders (1961) Mumps Cantell (1961); Waddell, Wilbur, and Merigan (1968) Rubella Neva and Weller (1964) Respiratorysyncytial Ray, Gravelle, and Chin (1967); Moehring and Forsyth (1971) Rabiesvirus Wiktoretal. (1972) Vesicular stomatitis Marcus and Sekellick (1977); Vilcek, Yamazaki and Havell (1977) Chikungunya Zimmerman et al. (1972) Sindbis Gresser and Enders (1962) Western Equine encephalitis Luby, Sanders, and Sulkin (1971) Yellow Fever Wheelock and Sibley (1965); Wheelock and Edelman (1969) Poliovirus-Type 1 Gresser, Chany, and Enders (1965) Poliovirus-Type-2 Smorodintsev et al (1970);Ho and Enders (1959a,b) Encephalomyocarditis Steward II, Gosser, and Lockart (1971 a) Rhinovirus-2 Smorodintsev et al (1971 a); Fiala (1972) Rhinovirus-1 2 Gatmaitan, Stanley, and Jackson (1973) Reovirus-2 Oie, Loh, and Ratnayake (1973) Blue tongue Jameson and Grossberg (1977) Adenoviruses Lysovetal. (1971) Varicella-zoster Vaczi, Horvath, and Hadhazy (1965) Human cytomegalovirus Vaczi, Horvath, and Hadhazy (1965); Glasgow (1974) Herpes simplex Rasmussen et al. (1974) Vaccinia Wheelock (1964); Epstein, Stevens, and Merigan (1972) It has also been discovered that inteferons produced from a selected type of human cell such as epithelial, fibroblast, vascular, hepatic, renal, etc., are particularly effective as antiviral or antitumor agents with respect to treatment of human tissue consisting of the selected cell type.Thus, IFN-E is most effective when used to treat epithelial cells, and will be found to have a lower level of activity when used to treat other human cell types. Similarly, lFN-p (of human fibroblast cell origin) is most effective in treating human fibroblast cells or tissue, and likewise has lower antitumor and antiviral properties with respect to cells of other human tissue types.

Accordingly, the following procedure may be employed to protect a given cell type or tissue from viral infestion, to arrest multiplication of a virus which has infected a given cell type or tissue, or to arrest growth of a tumor in a given tissue.

First, a culture of cells of the type in question is grown by conventional techniques such as (in the case of epithelial cells) the Green technique referred to above. The cells of the culture are then treated to induce expression of the portion of its DNA code responsible for the production of interferon E. This may be done, for example, by the lFN induction techniques discussed hereinafter. The product is then harvested, typically from the medium after incubation. The product may be used as an effective antiviral or antitumor substance for treatment of the cell or tissue type in question. In vitro, cultures are protected against viral attack by the addition of 2-5 units IFN-Jml. This concentration is effective in protecting epithelial cells having a density of O. 5x105 to 2.5 X105 cells/cm2.

Significant quantities of this new antiviral substance can also be produced by two additional currently known cellular techniques. The first involves the use of chemically, virally, or spontaneously transformed eukaryotic cell types similar to those previously employed in the production of lFN-afrom human lymphoblastoid cell lines. The second employs currently known recombinant DNA techniques similar to those currently employed in the production of IFN-a and IFN-/3.

In the first method, transformed cells capable of producing IFN-E can be obtained by employing one of two techniques: in vivo or in vitro. The first technique (in vivo) involves the direct isolation from fresh tissue of transformed cell types; the second technique (in vitro) involves a selection procedure in which a strain of normal human diploid cells is either cultured on a long term basis until a small but detectable number of the population undergoes spontaneous transformation, or the initial strain is treated for a brief period of time with a virus or a known mutagen such as EMS, MMS, or MNNG to increase the frequency of transformation. Certain of the cells transformed in either of the two in vitro ways described will contain a portion of the DNA code of epithelial cells responsible for the production of IFN-E.

Transformed cultures thus obtained may be induced to produce IFN-E by employing standard induction techniques as described previously for normal human diploid epithelial cells.

In the second technique, the mRNA synthesized in epithelial cells in reaponse to the IFN induction is extracted from the cells optionally purified by ultracentrifugation or the like to obtain an mRNA fraction of increased specificity, and the extract is used as a template for the synthesis of complementary DNA (cDNA). The cDNA may be produced using the Avian Myoblastosis reverse transcriptase enzyme. The final resulting product of this process, double standard complementary DNA (dscDNA), which includes the sequence coding for the synthesis of IFN-E, and a bacterial or eukaryotic vector are then treated with a restriction enzyme which cleaves the DNA and produces bonding sites by which the dscDNA and the vector may be attached. The vector and DNA are then mixed together, annealed, and covalently bonded employing a ligase enzyme.At this point the recombinant vector preparation is used to transform either a bacterial or eukaryotic cell. The transformed cells thus contain the DNA complementary to all or part of the RNA originally contained in the epithelial cells during their IFN-E production stage.

These recombinant vectors are then incubated with an appropriate cell type which permits the vector to operate. This results in a cell population which includes individual cells capable of synthesizing and secreting IFN-E. The cell population is then screened for a subpopulation of cells which produce IFN-E, and this subpopulation is cultured to establish a cell line capable of producing relatively large quantities of IFN-E.

Further particulars of this recombinant DNA technique may be found, for example, in the following references, the disclosures of which are incorporated herein by reference.

1. U.S. Patent No. 4,237,224 to Cohen et al entitled "Process for Producing Biologically Functional Molecular Chimeras".

2. Scheller, R. et al. 1977, Clones of Individual Repetitive Sequences from Sea Urchins DNA Constructed with Synthetic Eco R1 Sites, Science, Vol. 196, pp. 197-200; 3. Blattner, F. et al. 1977, Charon Phages: Safer Derivatives of Bacteriophage Lambda for DNA Cloning, Science, Vol. 196, pp.161-169; 4. Broach, J. R., and Hicks, J. B., 980, Replication and Recombination Functions Associated with YeastPlasmid, 2 Micron Circle, Cell, Vol. 21, pp. 501-508; and 5. Hamer, D. 1981, Synthesis Processing and Secretion of Eukaryotic Proteins in the SV40- Monkey Cell System, Recombinant DNA Abstracts, Vol. 1, p. 4.

The invention will be further understood from the following, non-limiting examples.

Example 1 An epithelial cell culture of human epidermal origin obtained from the laboratory of Howard Green at Massachusetts Institute of Technology was grown to a cell density of 1-2 x 05 cells per cm2 in a minimal essential medium (MEM, Gibco) containing 10% heat inactivated fetal calf serum (incubated at 560C for 30 minutes: HIFCS).Four equivalent cultures incubated in MEM and 2% HIFCS were exposed for 1 hour at 37CC to, respectively, 0, 1, 10, and 100 micrograms/ml (hereinafter ug/ml) poly (l.C)-(9S), a high molecular weight nucleic acid commercially available from PL Laboratories (Milwaukee, Wisconsin). (Field et al. induction technique, supra.) The cultures were then washed twice with MEM and incubated in the same medium supplemented with 2% HIFCS for 24 hours. The medium was then collected and assayed by the method of Stewart II, The Interferon System, pp. 17-1 8. No detectable anti-viral activity was observed in the cultures incubated with 0 and 1 ug/ml poly (l.C).

Approximately 10 units/ml IFN-E were detected in the samples obtained from cultures induced with 10 and 100 ug of poly IC).

Example 2 The procedure of Example 1 was repeated except that a modified "superinduction" technique was employed (Havell and Vilcek, supra) in an attempt to increase the IFN-E production. 50 ug/ml cyclohexamide (protein synthesis inhibitor) was included together with the poly (I-C). After incubation for one hour the cells were washed and incubated for an additional 3 hours in MEM containing 2% HIFCS and 50 ug/ml cyclohexamide. The cells were then again washed and incubated with MEM containing 50 ug/ml cyclohexamide and 5.0 ug/ml Actinomycin D (RNA synthesis inhibitor) for an additional 2 hours. At the conclusion of this incubation the cells were washed twice and then incubated for 24 hours at 370C in normal medium.Harvest of the medium and assay as set forth in Example 1 resulted in the production of no detectable IFN-E in the culture induced with 0 and 1 ug/ml poly (I-C). The culture induced with 10 ug/ml poly (I-C) resulted in 100 units/ml IFN-E. The cultures induced with 1 00 ug/ml poly (I-C) resulted in the production of greater than 330 units/ml IFN-E.

Example 3 An epithelial cells culture of epidermal origin containing 1-2 x 05 cells (Example 1) was used to produce IFN-E employing the Newcastle Disease Virus (NDV) induction method (Baron and Issacs, supra). The virus used was the Bankowski strain of NDV available from Poultry Health Laboratories, Davis, California.Four 1 ml samples of MEM containing 2% HIFCS and sufficient NDV to provide a final concentration of, respectively, 0, 100-200, 25-50, and 1-1 6 virus pfu/cell in the test samples were prepared. 0.2 ml of the respective 1 ml virus preparations (virus inducer) were then added to the cell cultures which were shaken at 5 minute intervals for 30 minutes at 370C. The remaining 0.8 ml portion of the NDV preparations were then added and the cultures were incubated for an additional 30 minutes. The cell cultures were then washed twice with normal medium, 1 ml of fresh medium was added to each culture, and the cultures were incubated for 24 hours.

The medium was then harvested, acidified to pH 2 with 0.1 N HCI, and stored at 40C for 5-6 days to inactivate the NDV. Thereafter, the harvested medium was neutralized with 0.1 N NaOH and assayed for IFN-E. In the sample containing no NDV (control), antiviral activity was absent. The sample induced with 100--200 virus particle/cell of NDV contained approximately 500-1000 units of IFN E/ml. The culture induced with the NDV preparation containing 25-50 virus particles/cell contained between 170 and 330 units IFN-E/ml. The culture induced with the NDV preparation containing 1-16 virus particles/cell contained between 50 and 1 70 units IFN-r/ml.

As can be seen from this experiment, a cell culture containing 1-2 xl 05 cells/ml can produce about 500-1000 units IFN-E/ml. The yield is accordingly on the order of 1 unit IFN-E produced per 100--500 cells in the culture.

Example 4 The procedure of Example 3 was repeated except that parainfluenza-1 (Sendai) virus, at various concentrations, was used as an inducer in place of NDV (see, Gresser, supra). As purchased (Flow Laboratories), the Sendai virus contained 1 0,000-40,000 heamaglutinating units per ml.

In a culture wherein the commercial virus preparation was diluted 1-3, 1000 units/ml IFN-E were produced. A 1-10 dilution resulted in 200 units/mi, and a 1-33 dilution resulted in 100 units/ml. No antiviral activity could be detected in the control culture wherein the virus was omitted.

The table set forth below summarizes the results of experiments (Examples 1-4) using human conjunctival cell cultures and human epidermal cell cultures using various induction techniques.

Table II Results of interferon experiments Units IFN-E/ Cell type Induction method Units IFN-/ml cell Human polylC* 33 3.3x10-5 Conjunctival polylCt 330 3.3x10-4 None (control) O 0 Human poly l.C* O 0 Epidermal Sendaitt 330 3.3 x 10-4 NDV** 1000 1.Ox 10-3 None (control) O 0 *Induction method of Example 1 using 100 ug/ml poly I.C tlnduction method of Example 2 using 100 ug/ml poly l.C **lnduction method of Example 3 using 1::3 dilution of NDV (100-200 virus particles/cell) ttlnduction method of Example 4 using a 1:3 dilution of the stock solution.

Example 5 Since interferons have classically, by definition, been shown to be proteins, samples of IFN-E were tested in order to determine whether the antiviral activity present in the samples was due to a protein component. The experiments were performed by incubating samples of IFN-a (NIH standard), IFN-p (NIH standard), and IFN-E (derived from NDV-induced epithelial keratinocytes) in the presence of known proteolytic enzymes and then analyzing the treated material for remaining antiviral activity.

Samples of IFN-a (16 units/ml), IFN- (16 units/ml), and IFN-E (32 units/ml) in MEM were incubated at 370C for one hour in the presence of either trypsin (Signa Chemical Co., St. Louis, Missouri) or Pronase (Sigma) at concentrations of 3.3-100 ug/ml. after completion of the reaction, remaining proteolytic activity in the samples was neutralized by the addition of HIFCS to a final concentration of 33% (v/v). The samples were then assayed for remaining IFN-induced antiviral activity by microtitration on GM-2767 cells.The data is shown below: Proteolytic sensitivity of IFN-E Enzyme Remaining IFN antiviral activity Proteolytic concentration enzyme (ug/mI) IFN-a IFN-p IFN-E Trypsin 100 0 0 0 33 0 0 0 10 1 2 4 3.3 8 8 16 0 16 16 32 Pronase 10 0 0 0 3.3 0 0 0 0 16 16 32 These data demonstrate that the antiviral activity observed in samples of IFN-E is due to the presence of a protein component or components, based on complete elimination of activity by treatment with proteolytic enzymes. The proteinaceous character of IFN-E is also apparent from the fact that RNA and protein synthesis inhibitors block production of IFN-E.

Example 6 This example shows the stability of IFN-E to acid pH. IFN-E was produced by induction of human epidermal cells with NDV in accordance with Example 3. When the medium was harvested, it was divided into three portions. The virus was inactivated in the first portion as described in Example 3 by acidifying to pH 2 with HCI. The second portion was filtered for 5 days through a Millipore 01310 filter (100,000 molecular weight limit, Millipore Corp., Bedford, MA) to remove the virus. The third portion was filtered through a Millipore PTMK01310 filter,100,000 molecular weight limit, which has been previously soaked for 4 hours in a 1% solution of BSA.

In each case the resulting medium was assayed for both IFN and for the presence of resdiual virus. The results are set forth below: Stability of IFN-E to acid pH Treatment Initial virus Final virus IFN Titre oflFN- titre zpfu/ml) titre (pfu/ml) (u/ml) pH2,5days 1.1x108 0 128 Filter through 1.1x108 0 128 PTHK01310 Filterthrough 1.1x108 0 128 PSVP01310 None of the samples showed any viral activity and each sample contained the same amount of IFN-E activity. This shows that IFN-E activity is stable at acid pH.

Example 7 The immunological properties of IFN-E (produced by NDV induction of human epidermal cells in accordance with Example 3), lFN-a', IFN-P (NIH), and IFN-y (obtained from S. Baron, University of Texas Medical School, Galveston, Texas) were compared. Neutralization titrations of each interferon were carried out using anti-lFN-a (NIH), anti-íFN-,B (NIH and Y. H. Tan, Calgary University, Calgary, Alberta, Canada), anti-IFN-y (S. Baron), and mixtures of these antisera. Serial 2-fold dilutions of the appropriate antisera were made in MEM using a 96-well microtiter plate. Dilutions ranging from 32 lU/ml 2 IU/ml-of each IFN preparation were added to the wells and the plates were incubated first at 370C for one hour and then at 40C for one hour to allow the formation of antibody-lFN complex. Human fibroblast cells (GM-2767-Human Mutant Cell Repository) were then seeded at 2x 104 cells per well, and after 2024 hours incubation at 370C, were challenged with a standard dose of Vesicular Stomatitis Virus (1 pfu/cell). After 24-48 hours at 370C each well was observed under the microscope for interferonmediated inhibition of virus-induced cytopathic effect (CPE). The results are summarized below: Amount of antiserum required to neutralize IFN (neutralizing units per ml) IFN Conc. Anti-a 8 IFN Sample units/ml Anti-aa Anti-/3a anti-pa IFN-a 16 N.D. > 3000 N.D.

(NIH std) 8 N.D. > 3000 N.D.

4 62 > 3000 125 lFN-p 8 > 3000 375 750 (NIH std) 4 > 3000 188 750 2 > 3000 188 47 IFN-E 8 > 3000 > 3000 188 AR-IVb 4 > 3000 375 188 2 > 3000 375 47 IFN-E 16 > 3000 > 3000 375 AOBC 8 > 3000 > 3000 375 2 > 3000 1500 94 Control sheep serum to fibroblast IFN has no neutralizing activity.

N.D.=not determined.

a-Antisera prepared in sheep against respective IFNs.

,--Derived from epidermal keratinocytes, not necessarily fibroblast-free (NDV-induced).

,derived from fibroblast-free epidermal keratinocytes (NDV-induced).

It can be seen from the foregoing table that a) IFN-a is neutralized by both anti-a and a mixture of anti-a/anti-p antisera. As expected, it takes twice as much anti-a/p as anti-a to neutralize the same amount of IFN-a. (Like quantities of antisera neutralize like quantities of lFN-a.) b) lFN-p is neutralized by both anti-p and a mixture of anti-a/anti-p antisera. Again, it takes twice as much anti-a/p as anti-p to neutralize the same amount of lFN-p.

c) It takes less than half as much anti-a/anti-p mixture as anti-p to neutralize the same amount of IFN-E. IFN-E does not react with antisera directed against IFN-y.

This example demonstrates that IFN-E is immunologically distinct from IFN-cr and IFN-/3, and IFNy and is therefore chemically distinct from these known human interferons.

Example 8 In addition to serological evidence presented in Example 7, IFN-E which had been immunoprecipitated in part was used in a neutralization experiment. To samples of this interferon were added anti-p IFN (polyclonal obtained from NIH or monoclonal obtained from Y. H. Tan). The mixture was incubated for one hour at 370C followed by one hour at 40C to allow antibody-lFN complex formation, and then the complex, together with any excess antibody, was removed by incubation of complexed and uncomplexed antisera in the presence of protein A Sepharose (Sigma) at a ratio of 30 ul to 1.5 ul of reaction mixture. Remaining uncomplexed IFN was subjected to a second round of immunoprecipitation, and the samples were assayed for IFN activity as described using the CPE inhibition assay. The results are set forth below: IFN IFN Orig.IFN l stAnti- titre 2ndAnti- titre IFN titre body precip. after body precip. after sample units/ml using: 1st using: 2nd AOB-lFN-E 1024 Anti-p 256 Anti-p 256 polyclonal polyclonal AOB-IFN-E 1024 Anti-p 1024 Anti-p 1024 monoclonal monoclonal AOB-IFN-E 1024 Anti-p 1024 Anti-p 1024 monoclonal polyclonal AOB-IFN-E 1024 Anti-p 256 Anti-p 256 polyclonal monoclonal When the pre-precipitated AOB-IFN-E was used in the neutralization test, performed exactly as in Example 7, the following results were obtained:: Amount of antisera required to neutralize IFN (neutralizing units) Mixture IFN Conc. of IFN Sample u/ml Anti-p Anti-a/p AOB lFN-E absorbed 16 > 3000 1 500 with anti-p 8 > 3000 375 antiserum 4 > 3000 94 AOB IFN-E absorbed 16 > 3000 1500 with control 8 > 3000 188 antiserum 4 1 500 94 AOB IFN-E 8 > 3000 1500 not absorbed 4 375 94 1 12 24 NlHIFN-p 16 750 750 8 188 375 4 94 188 Control sheep serum to fibroblast IFN had no neutralizing activity.

The data of the foregoing tables show that a portion of the IFN-E preparation cannot be neutralized by anti-p antiserum even when two different anti-p antibodies are employed. However, even after thorough precipitation with anti-p antisera, an additional portion of IFN-E can be precipitated with mixed anti-a/p antiserum. These data constitute additional evidence that IFN-E is immunologically and chemically distinct from other known human interferons.

Example 9 IFN-E produced by the induction of human epidermal cells with NDV in accordance with Example 3 can be purified and concentrated by affinity chromatography on a number of immobilized ligands.

These include, but are not limited to, reactive red agarose, phenyl sepharose (Sigma Chemical Co., St.

Louis, Mo.), procion red agarose, phenyl agarose (BRL, Gaithersburg, Md.), Glycogel B, N-(3carboxypropionyl)aminodecane (CPAD) agarose and N-pyromellitylaminodecane (PMAD) agarose (Pierce Chemical Co., Rockford, ill.).

8 ml IFN-E in MEM containing 900 units of IFN-E were applied to a 0.5 ml reactive red agarose column. The column was washed with 10 ml of 20 mM phosphate buffer (pH 7.4). The IFN was then eluted with 3 2 ml aliquots of 50% ethylene glycol in 20 mM phosphate buffer containing 1 M sodium chloride. Each of these fractions was collected in an equal volume of 20 mM phosphate buffer to dilute the ethylene glycol. The column was finally washed with an additional 5 ml of 20 mM phosphate buffer. The protein concentrations were measured by the absorbance at 280 nm (OD-280).The results are shown in the following table: Purification of IFN-E using reactive red agarose IFN Per Titre Total Purifi- (Total centage Column (units Volume OD cation IFN recovery fraction Or280 /ml) (my) units (fold) units) of IFN Original 2.95 128 7 20.65 0 896 100 material Column 1.995 8 7 13.97 - 56 0.5 pass-through 1st20mM 0.911 4 5 4.6 - 20 0.02 phosphate wash 2nd 20 mM 0.088 2 5 0.44 - 10 0.01 phosphate wash lst2mlethy- 0.721 128 4 2.88 7.2 512 57 lene glycol 2nd 2 ml ethy- 0.123 256 4 0.49 43 1024 114 lene glycol 3rd 2 ml ethy- 0.021 16 4 0.084 250 64 0.7 lene glycol 20 mM phos- N.D. 0 5 - - 0 0 phate wash The first two ethylene glycol fractions were combined and the resulting fraction contained all the IFN-E applied to the column and showed a 6-fold purification over the original material.

In other similar experiments up to 68 ml of IFN-E were applied to the column and the IFN was recovered in a final volume of 8 ml. This represents an 8.5-fold concentration of the material.

If the IFN-E is applied to two or more of these immobilized ligands in sequence, then a greater purification can be obtained. For example, an IFN-E sample was applied to a CPAD column and eluted as described in the above procedure. The combined IFN-containing fractions showed approximately a 9-fold purification and were dialysed against 20 mM phosphate buffer to remove the ethylene glycol.

The sample was then applied to a PMAD agarose column which was developed in the same way. The IFN-E was recovered in 70-1 00% yield with a further 5-fold purification with respect to total protein.

This example shows that IFN-E can be purified by any one or a combination of immobilized ligands.

Example 10 Unfractionated IFN-E is first centrifuged at 300 xg for 20 minutes. The supernatant is mixed end over end on a Fisher Rota rack with controlled pore glass (C.P.G.) (Electronucleonics, Inc.) at 40C. 0.28 gms of controlled pore glass is used per 30 ml of culture supernatant. After 3 hrs, the supernatant is removed and the controlled pores glass beads are packed in a column (2.5x3.1 cm). The column is first washed with 4 column volume of PBS followed by 4 column volumes of 1.0 M NaCI-20 mM phosphate buffer, pH 7.4. The IFN-E is eluted with 4 columns volumes of 50% v/v ethylene glycol -1.0 M NaCI-20 mM phosphate buffer pH 7.4 and collected into an equal volume of 1.0 M NaCI-20 mM phosphate buffer to give a final concentration of ethylene glycol of 25%. The IFN-E containing fraction is concentrated by ultrafiltration to 1.0 to 1.6 mis and applied to a polyacrylamide-agarose (Ultragel AcA54) column (1 .6x85 cm) equilibrated with 25% ethylene glycol (V/V)--1.0 M NaCI-20 mM phosphate buffer 7.4. The IFN- was then eluted with equilibrating buffer at a flow rate of 6.0 mls per hr 2.0 ml fractions were collected and assayed for their IFN and protein content (Loury assay).The results are shown below: Purification of IFN- using controlled pore glass and Ultragel AcA 54 IFN Total Specific Purifi- % Column Protein Titer Volume Total IFN protein activity cation Recovery fraction mg/ml u/m. (mI) units (mg) {x 104) fold oflFN Original .034 2.560 420 1,075,200 14.28 7.53 - 100 material (after centri fugation) C.P.G. super- N.D. 32 405 12,960 - - - 1.20 natant 1stwash PBS N.D. 32 55 1760 - - - 0.16 2ndwash 1.0 M N.D. 16 56 896 - - - 0.08 NAcl-20 mM PO4 7.4 50% Ethylene 0.094 7776 112 870,912 10.53 8.27 1.10 81.0 glycol 1.0 m NaCI20 mM PO4 7.4 Concentrated N.D. 512,000 1.6 819,200 N.D. - - 76.2 50% ethylene glycol fraction Ultragel AcA 0.05 25,600 31.0 793,600 1.55 51.2 7.0 73.8 pool Example 11 Preparations of IFN-E, lFN-a, and IFN-p, including crude, partially purified in accordance with Example 10, (p-lFN-E), and anti-mouse interferon-absorbed p-lFN-E were subjected to molecular weight analysis using SDS gel electrophoresis according to the method of Lamelli. The samples were incubated at room temperature for 1 hour in the presence of 1.0% SDS, 0.05 M tris-HCI (pH 6.8) buffer, 10% (v/v) glycerol, and 0.001% bromo phenol blue, loaded onto 12.5% polyacrylamide gels, and run for approximately 1 6 hours. After electrophoresis the lanes containing the molecular weight standards were stained. IFN containing lanes were sliced into 3 mm slices, and extracted by shaking with 0.5 ml PBS containing 0.5% SDS for 20 hours at room temperature.Assay of these fractions was performed in GM-2767 cells and the molecular weight of each activity peak was calculated by comparison with its position of the gel with the molecular weight standards. The results are set forth below: Molecular weight of IFNs IFN M. W. of activity peaks {% of activity recovered) a 18K (27.6) 19.7K (44.8) 22.5K (27.6) ,B ~ (1.83) - - 22.0K (98.2) E (crude) 17.5K (41.0) 20.0K (7.4) 22.0K (51.6) p-lFN-E 17.5K (35.4) 20.0K (2.53) 22.0K (62.0) Unbound p-lFN-E 17.5K (19.2) 20.0K 0 22.0K (90.8) Bound p-lFN-E 17.5K (68.0) 20.0K (15.3) 22.0K (16.8) All three fractions obtained with IFN-E samples show antiviral activity on GM-2767 cells.The 22.0x 103 mw fraction obtained from IFN-E exhibits no activity on bovine kidney cells (MDBK). The other two IFN-E fractions (17.5K, 20.0K) were approximately 25% as active on MDBK cells as on GM2767 cells.

Example 12 The antiviral potency of IFN-E (produced with NDV induction from human epidermal cells in accordance with Example 3), IFN-a: (obtained from the National institute of Health, NIH), and IFN-p (NIH) was compared in cultures of normal human epidermal cells versus normal human diploid fibroblast cells. Serial dilutions of the three IFN preparations were added to the wells of a microtiter plate together with standard cultures of normal human epidermal cells and normal human fibroblast cells. After an 1 8-24 hour incubation at 370C, the cells in each of the wells were challenged with a standard dose of Vesicular Stomatitis Virus (1 pfu/cell).After 36-96 hours (when controls showed 100% cell death), each well was observed under the microscope for assay of resistance to the virus.

The results are summarized below.

Table Ill Units of activity/ml of IFN on: Interferon Fibroblasts {FS-4)a Epithelia IFN-E 32-64 512-2048 IFN-a 128-512 64 lFN-p 32-256 8-32 aNormal human diploid fibroblasts obtained from J. Vilcek (N.Y.U., N.Y., N.Y.).

As is evident from the foregoing table, IFN-E produced by the DNA code of epithelial cells is significantly more potent (up to approx. 1 6x) as an antiviral agent with respect to epithelial cells than with respect to fibroblasts. IFN-a, produced from the DNA code derived from human leukocytes and lymphoblastoid cells, is more potent (up to 8x) with respect to fibroblasts than with respect to epithelia, and lFN-p, produced from the DNA code derived from fibroblasts is more potent (up to 32x) with respect to fibroblast cells than with respect to epithelial cells.

Example 13 The procedure of Example 11 was used to assay the relative antiviral activity of IFN-a, p, y, and E on bovine kidney cells (MDBK cells ATCC CCL22) versus human trisomic fibroblasts (3 copies of chromosome 21 -GM-2767 Human mutant cell repository 47, XX, t2 1). The results of the assay are set forth below: Units of activity/ml of IFN on IFN Type MDBK GM-2767 a > 512 512 p 0 128 Y 0 128 E 8-16 32 The results for lFN-a, p, and y are consistent with previously reported data. The results for IFN-E indicate that this new interferon is between about 2 and 4 times as active on GM-2767 as on bovine kidney cells.

Example 14 The novel interferon also exhibits antiproliferation activity as judged by the inhibition of multiplication of human lymphoblastoidcells (Daudi cells). Approximately 10,000 human lymphoblastoid cells were plated together with 200 ul of media in each well of a microtiter plate.

Various dilutions of IFN-a, p, and E were then added to each well and the number of viable cells were counted after 1,2, and 4 days. The results are set forth in the following table: Daudi cell assay Number of viable cellsfwell (x 1 O-) IFN Type and concentration Day 1 Day 2 Day 4 100 00 u 6.0 4.8 6.0 10 u 4.8 5.2 6.0 -1 00 u 7.2 9.6 6.4 lOu 5.2 9.6 18.4 E-1 00 u 6.0 5.2 4.8 lOu 5.6 5.2 6.0 Control (no N) 8.0 21.5 67 These results demonstrate that 10 units of IFN-E can suppress Daudi cell growth by greater than 90%.

Example 15 Samples of IFN-α, p, and were each equilibrated with phosphate buffered saline (PBS), sodium dodecyl sulphate (SDS), and a mixture of SDS, p mercaptoethanol (BME), and urea at 100 C. The detergent SDS is known to denature proteins by altering their conformation, BME reduces disulfide bonds. It was discovered that IFN-a, p and E differ markedly in their sensitivity to these reagents.

Specifically IFN-a was stable in SDS alone but was substantially inactivated in the presence of an SDS, urea, BME mixture; lFN-p was mostly inactivated in SDS alone but stable in SDS/urea/BME mixture; and IFN-E is relatively stable in SDS alone and in an SDS/BME urea mixture. In PBS at 100 C, less than 5% of the activity of all three interferons remains.

In similar experiments it was determined that IFN-a which had previously been inactivated with an SDS-BME mixture could be reactivated by equilibration at 100C with an SDS-PBS mixture, that lFN-p which had previously been inactivated in SDS alone could be reactivated in a mixture of SDS, BME, urea, and PBS, and that IFN-E which had previously been inactivated in PBS could be reactivated by equilibration at 100C in the presence of PBS/SDS with or without BME/urea. These observations are further evidence of the uniqueness of the IFN-E preparation of the invention.

Example 16 This experiment compared the binding and elution properties of IFN-E and IFN-p on a Blue Sepharose column (Pharmcia). The columns were prepared in accordance with the manufacturers instructions and loaded at room temperature with either lFN-p or IFN-E samples in PBS. Elutions were conducted using 1.0 M NaCI in PBS and 50% (v/v) ethylene glycol in 20 mM sodium Phosphate Buffer.

Each interferon preparation was run on 2 columns containing different Blue Sepharose lots, and the following data were obtained: Binding of IFNs to Blue Sepharose Percent activity recovered with Fraction IFN-p IFN-E Column pass-through 44.0-44.2 2.5-10.2 +20 mM phosphate buffer Elution with 1.0 M NaCI 2.7-4.3 19.3-30 Elution with 50% ethylene 51.9-52.3 60.0-78.2 glycol/1.0 M NaCI in 20 mM phosphate buffer This example shows that IFN-E and IFN-p have differing affinities to Blue Sepharose and differing elution behaviors. Blue Sepharose binds approximately 9097% of the IFN-E activity whereas only about 60% of IFN-p is bound under the same conditions. Additionally, whereas 2030% of the IFN-E is recovered using 1.0 M NaCI, only 34% of the lFN-p activity is recovered under similar conditions.

Finally, up to approximately 80% of the IFN-E activity is eluted with a combination of 50% ethylene glycol/1.0 M NaCI whereas only about 50% of the IFN-p activity is recovered under similar conditions.

Other embodiments are within the following claims.

Claims (22)

Claims

1. An interferon preparation having antiviral activity, said preparation being produced by the steps of culturing living cells containing at least a portion of the DNA code of a human epithelial cell to synthesize said interferon and harvesting a fraction rich in said interferon from said culture, said interferon preparation being characterised in that i) it is stable at pH 2, ii) its antiviral activity is destroyed by proteolytic enzymes, iii) it exhibits no cross reactivity with antibody to interferon-a or interferon-y, and, iv) its antiviral activity on MDBK bovine kidney cells is 1/4 to 1/2 of its activity on human trisomic fibroblasts (GM-2767).

2. The interferon preparation of claim 1 wherein said culture is a human epithelial cell culture.

3. The interferon preparation of claim 1 wherein said culture comprises transformed eukaryotic cells.

4. The interferon preparation of claim 1 wherein said culture comprises cells wherein said portion of the DNA code is inserted into said cells by recombinant DNA techniques.

5. The interferon preparation of claim 1 characterized by a unique profile of cross reactivity to a mixture of anti-a and anti-p antisera such that twice as much of said interferon is neutralized by the mixture as is neutralized by anti-p antisera alone based on equal units of neutralizing activity.

6. A process for producing an interferon preparation comprising the steps of: A. producing a culture of living cells containing a portion of the DNA code of human epithelial cells; B. incubating said cells in a medium under conditions to promote synthesis of interferon; and C. harvesting a fraction rich in said interferon from said cells, said interferon being characterized in that i) it is stable at pH 2, ii) its antiviral activity is destroyed by proteolytic enzymes, iii) it exhibits no cross reactivity with antibody to interferon a or interferon 1^, and, iv) its antiviral activity in MDBK bovine kidney cells is 1/4 to 1/2 of its activity on human trisomic fibroblasts (GM-2767).

7. The process of claim 6 wherein said culture of living cells is a human epithelial cell culture and, prior to step B, a virus is used to induce said cells to produce said interferon.

8. The process of claim 6 wherein the culture of step A comprises transformed eukaryotic cells.

9. The process of claim 6 wherein the culture of step A comprises cells wherein said portion of the DNA code is inserted into said cells by recombinant DNA techniques.

1 0. The process of claim 7 wherein the virus is selected from the group consisting of Newcastle Disease Virus and Sendai virus,

11. The process of claim 1 0 wherein the virus is the Bankowski strain of Newcastle Disease virus.

1 2. The process of claim 7 wherein said cell culture is grown from cells sampled from a human epithelial tissue selected from the group consisting of epidermis, conjunctiva, vagina, and esophagus.

13. A method of treating living cells to inhibit virus infection, said method comprising the step of: contacting cells to be treated with an effective amount of the interferon preparation of claims 1, 2, 3, 4 or 5 to inhibit said virus infection.

14. The method of claim 1 3 wherein said living cells comprise human epithelial cells.

1 5. A method of treating neoplastic living cells to inhibit proliferation of said cells, said method comprising the step of: contacting cells to be treated with an effective amount of the interferon preparation of claims 1, 2, 3, 4 or 5 to inhibit said proliferation.

1 6. The method of claim 1 5 wherein said neoplastic living cells comprise human epithelial cells.

1 7. An interferon E preparation for use in inhibiting virus infection.

1 8. An interferon E preparation for use in inhibiting proliferation of neoplastic living cells.

1 9. An interferon preparation substantially as hereinbefore described.

20. A process for producing an interferon preparation substantially as hereinbefore described.

21. A method of treating living cells substantially as hereinbefore described.

22. Interferon E.