AU707739B2 - Suppressor T-cell hybridoma - Google Patents
Suppressor T-cell hybridoma Download PDFInfo
- Publication number
- AU707739B2 AU707739B2 AU65731/96A AU6573196A AU707739B2 AU 707739 B2 AU707739 B2 AU 707739B2 AU 65731/96 A AU65731/96 A AU 65731/96A AU 6573196 A AU6573196 A AU 6573196A AU 707739 B2 AU707739 B2 AU 707739B2
- Authority
- AU
- Australia
- Prior art keywords
- gif
- antigen
- human
- cell line
- specific
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Landscapes
- Preparation Of Compounds By Using Micro-Organisms (AREA)
Description
4oa.
SUPPRESSOR T-CELL HYBRIDOMA This application is a continuation-in-part of 709,375, filed June 3, 1991, which is a continuation-in-part of Serial No. 533,889, filed June 4, 1990.
FIELD OF THE INVENTION The invention relates to human glycosylation inhibiting factor (GIF) specific for an antigen and which can be used to suppress the human immune response to the antigen.
DESCRIPTION OF THE BACKGROUND ART Although the immune response is often seen as beneficial, in certain circum- 10 stances the immune response to an antigen can actually be harmful to the animal in which the immune response occurs. An example where the immune response creates a condition wherein the host is subject to serious pathologic sequelae is in such autoimmune diseases as lupus erythematosus. In lupus erythematosus, antibodies are often present which react with 15 determinants in the thyroid, erythrocytes, DNA, and platelets of the host.
Another example of where the suppression from immune response would be described is in the treatment of allergies. It has been established that IgE antibodies against allergens cause hay fever, and are involved in the other allergic diseases such as extrinsic asthma. The crucial role of IgE antibodies in the allergic diseases raised the possibility that the regulation and suppression of the IgE antibody formation against allergens would be one of the fundamental treatments for allergic diseases. For example, in the serum of hay fever patients sensitive to ragweed allergens, IgE antibodies against the allergens are always detected. The IgE antibody titer goes up after the pollen season, and then declines only slightly during the rest of the year. Since the half life of IgE in the serum is only 2 to 3 days, the persistence of the IgE antibody titer indicates that the antibodies are being synthesized continuously by the lymphoid cells of the patients in spite of the lack of allergen exposure.
Over the past 20 years, several different attempts were made to control the IgE antibody response in experimental animals. One of the approaches was to improve classical immunotherapy or desensitization treatment, in which allergic patients receive repeated injections of a minute dose of allergen. It was shown that the desensitization treatment can improve clinical symptoms in some patients. However, the IgE antibody titer in the serum of hay fever patients did not decline after the treatment. The major immunological effects of the treatment is an enhancement of the IgG antibody formation, and the suppression of an increase in the IgE antibody titer after the pollen season.
A limitation in the desensitization, or immunosuppression treatment is that patients cannot tolerate a large dose of allergen because of side effects. In order to overcome this difficulty, attempts were made to use a chemically modified allergen, such as urea-denatured antigen or polyethylene glycol (PEG)-conjugates of the antigen for the treatment. Since the modified antigens do not bind to antibodies against the native antigen, relatively large doses of the modified antigen can be injected without causing allergic symptoms. However, the modified antigen can stimulate antigen-specific Tcells. Evidence was obtained that intravenous injections of the modified antigen into mice resulted in the generation of antigen-specific suppressor T-cells which suppressed the primary IgE antibody response to the native antigen. However, the treatment had minimal effects on the on-going IgE antibody formation, if the treatment were initiated after the antibody titer reached maximum (Takatsu and Ishizaka, J. Immunol., 117: 1211, 1976).
In agreement with the observations in the mouse, clinical trials of polyethylene-glycol-conjugated allergen in hay fever patients showed that the treatment failed to diminish the IgE antibody titer. Failure of the repeated injections of the modified antigen to suppress the on-going IgE antibody formation is probably due to the presence of a relatively large population of antigen-specific helper T-cells in the allergic patients. Since the modified antigen not only induces the generation of antigen-specific suppressor Tcells, but also expands the population of helper T-cells, this latter effect of the treatment might have overcome the effect of suppressor T-cells. This interpretation is supported by the fact that transfer of antigen-specific suppressor T-cells into immunized mice resulted in the suppression of the on-going IgE antibody formation (Takatsu and Ishizaka, J. Immunol., 117: 1211, 1976). The results collectively suggested that the persistent IgE antibody formation in hay fever patients could be suppressed, if it were S* possible to generate the antigen-specific suppressor T-cells without expanding the helper T-cell populations.
15 Since 1980, the inventors have investigated various ways in which IgE synthesis is selectively regulated in an immunoglobulin isotype-specific manner. As a result of this research, two types of T-cell factors have been found which have affinity for IgE and selectively regulate IgE synthesis. One of the IgE-binding factors (IgE-BF) selectively enhances the IgE response, 20 while the other type of IgE-BF selectively suppresses the response. The major difference between the IgE-potentiating factors and IgE-suppressive factors appears to be carbohydrate moieties in the molecules. The IgEpotentiating factors bind to lentil lectin and concanavalin A, while IgEsuppressive factors fail to bind to these lectins (Yodoi, et al., J. Immunol., 128: 289, 1982). Analysis of the cellular mechanism for the selective formation of either IgE-potentiating factors or IgE-suppressive factors, as well as gene cloning of the factors, indicated that the IgE-potentiating factor and IgE-suppressive factor share a common structural gene and that the nature of the carbohydrate moieties and biologic activities of the factors are established during the post-translational glycosylation process (Martens, et al, Proc. Nat'l Acad. Sci., 84: 809, 1987). Under the physiological conditions, this glycosylation process is controlled by two T-cell factors which either enhance or inhibit this process. These factors are denominated glycosylation inhibiting factor (GIF) and glycosylation enhancing factor
(GEF).
A unique property of GIF is its biochemical activity. This lymphokine binds to monoclonal antibodies against lipomodulin (a phospholipase inhibitory protein) and appears to be a phosphorylated derivative of a phospholipase inhibitory protein (Uede, et al., J. Immunol., 130: 878, 1983). It was also found in the mouse that the major source of GIF is antigen-specific suppressor T-cells (Ts) (Jadieu, et al., J. Immunol., 133: 3266, 1984).
Subsequent experiments on ovalbumin (OVA)-specific suppressor T-cell hybridomas indicated that stimulation of the hybridoma cells with antigen (OVA)-pulsed syngeneic macrophages resulted in the formation of GIF that 15 has affinity for OVA (antigen-binding GIF). However, the same hybridomas constitutively secreted GIF having no affinity for OVA (nonspecific
GIF).
Studies on the relationship between nonspecific GIF and OVA-binding GIF indicated that the antigen-binding GIF is composed of an antigen-binding polypeptide chain and a nonspecific GIF (Jardieu, and Ishizaka, in Immune o 20 Regulation By Characterized Polypeptides, Goldstein, et al., eds., Alan R.
Liss, Inc., p595, 1987). It was also found that the antigen-binding
GIF
shares common antigenic determinants with antigen-specific suppressor Tcell factors (TsF) described by the other investigators, and suppressed the antibody response in an antigen (carrier)-specific manner. Furthermore, not only antigen-binding GIF, but also antigen-specific TsF described by other investigators, bound to immunosorbent coupled with monoclonal anti-lipomodulin (141-B9), and were recovered by elution of the immunosorbent at acid pH.
Despite the major limitations of desensitization in treating allergy, this technique continues to be the method of choice. Consequently, there is significant need for a technique which is antigen-specific yet does not have associated with it the side effects seen with existing desensitization regimens.
The suppression of the immune response is crucial in order to prevent host versus graft (HVG) and graft versus host rejection (GVH). Unfortunately, in the case of both autoimmune disease as well as in HVG and GVH, the immune response suppression uses highly toxic drugs which are of limited effectiveness and act systemically, rather than specifically. The severe limitations of such therapy point to the need for immunosuppressive agents which have less toxicity, but greater specificity.
*An improved way to suppress an immune response to an antigen in a human would be to administer an immunosuppressively effective amount of human GIF which can specifically bind to the antigen. In so doing, the concentration of T suppressor factor is favored and, as a result, the immune response to the antigen is decreased. The present invention provides a means for accomplishing this result.
SUMMARY OF THE INVENTION The inventor has substantially purified human antigen-specific GIF which suppresses the immune response to the homologus antigen. The human antigen-specific GIF, if desired, can be used therapeutically to suppress the human immune response to the antigen.
DETAILED DESCRIPTION OF THE INVENTION The present invention relates to substantially pure human antigen-specific GIF with specificity for an antigen associated with an undesirable immune response. This human antigen-specific GIF is highly useful for the 10 immunosuppression of the undesirable immune response in an antigenspecific manner.
.:'-Preferred in the present invention are human antigen-specific GIFs which can specifically bind allergens. In an especially preferred embodiment of the invention, a human antigen-specific GIF is disclosed which binds to an S: 15 epitope on phospholipase
A
2
(PLA
2 the major allergen in honey bee venom.
This specificity enables this antigen-specific GIF, and like antigen-specific GIFs with the same specificity, to be used to suppress the human immune response to PLA 2 In another especially preferred embodiment of the invention, a human antigen-specific GIF is disclosed which binds to an epitope on Japanese cedar pollen which can be used to suppress the human immune response to this antigen.
The teaching used to produce GIF with antigen-specificity to PLA 2 can be readily extended to other antigens by those of skill in the art to prepare and purify other GIF molecules with antigenic specificity for those other antigens without undue experimentation. As a consequence, the broad pioneering nature of the invention enables the preparation of human antigen-specific 1111 'M GIFs for other allergens which can be used to suppress such immune response mediated disorders as autoimmune disease and allergy. The production of various human antigen-specific GIFs is especially facilitated where the antigen of the undesirable immune response is known such as with most allergies and various autoimmune diseases.
The human PLA,-specific GIF of the invention is obtained from, or has the identifying characteristics of, an antigen-specific GIF obtained from the cell line AC5 having ATCC accession number HB 10473. The human Japanese cedar pollen-specific GIF of the invention is obtained from, or has the identifying characteristics of, an antigen-specific GIF obtained from the cell line 31E9.
S Methods of Producing and Characterizing Hybridomas The general method used for production of hybridomas is well known (Kohler, et al., European J. Imm., 6: 292, 1976). Briefly, peripheral blood 15 mononuclear cells (PBMC) from a human allergic to honey bee venom were cultured in the presence of chemically modified PLA.. Non-adherent cells were recovered and then cultured with IL-2 and lipocortin-1 prior to fusion with lymphoblastoid cell line BUC. Hybriddmas were screened for production of human GIF specific for PLA.
20 More generally, the invention is directed to a method of producing a continuous hybridoma cell line which produces human antigen-specific GIF comprising: obtaining human antigen-primed T-cells which are activated to the antigen and cultured in the presence of IL-2 and phospholipase A 2 inhibitor; and combining the activated T-cells by fusion with a fusion partner cell line to produce hybridomas capable of producing human antigen-specific GIF.
The antigen-primed T-cells can be obtained from any sample, including the mononuclear cell fraction of peripheral blood. The antigen-primed T-cells can then be activated by culturing in the presence of the antigen to which they have been primed, followed by expanding the activated T-cells in the presence of interleukin-2 (IL-2) and a phospholipase A 2 inhibitor. An especially useful phospholipase A 2 inhibitor for such purposes is lipocortin.
Alternatively, synthetic compounds with PLA 2 inhibitory activity can be used such as 2-(p-amylcinnamoyl)-amino-4-chlorobenzoic acid, (ONO-RS-082, ONO Pharmaceutical Co.).
Under certain circumstances, such as where the primary antigen is toxic to the T-cells, it is desirable to chemically modify the antigen. Agents useful for 15 such modification include guanidine HCI and cyanogen bromide, but those *oO.
of skill in the art can easily ascertain similar agents without undue experimentation. Generally, it is preferred to use agents which do not destroy the external structure of the antigen, since it is thought that such external structures are important in suppressor T-cell epitopic recognition of the antigen. However, this issue is not significant for most antigens, such as many allergens, which are not cytotoxic. Consequently, with typical allergens, the native molecules can be used to stimulate the T-cells.
The present invention is directed to a method for generating antigen-specific human T-cells and T-cell hybridomas which produce human antigen-specific GIF, which are specifically reactive with an antigen which is associated with an immune response to be immunosuppressed.
The isolation of T-cell hybridomas producing a human antigen-specific
GIF
with the antigenic specificity of the human antigen-specific GIF of the invention can be accomplished using routine screening techniques to determine the elementary reaction pattern of the human antigen-specific
GIF
of interest. Thus, for example in the case of human GIF specific for PLA, if a human antigen-specific GIF being tested suppresses the immune response of cells from a patient allergic to PLA 2 then the human antigenspecific GIF being tested and the human GIF specific for PLA, produced by the hybridoma of the invention are equivalent.
Still another way to determine whether a human antigen-specific GIF has the specificity of a human antigen-specific GIF of the invention is to pre-incubate the human antigen-specific GIF of the invention with the antigen with which i it is normally reactive (for example, bee venom PLA 2 and determine if the human antigen-specific GIF being tested is inhibited in its ability to bind the 15 antigen. If the human antigen-specific GIF being tested is inhibited then, in all likelihood, it has the same epitopic specificity as the human antigenspecific GIF of the invention.
As used in this invention, the term "epitope" is meant to include any Bdeterminant capable of specific interaction with a human antigen-specific GIF or the monoclonal antibodies of the invention. Epitopic determinants usually consist of chemically active groupings of molecules such as amino acids or sugar side chains and usually have specific three dimensional structural characteristics, as well as specific charge characteristics.
In still another aspect, the invention relates to a method of producing substantially pure human antigen-specific GIF comprising: culturing a continuous hybridoma cell line capable of producing human antigen-specific GIF such that the cell line produces human antigen-specific GIF; and isolating substantially pure human antigen-specific GIF from the culture.
The continuous hybridoma cell lines so used can themselves be produced as described above. In addition, during culturing the hybridoma cell line is preferably stimulated to produce human antigen-specific GIF by exposing the hybridoma cells to syngeneic macrophages which have been pulsed with the antigen to which the antigen-specific GIF binds, or with antibodies to the CD3 or T-cell receptor.
Various techniques can be used to isolate or substantially purify the human antigen-specific GIF from the culture. A particularly useful technique is affinity purification using the antigen, for example attached to a solid phase, to which the antigen-specific GIF binds. A modification of this technique is to use two affinity absorption steps, if desirable, to substantially purify the human antigen-specific GIF. In such a process, the step of isolating 15 substantially pure human antigen-specific GIF includes: reacting the hybridoma cell line culture with a monoclonal Santibody specifically reactive with human GIF; (ii) eluting the human GIF from the monoclonal antibody; (iii) reacting the eluted GIF with the antigen to which the human 20 antigen-specific GIF binds; (iv) eluting the human antigen-specific GIF from the antigen; and recovering the human antigen-specific
GIF.
Alternatively, the order of the two absorption steps can be reversed, wherein the step of isolating substantially pure human antigen-specific GIF includes: reacting the hybridoma cell line culture with the antigen to which the human antigen-specific GIF binds; (ii) eluting the human GIF from the antigen;
I
(iii) reacting the eluted GIF with a monoclonal antibody specifically reactive with human GIF; (iv) eluting the human antigen-specific GIF from the monoclonal antibody; and recovering the human antigen-specific
GIF.
Purification of human antigen-specific GIF is facilitated by adjusting the hybridoma cells to serum-free culture medium, such as ABC. After treatment of the subclone with anti-CD3, followed by culture of the hybridoma cells in Protein A-coated tissue culture dishes, antigen-binding GIF in culture supernatants can be purified by ion-exchange chromatography, described above. Under such conditions, the process of isolating substantially pure human antigen-specific GIF includes: contacting the hybridoma cell line culture supernatant with an anionic exchange matrix; 15 (ii) eluting the human GIF from the matrix; (iii) reacting the eluted GIF with a monoclonal antibody specifically reactive with human GIF or with the antigen to which the human antigen-specific GIF binds, or both; (iv) eluting the human antigen-specific GIF; and 20 recovering the human antigen-specific GIF.
°Thus, ion-exchange chromatographic purification is used in combination with affinity-purification to isolate human antigen-specific GIF, for example, by using the antigen to which the antigen-specific GIF binds or by using an antibody specifically reactive with human GIF, or both, as described above.
In the preferred embodiment, DEAE (diethylaminoethyl) Sepharose is the matrix utilized for purification of antigen-specific human GIF. Other ionexchange materials which can be utilized include virtually any of the commercially available anion exchange agaroses and celluloses, such as polysulfated agaroses, specifically including but not limited to QAE (quaternary amine) derivatives, ecteola (epichlorohydrintri-ethanolamine), TEAE (triethylaminoethyl) derivatives, and AE (aminoethyl) cellulose. The specific parameters for binding and eluting from these various ion-exchange materials can be known to those of skill in the art, or can be readily ascertained, without undue experimentation.
When the hybridoma cell line culture supernatant is added to the anionexchange matrix equilibrated with about 20mM salt, NaCI, much of the GIF will pass through the column and the remainder are eluted with salt concentrations up to about 60mM. Preferred for elution from DEAE are concentrations of NaCI from about 20mM to about 60mM contained in Tris.
A monoclonal antibody which is particularly useful in the affinity purification of human GIF is the monoclonal antibody produced by a cell line 388F, or 15 monoclonal antibodies having the specificity of a monoclonal antibody produced by cell line 388F,.
THERAPEUTIC USES OF HUMAN ANTIGEN-SPECIFIC GIF The term "suppressive" denotes a lessening of the detrimental effect of the undesirable immune response in the human receiving therapy. The term "immunosuppressively effective" means that the amount of human antigenspecific GIF used is of sufficient quantity to suppress the cause of disease or symptoms due to the undesirable immune response.
*o The dosage ranges for the administration of the human antigen-specific GIF of the invention are those large enough to produce the desired effect in which the symptoms of the immune response show some degree of suppression. The dosage should not be so large as to cause adverse side effects, such as unwanted cross-reactions, anaphylactic reactions, and the like. Generally, the dosage will vary with the age, condition, sex and extent of the disease in the patient and can be determined by one of skill in the art. The dosage can be adjusted by the individual physician in the event of any counterindications. Dosage can vary from about 0.001 mg/kg/dose to about 2 mg/kg/dose, preferably about 0.001 mg/kg/dose to about 0.2 mg/kg/dose, in one or more dose administrations daily, for one or several days.
The human antigen-specific GIF of the invention can be administered parenterally by injection or by gradual perfusion over time. The human antigen-specific GIF of the invention can be administered intravenously, intraperitoneally, intramuscularly, subcutaneously, intracavity, or transdermally.
Preparations for parenteral administration include sterile aqueous or nonaqueous solutions, suspensions, and emulsions. Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. Aqueous carriers include water, alcoholic/aqueous solutions, emulsions or suspensions, including saline and buffered media. Parenteral vehicles include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's, or fixed oils. Intravenous vehicles include fluid and nutrient replenishers, electrolyte replenishers (such as those based on Ringer's dextrose), and the like. Preservatives and other additives may also be present such as, for example, antimicrobials, anti-oxidants, chelating 25 agents, and inert gases and the like.
The invention also relates to a method for preparing a medicament or pharmaceutical composition comprising the human antigen-specific GIF of the invention, the medicament being used for therapy of an undesirable immune response to an antigen wherein the antigen is capable of binding by the human antigen-specific GIF of the invention.
The present invention is also directed to monoclonal antibodies, and B-cell hybridomas which produce them, which are specifically reactive with human
GIF.
As stated above, techniques for production of hybridomas are well known to those of skill in the art. In brief, the B-cell hybridomas of the invention were prepared by immunizing BALB/c mice with affinity-purified human GIF and later boosted. Two weeks after the last immunization, spleen cells were obtained from the animals and transferred to syngeneic BALB/c mice which had been lethally irradiated. The syngeneic recipients were immunized twice with purified human GIF and 2 weeks after the last immunization the spleen cells were fused with SP 2/0-14AG myeloma cell line. Hybridomas were screened for monoclonal antibody production to human GIF.
15 The isolation of hybridomas producing monoclonal antibodies with the reactivity of the monoclonal antibodies of the invention can be accomplished using routine screening techniques to determine the elementary reaction pattern of the monoclonal antibody of interest. Thus, if a monoclonal antibody being tested reacts with human GIF, but does not react with 20 mouse GIF, then the antibody being tested and the antibody produced by the hybridomas of the invention are equivalent.
The isolation of other hybridomas producing monoclonal antibodies with the specificity of monoclonal antibody 388F 1 or any other monoclonal antibody of the invention, can be accomplished by one of ordinary skill in the art by producing anti-idiotypic antibodies (Herlyn, et al.,Science, 232: 100, 1986).
An anti-idiotypic antibody is an antibody which recognizes unique determinants present on the monoclonal antibody produced by the hybridoma of interest. These determinants are located in the hypervariable region of the antibody. It is this region which binds to a given epitope and, thus, is responsible for the specificity of the antibody. The anti-idiotypic antibody can be prepared by immunizing an animal with the monoclonal antibody of interest. The animal immunized will recognize and respond to the idiotypic determinants of the immunizing antibody by producing an antibody to these idiotypic determinants. By using the anti-idiotypic antibodies produced by the second animal, which are specific for the monoclonal antibodies produced by a single hybridoma which was used to immunize the second animal, it is now possible to identify other clones with the same idiotype as the antibody of the hybridoma used for immunization and thereby greatly simplify and reduce the amount of screening needed to find other hybridomas producing monoclonal antibodies with the specificity of the monoclonal antibodies of the invention.
o 15 Idiotypic identity between monoclonal antibodies of two hybridomas demonstrates that the two monoclonal antibodies are the same with respect ::to their recognition of the same epitopic determinant. Thus, by using antibodies to the epitopic determinants on a monoclonal antibody it is possible to identify other hybridomas expressing monoclonal antibodies of the same epitopic specificity.
Alternatively, it is possible to evaluate, without undue experimentation, a monoclonal antibody to determine whether it has the same specificity as 388F 1 of the invention by determining whether the monoclonal antibody being tested prevents 388F 1 from binding to a particular antigen, for 25 example human GIF, with which 388F, is normally reactive. If the monoclonal antibody being tested competes with 388F,, as shown by a decrease in binding by 388F,, then it is likely that the two monoclonal antibodies bind to the same epitope.
Still another way to determine whether a monoclonal antibody has the specificity of 388F, is to pre-incubate 388F, with an antigen with which it is normally reactive, for example, human GIF, and determine if the monoclonal antibody being tested is inhibited in its ability to find the antigen. If the monoclonal antibody being tested is inhibited then, in all likelihood, it has the same epitopic specificity as the monoclonal antibody of the invention.
Under certain circumstances, monoclonal antibodies of one isotype might be more preferable than those of another in terms of their diagnostic or therapeutic efficacy. Particular isotypes of a monoclonal antibody can be prepared either directly, by selecting from the initial fusion, or prepared secondarily, from a parental hybridoma producing monoclonal antibody of different isotype, by using the sib selection technique to isolate class-switch variants (Steplewski, et al., Proceedings of National Academy of Sciences, USA, 82; 888653, 1985; Spira, et al., Journal of Immunological Methods, 74: 15 307, 1984). Thus, the monoclonal antibodies of the invention would include class-switch variants having the specificity of monoclonal antibody 388F 1 which is produced by ATCC HB 10472. This cell line was placed on deposit for 30 years at the American Type Culture Collection (ATCC) in Rockville, Maryland prior to June 4, 1990.
S S The term "antibody" as used in this invention is meant to include intact S: molecules as well as fragments thereof, such as, for example, Fab and F(ab,) 2 which are capable of binding the epitopic determinant.
The monoclonal antibodies of the invention can also be used in immunoaffinity chromatography for the purification of the various types of human GIF mentioned herein. One way by which such immunoaffinity chromatography can be utilized is through the use of, for example, the binding of the monoclonal antibodies of the invention to CNBr-Sepharose-4B or Tresyl-activated Sepharose (Pharmacia). These solid phase-bound monoclonal antibodies can then be used to specifically bind human GIF from mixtures of other proteins to enable its isolation and purification. The bound IFN-gamma can be eluted from the affinity chromatographic material using techniques known to those of ordinary skill in the art such as, for example, chaotropic agents, low pH, or urea.
The above disclosure generally describes the present invention. A more complete understanding can be obtained by reference to the following specific examples which are provided herein for purposes of illustration only, and are not intended to limit the scope of the invention.
EXAMPLE 1 PREPARATION-OF HYBRIDOMA CELL LINES PRODUCING HUMAN 1 ANTIGEN-SPECIFIC GLYCOSYLATION INHIBITING FACTOR (GIF) AND PURIFICATION TECHNIQUES A. ANTIGENS 15 Lyophilized phospholipase A 2
(PLA
2 from bee venom was purchased from Sigma Chemical Co., St. Louis, MO. Denatured PLA, (-PLA 2 and cyanogen bromide-treated PLA, were prepared by the method described by King, et al., Arch. Biochem and Biophys., 172: 661, 1976. For the preparation of D-
PLA
2 5 mg of PLA. were dissolved in 0.1 M Tris HCI buffer, pH 8.6, and denatured in 6M guanidine HCI in the presence of 5 mg/ml dithiothreitol.
After 18 hours at room temperature, sulfhydryl groups were carboxymethylated with iodoacetic acid. The denatured protein was dialyzed against 0.02 M acetic acid and kept at -40 C until use. For the cleavage of methionine bonds in PLA., 10 mg bee venom PLA, was dissolved in 0.4 ml distilled water, and 1.2 ml formic acid containing 100 mg CNBr were added.
After 2 hours at room temperature, the mixture was diluted two-fold with and lyophilized in Speed Vac. Native PLA, was coupled to Tresyl activated Sepharose (Pharmacia) following the procedures recommended by the manufacturer. Unless otherwise stated, 1 mg protein was coupled to 1 ml Sepharose.
B. ANTIBODIES Purified human E myeloma protein PS, monoclonal mouse IgE from the hybridoma H-1 DNP-E-26 (Liu, et al., J. Immunol., 124: 2728, 1980) and monoclonal anti-CD3 (OKT 3) were the same preparations as those described in a previous article (Carini, et al., J. Immunol. Methods, 127: 221, 1990). Ascitic fluid containing the monoclonal anti-T-cell receptor ap, WT 31C, (Spits, et al., J. Immunol., 135: 1922, 1985) was kindly supplied by Dr.
J. DeVries, DNAX Institute of Molecular and Cellular Biology, Palo Alto, CA.
The mouse monoclonal antibody against rabbit lipomodulin 141B9 (Iwata, et al., J. Immunol., 132: 1286, 1984) was the same preparation as that 15 described in a previous article (Askasaki, et al., J. Immunol., 131: 3172, 1986). Specifically-purified goat antibodies against mouse IgG, which contained both anti-heavy chain and anti-light chain, were previously described (Suemura, et al. J. Immunol., 125: 148, 1980). Fluoresceinated goat anti-mouse IgG antibodies were purchased from Cappel. Human IgE 20 and anti-lipomodulin antibody (141B9) were coupled to CL-Sepharose 4B; approximately 5 mg of protein were coupled to 1 ml Sepharose.
C. CELL LINES RPMI 8866 lymphoblastoid cells were cultured in RPMI 1640 medium enriched with 10% fetal calf serum, 2 mM L-glutamine, 50/M 2mercaptoethanol and antibiotics (RPMI 1640 culture medium). The mouse T-cell hybridoma 12H5 cells (Iwata, et al., J. Immunol., 140: 2534, 1988) were maintained in high glucose Dulbecco's modified Eagle's medium (DMEM) described in a previous article (Huff, et al., Proc. Natl. Acad. Sci., USA, 129: 509, 1982). A hypoxanthine guanine phosphoribosyltransferasedeficient mutant of the human lymphoblastoid cell line CEM (BUC) cells were previously described (Huff Ishizaka, Proc. Natl. Acad. Sci., USA, 81: 1514, 1984).
D. CELL CULTURE AND CONSTRUCTION OF HYBRIDOMAS Peripheral blood was obtained from a patient allergic to honey bee venom, and mononuclear cells of the blood (PBMC) were obtained by centrifugation on Ficoll-Pague (Pharmacia). To activate antigen-primed T-cells, PBMC were suspended in RPMI 1640 culture medium at the concentration of 3x106 nucleated cells/ml, and cultured for three days in the presence of
D-PLA
2 or CNBr-treated
PLA
2 Non-adherent cells were recovered, resuspended in fresh culture medium (2x10 5 cells/mi), and then cultured for four days with 60 units/ml purified IL-2 (chromatographically purified human 15 IL-2, Electro-nucleonics, Silverspring, MD), in the presence of 3 pg/ml recombinant human lipocortin 1, which was kindly supplied by Drs. J.
Browning and B. Pepinsky, Biogen.
To construct T-cell hybridomas, 1.2x107 T-cells, which had been propagated by IL-2, were mixed with twice the number of BUC cells. Mixed cells were pelleted together and fused by using polyethylene glycol (1300 1600 MW., e Sigma). Detailed procedures for cell fusion were as previously described (Huff, et al., Proc. Nat'l. Acad. Sci., U.S.A. 81: 1514, 1984). Cells were resuspended in hypoxanthine/aminopterin/thymidine (HAT)-containing DMEM, and 5x104 cells were seeded.in each well of 96 well plates. Hybrid S* 25 clones were maintained in complete DMEM with biweekly subcultures. In order to stimulate the T-cell hybridomas, the cells were treated with 8 pg/ml OKT 3 for 40 minutes at 0C, and the antibody-treated cells (1x10 6 /ml) were seeded in Limbro tissue culture wells (Flow Labs, McLean, VA) which had been coated with 10 pg/ml anti-MGG. Culture supernatants were obtained after 24 hour culture.
E. DETECTION OF CD3 AND TcR The hybridoma cells (1xl10/sample) were incubated 40 minutes at 0*C with 8 pg/ml OKT 3 or a 1:1000 dilution of anti-TcRaB(WT31 )-containing ascitic fluid in RPMI 1640 medium supplemented with 5% FCS and 10 pM NaN3.
As controls, aliquots of the same cells were treated with the same concentration of mouse IgG2. (Becton-Dickinson, isotype control). Cells were washed twice with PBS containing 5% FCS, and then incubated with fluoresceinated anti-mouse IgG for 40 minutes. After washings, cellassociated fluorescence was analyzed by using FACScan from Becton- Dickinson.
The CD3 hybridoma cells were identified by rosetting using ox erythrocytes coated with anti-mouse IgG. The antibodies were coupled to erythrocytes So. 15 by the method of Wilhelm, et al., Immunol. Methods, 90: 89, 1986).
Briefly, 0.5 ml of packed ox erythrocytes were washed 4 times with saline, and resuspended in 0.75 ml of 0.5pg/ml purified anti-MGG; 25 pl of CrCI, (16.5 mg CrCI, dissolved in 5 ml saline) were added to the cell suspension under gentle mixing, and the cell suspension was incubated for 1 hour at 30" C. The anti-MGG-coupled erythrocytes were washed 4 times with saline and resuspended in 5 ml FCS (approximately 1x10 9 erythrocytes/ml). To detect the CD3' cells, pellets of 106 hybridoma cells were suspended in pl of DPBS containing 5% FCS and 8pg/ml OKT 3. After 45 minutes at 0* C, the cells were washed twice and resuspended in 80 p1 DPBS-5% FCS, and 25 20 pl of a suspension of anti-MGG-coated erythrocytes and crystal violet were added to the cell suspension. The mixtures were centrifuged at 200g for 5 minutes and tubes were incubated for 2 hours at 0°C. The pellets were gently resuspended and examined for rosetting cells under microscope.
F. ENRICHMENT OF CD3' CELLS Hybridoma cells treated with 8 pg/ml OKT 3 (1.5x106 cells) were mixed with anti-MGG coupled erythrocytes (ca 4x108 erythrocytes) to form rosettes by the procedures described above. Pellets were resuspended and applied to the top of Percoll gradient consisting of 60% and 50% Percoll layers. Tubes were centrifuged for 20 minutes at 1200 RPM (700g) at room temperature.
The pelleted cells were washed twice with culture medium, and the erythrocytes were lysed by treatment with 0.83% NH 4 CI buffer for 1 minute at '0C. The cells were washed with and resuspended in DME culture medium and cultured to expand the cell population.
Further enrichment of CD3' cells was carried out by cell sorting. Hybridoma cells were treated with OKT 3 and stained with fluoresceinated anti-MGG.
The positively stained cells were selected by sorting the cells by using FACSTAR (Becton-Dickinson).
G. PURIFICATION AND DETECTION OF IgE-BF Culture supernatant of T-cell hybridomas were filtered through Diaflo YM 100 membranes (Amicon Corp., Lexington, MA) and the filtrates were 20 concentrated ten-fold by ultrafiltration through YM5 membranes. IgE-BF in the filtrates were purified by using IgE-coupled Seoharose by the described procedures (Ishizaka Sandberg, J. Immunol., 126: 1692, 1981). The presence of IgE-BF in culture filtrates or acid eluate fraction from IgE- Sepharose was assessed by inhibition of rosette formation of FceR'B iymphoblastoid cell line, RPMI 8866 cells with human IgE-coated ox erythrocytes (E-lgE) by the procedures previously described (Kisaki, et al., J. Immunol., 138: 3345, 1987). The proportion of rosette forming cells (RFC) in 300 RPMI 8866 cells was determined in triplicate and was expressed as the average -SD.
Rodent IgE-BF formed by the 12H5 cells were detected by the same procedure, except that rat IgE-coated ox erythrocytes were employed as indicator cells, and mesenteric lymph node cells of Lewis strain rate infected with Nipportronngylus brasiliensis were used as a source of FceR'B cells (Yodoi Ishizaka, J. Immunol., 124: 1322, 1980).
H. DETECTION OF GIF GIF was detected by using T-cell hybridoma 12H5 cells (Iwata, et al., J.
Immunol., 140: 2534, 1988). A suspension of the hybridoma calls was mixed with an equal volume of a test sample, and the cell suspensions were cultured for 24 hours with 10 pg/ml mouse IgE. Culture supernatants were filtered through CF50A membranes, and filtrates containing IgE-BF were 15 fractionated on lentil lectin Sepharose (Yodoi, et al., J. Immunol., 125: 1436, 1980). Both unbound proteins (effluent fraction) and those eluted with 0.2 M a methylmannoside (eluate fraction) were assessed for the presence of IgE-BF by rosette inhibition technique. When the 12H5 cells were cultured with mouse IgE alone, essentially all IgE-BF formed by the cells bound to lentil lectin Sepharose and were recovered by elution with a methylmannoside. Thus, the ratio of the percent rosette inhibition between the effluent/eluate fraction is less than 0.2. If a sufficient amount of GIF were added to the culture of 12H5 cells together with mouse IgE, the majority of IgE-BF formed by the cells lacked affinity for lentil lectin and were recovered 25 in the effluent fraction (Iwata Ishizaka, J. Immunol., 141: 3270, 1988).
Thus, GIF was taken as if the ratio oi the percent rosette inhibition between the effluent/eluate fraction were 3.0 or higher.
1. FRACTIONATION OF GIF In order to determine whether GIF from hybridomas has affinity for bee venom PLA 2 culture filtrates of hybridoma cells were fractionated on antigencoupled Sepharose. Hybridoma cells were treated with OKT 3 antibody (8 pg/ml) and 8 ml aliquots of the antibody treated or untreated cell suspension (1.5x10 6 cells/mi) were cultured in anti-MGG-coated tissue culture flasks.
Culture supernatants were concentrated four-fold, and a 2 ml sample was absorbed with 0.4 ml IgE-Sepharose. The effluent fraction was mixed with ml PLA,-Sepharose overnight, and immunosorbent was packed into a small column. After effluent fraction was recovered, the column was washed with DPBS, and then eluted with 1.0 ml glycine HCI buffer, pH 3.0. Partial purification of GIF on anti-lipomodulin (141B9) Sepharose was carried out by the procedures previously described (Akasaki, et al., J. Immunol., 136: 3172, 1987).
15 J. DETERMINATION OF PHOSPHOLIPASE INHIBITORY ACTIVITY Affinity-purified GIF was treated with alkaline phosphatase as previously described (Uede, et al, J. Immunol., 139: 898, 1983). Briefly 1 ml of the preparation was dialyzed against Tris-HCI buffer, pH 8.2 and was mixed with unit of insoluble alkaline phosphatase (calf intestinal, Sigma) for 2 hours at room temperature. After centrifugation, the supernatant was dialyzed against 0.1 M Tris-HCI buffer, pH 8.0. Phospholipase A 2 inhibitory activity of the alkaline-phosphatase treated samples was determined using E coli which were biosynthetically labeled with 3 H-oleic acid and porcine pancreatic
PLA
2 (Sigma) (Rothut, et al., Biochem. Biophys. Res. Commun., 117: 878, 25 1983). Detailed procedures were described in Ohno, et al. (Internat.
Immunol., 1: 425, 1989). Briefly, porcine pancreatic PLA 2 (1x10 5 units) was mixed with GIF in a total volume of 150 pl. After 5 minutes at 25 C, 50 pl cf a suspension of 3 H-labeled E coili (5000 cpm) was added, and the I 1 3 mixtures were incubated for 5 minutes at 25 C. The reaction was stopped by the addition of 50 l1 2 M HCI, and 50 pl of 100 mg/ml BSA was added to the mixtures. The suspensions were centrifuged for 1 minute at 5500 g, and radioactivity in 250 pl of supernatant was measured in a scintillation spectrometer.
K. ION EXCHANGE COLUMN CHROMATOGRAPHY Culture supernatant of AC5 cells in serum-free medium was concentrated to 100 fold by ultrafiltration. After centrifugation at 10,000 rpm for 20 min, the supernatant was diluted 8-fold with distilled water, adjusted to pH with Tris, and immediately applied to a DEAE-Sepharose CL-6B (Pharmacia) column (3 ml volume) which was equilibrated with 10mM Tris HCI buffer, pH After effluent (passed-through) fraction was recovered, the column was washed with 4 column volumes of 10 mM Tris-HCl buffer containing NaCI, and the washing was combined to the passed through fraction.
15 Proteins bound to the column were eluted successively with 4 column volumes of 10 mM Tris HCI buffer, pH 8.0 containing 50mM, 75mM, 100mM, 150mM, and 200 mM NaCl. Each eluate fraction was concentrated and dialyzed against Dulbecco's phosphate buffered saline (DPBS).
L. GEL FILTRATION 20 One ml sample in DBPS was applied to a Superose 12 column (1.6 x cm, Pharmacia), connected to HPLC (Beckman, System Gold). Proteins were eluted from the column with DPBS at a flow rate of 1 ml/min, and appropriate fractions were collected. The column was calibrated with human IgE (PS protein, MW: 185,000), bovine serum albumin (BSA, MW: 67,000), ovalbumin (MW: 43,000), soybean trypsin inhibitor (MW: 20,100), and cytochome C (MW: 12,500). All standard proteins except IgE were obtained from Sigma. Retention time for the standard proteins were 41.97, 52.08.
55.135, 62.097, and 71.67 min, respectively.
M. AFFINITY-PURIFICATION OF GIF Culture supernatant of CL3 clone in complete DME medium was concentrated 5-fold by ultrafiltration, and GIF in the supernatant was absorbed to 141B9-Sepharose or anti-GIF Sepharose by recycling the supernatant overnight through the immunosorbent column (5 ml volume) (Iwata, et al., J.Immunol., 141:3270, 1988). The immunosorbent was washed with 20 column volumes of DPBS, and proteins bound to the beads were recovered by elution with 0.1 M glycine HCI buffer, pH 3.0. Murine GIF from the 231F1 cells was purified by the same technique using the 141B9- Sepharose.
In order to isolate GIF in culture supernatant of AC5 cells in protein-free medium, the supernatant was concentrated 50 to 100-fold by ultrafiltration.
15 An appropriate fraction of the supernatant from a DEAE-Sepharose column was concentrated to 5-6 ml and mixed overnight at 4'C with 1.0 to 1.5 ml of Affigel 10-immunosorbent coupled with monoclonal anti-GIF antibody.
The suspension was packed into a small column and the immunosorbent was washed with 40 column volumes of DPBS. In some experiments, the immunosorbent was washed with 40 column volumes of DPBS and column volumes of PBS containing 0.5 M NaCI. Proteins bound to the mmunosorbent were eluted with 0.05 M glycine HCI buffer containing 0.15 M NaCI, pH 3.0-3.2.
N. DETECTION OF GIF BY SDS-PAGE Affinity-purified GIF was dialyzed against 0.01% SDS in deionized water, and lyophilized in a Speed vac (Savant Instruments, Hicksville, NY). Samples were then analyzed by SDS gel electrophoresis in 15% poiyacrylamide slab gel by using the Laemmli system (Laemmli, Nature, 227:680, 1970).
Gels were fixed and protein bands were detected by silver staining (Ochs, et al., Electrophoresis, 2:304, 1981). Molecular weight standards were obtained from Pharmacia.
O. ELISA ASSAYS In order to detect monoclonal anti-GIF antibody, the method described by Steele, et al. (J.Immunol., 142:2213, 1989) was employed with slight modifications. Briefly, Immulon I plates (Dynatech) were coated overnight with 1001p of affinity-purified GIF diluted with 0.1M carbonate coating buffer, pH 9.6. Plates were washed 3 times with phosphate buffered saline (PBS) containing 0.05% Tween 20 between each of the following steps. Plates were blocked with 2% BSA in PBS for 6-9 hours. One hundred microliters of each test sample was then added to the well, and plates were kept 15 overnight at 4 C. Binding of mouse Ig to the plate was detected by using alkaline phosphatase-coupled goat anti-mouse Ig (Zymed Lab, So. San Francisco, CA) and alkaline phosphatase substrate (Sigma). ELISA signal was read in a microplate reader MR 5000 (Dynatech Lab) with a 410 nm filter 30 min after the addition of substrate. Isotype of monoclonal antibodies 20 was determined with ELISA assay by using an isotyping kit for mouse mAb (Zymed Lab).
For the detection of GIF in fractions of an affinity-purified GIF preparation, a biotin-avidin system and amplification method (Stanley, et al., J.Immunol.
Methods, 83:89, 1985) were employed to increase the sensitivity. Maxi-Sorp 25 microtiter plates (Nunc, Copenhagen, Denmark) were coated with 50 p1 of each fraction. After incubation for 2 hours at 37 C, plates were washed with Tween/PBS and blocked with 2% BSA overnight at 4 C. After washing, of biotin-coupled mAb 141-B9 (200 ng/ml) were added to each well and the plate was incubated for 2 hours at 37'C. The plate was washed and of a 1:1500 dilution of streptavidin-alkaline phosphatase conjugate (Zymed Lab) were added to each well. After incubation for 1 hour at 37 C, quantities of alkaline phosphatase bound to the wells were measured by amplification system (Stanley, et al., J.Immunol. Methods, 83:89, 1985), (GIBCO-BRL, Bethesda, MD). ELISA signal was determined at 490 nm.
EXAMPLE 2 CHARACTERIZATION OF HYBRIDOMAS PRODUCING HUMAN ANTIGEN-SPECIFIC
GIF
As described above, PBMC of a bee venom-sensitive patient were cultured for three days in the presence of 10 pg/ml D-PLA 2 and activated T-cells were propagated by IL-2 for four days in the presence of 3 ul/ml recombinant lipocortin. T-cells were then fused with BUC cells to construct hybridomas. In this experiment, 4 hybridoma clones were obtained. Each 15 hybridoma clone was cultured in complete DMEM and culture supernatants were filtered through YM100 membranes. Filtrates were concentrated ten-fold and assessed for the presence of GIF by using the 12H5 cells. The results shown in Table I indicate that two of the four hybridoma clones constitutively secrete GIF.
TABLE I Selection of GIF-Producing Hybridomas 3 H-Oleic Acid Releasee GIF Activity" Release Inhibition Hybridoma Effluent/Eluate (cpm) Cl 1 0/26 ND CI 2 2/33 390±27 4 Cl 3 29/0 257±25 37 Cl 7 27/5 303±17 26 Control 0 3 1 b 408±15 S. Culture filtrates of each clone were concentrated ten-fold. One volume of the filtrate was added to an equal volume of a suspension of the 12H5 cells, and the cells were cultured for 24 hours in the presence of 10 pg/ml mouse IgE. Numbers in the column represent the percent of rosette 15 inhibition by the effluent/eluate fractions from lentil lectin Sepharose. The proportion of IgE-RFC in the absence of IgE-BF was 24.4 0.3 b The 12H5 cells were cultured with 10 pg/ml mouse IgE alone, and IgE-BF in culture filtrates were fractionated on lentil lectin Sepharose.
Culture filtrates were fractionated on 141B9-Sepharose, and acid eluates from the immunosorbent were concentrated to 1/100 volume of the original culture supernatant. The samples were treated with alkaline phosphatase, and dephosphorylated materials were assessed for the ability to inhibit pancreatic phospholipase A2.
The presence of CD3 determinants on the hybridoma clone CL3 was assessed by fluorocytometry and the rosetting technique. The cells were treated with 8 pg/ml monoclonal antibody OKT3 and then stained with fluoresceinated goat anti-mouse Ig. Less than 10% of the total cells were stained. It was also found that only 6-8% of the OKT3-treated cells formed rosettes with anti-MGG-coupled erythrocytes. As a consequence, the CD3' cells were enriched using the rosetting procedures described in Example 1.
Cells which formed rosettes with anti-MGG coupled erythrocytes were separated from non-rosetting cells by density gradient centrifugation on Percoll layers and were expanded by culture in complete DMEM. The same procedures were repeated three times to enrich the CD3' cell population.
Treatment of the final cell preparation with OKT3 antibody followed by incubation with anti-MGG-coated erythrocytes showed that 80-90% of the cell population formed rosettes. Approximately 75% of the cells were stained by OKT3 in cytofluorometry. However, when culture of the cells for 2 weeks :with four passages resulted in the decline of CD3' cells to approximately 52% (as determined by cytofluorometry), the CD3' cell population was further enriched by cell sorting and expanding the cells by culture. After repeating the cell sorting twice, a CL3 population was obtained which stably expressed CD3. Fluorescent staining of the population with OKT3 and WT31 (anti-TcRap) indicated that essentially 100% of the cells expressed CD3 and the majority of the cells expressed TcRoa. The CD3' cell population and CD3' population were cultured and culture filtrates were assessed for the presence of GIF by using the 12H5 cells. The GIF activity 25 was detected in culture filtrates of CD3" cells, but not in the culture filtrates of CD3- population. The results indicated that the source of GIF is CD3cells.
Since one of the unique properties of mouse GIF is that the monoclonal anti-lipomodulin (141B9) binds the lymphokine, it was decided to determine whether human GIF from the CL3 cells would be absored with 141B9coupled Sepharose. The CD3-, CL3 clone was cultured to yield 1 liter of culture supernatant. After filtration through a YM100 membrane, the filtrates were concentrated to 5 ml, and fractionated on 1 ml 141-89 Sepharose.
After recovering the effluent fraction, the immunosorbent was washed with 10 column volumes of DPBS, and then eluted with 5 column volumes of glycine-HCI buffer, pH 3.0. After dialysis against DPBS, distribution of GIF activity in the fractions was determined by using the 12H5 cells. The results shown in Table II indicate that essentially alr GIF activity in the culture filtrate bound to 141-B9 Sepharose and was recovered by elution at acid pH.
V
V.
V*
V
V
V
TABLE II Human GIF From CL3 Clone Purified By Affinity Chromatography On Anti-Lipomodulin Sepharose a Fraction from 141B9- Sepharoseb Effluent Washing Eluate Dilution 1:10 1:10 1:10 1:40 1:80 GIF Activityc Effluent/Eluate 0/31 0/35 42/0 45/0 39/0 0/34 Media Control
S
S *c S.
S 5c S S S. Culture supernatants of the CL3 clone were filtered through YM100 membranes, and filtrates were concentrated 200-fold. 5 ml of the concentrated filtrate was fractionated on 1 ml 141B9-Sepharose.
b After recovering the effluent fraction, the immunosorbent was washed with column volumes of DPBS, and then eluted with 5 column volumes of glycine HCI buffer, pH GIF activity was assessed by using the 12H5 cells by the same procedures described in Table 1. Numbers in the column represent percent rosette inhibition by the effluent/eluate fractions from lentil lectin Sepharose. The proportion of IgE-RFC in the absence of IgE-BF was 22.9 0.6 in this assay. indicated the presence of GIF.
*5 5 Se
S
Previous experiments provided evidence that murine GIF is a phosphorylated derivative of a phospholipase inhibitory protein (Uede, et al., J. Immunol., 139: 898, 1983). Thus, GIF in the culture filtrates of CL3 clone was purified by using the 141B9-Sepharose. Culture filtrate of the three other clones, CL1, CL2, and CL7 were fractionated on the 141B9- Sepharose in a similar manner. The acid eluates from the immunosorbent were treated with alkaline phosphatase, and assessed for the ability to inhibit the release of 3 H-oleic acid from biosynthetically labeled E. coli by pancreatic phospholipase A 2 (Rothut, et al., Biochem. Biophys. Res.
Commun., 117: 878, 1983). The results included in Table I indicate that the affinity-purified GIF from CL3 and CL7 exerted phospholipase inhibiting activity, while the same fraction from CL1 and CL2 failed to inhibit phospholipase A 2 EXAMPLE 3 15 ANTIGEN-BINDING PROPERTIES OF GIF Previous experiments have shown that antigen-activated T-cells propagated with IL-2 in the presence of lipocortin constitutively released GIF that had no affinity for bee venom PLA, but cross-linking of CD3 on the same cells ~resulted in the formation of GIF having affinity for the antigen-coupled Sepharose together with IgE-BF. In view of these findings, it was decided to determine whether the CL3 clone produces antigen-binding GIF and IgE-BF. The cells were treated with OKT3 at 0° C, and the antibody-treated cells (1.5x106 cells/ml) were cultured in the anti-MGG-coated cells. As a control, untreated CL3 cells were cultured in the anti-MGG-coated wells.
25 Culture supernatants were filtered through YM100 membranes and concentrated seven-fold by ultra-filtration. The concentrated culture filtrates were absorbed overnight with 1 ml IgE-Sepharose, and unbound protein fraction and 2 ml of washings were combined. The IgE-Sepharose was thoroughly washed, and eluted with glycine HCI buffer. The eluate fractions from IgE-Sepharose were assessed for the presence of IgE-BF by using RPMI 8866 cells as the source of FcER- cells.
TABLE III Failure Of The GIF From The CL3 Clone To Bind To Bee Venom PLA 2 GIF Activity In PLA2-Sepharosec Treatmenta IgE-BFb Eluate Washing Eluate OKT 3 23 21/0 0/24(-) None 0 22/13(±) 0/26(-) 0 Untreated or CD3-treated cells were cultured in anti-MGG-coated wells.
0* b 30 ml culture supernatant were filtered through YM100, and filtrates were concentrated to 4 ml. The samples were absorbed with 1.0 ml IgE- Sepharose. Acid eluate fraction was adjusted to 4.0 ml and assessed for IgE-BF by rosette inhibition. The proportion of IgE-BF in the absence of IgE-BF was 37.7 1.0 mi of the effluent fraction from IgE-Sepharose was fractionated on PLA-Sepharose. The effluent, washing and acid eluate fractions were adjusted to 1.3 ml, and were assessed for GIF activity by using the 12H5 cells. Numbers in the column represent percent rosette inhibition by the effluent/eluate fractions from lentil lectin Sepharose. The proportion of IgE-RFC in the absence of IgE-BF was 21.7 0.6 The results shown in Table 11I indicate that anti-CD3-treated cells formed IgE-BF, while untreated cells failed to produce a detectable amount of IgE- BF. The effluent fraction from IgE-Sepharose was concentrated two-fold and 1 ml samples were fractionated on 0.25 ml PLA 2 -Sepharose. The effluent fraction, washing, and eluate fraction were adjusted to 1.5 ml. and the samples assessed for GIF activity. As shown in Table III, GIF from both unstimulated and anti-CD3 treated cells failed to bind to PLA 2 Sepharose.
It was thought that the failure of the GIF from anti-CD3 treated CL3 cells to bind PLA might be related to the use of D-PLA, for the activation of Tcells. In order to investigate this possibility, more T-cell hybridomas from PBMC of a bee venom sensitive patient were constructed. The protocol for the construction of T-cell hybridomas was exactly the same as that S: described above, except that PBMC were stimulated with 10 pg/ml CNBr- 15 treated PLA, instead of D-PLA 2 As the results of this experiment, 22 hybridoma clones were obtained. The GIF assay of culture filtrates of each clone indicated that 10 out of 22 clones constitutively formed GIF (results not shown). Seven GIF-secreting clones were treated with OKT3 and the antibody-treated cells were cultured in anti-MGG-coated dishes. Culture S" 20 filtrates were concentrated four-fold and absorbed with IgE-Sepharose.
TABLE IV Formation Of Antigen-Binding GIF By Anti-CD3-Treated Hybridoma Cellsa GIF Activity in PLA,-Seharose Clone laE-BF° Effluent Eluate 20 0/21 31/0 36 19/0 0/21 BA6 8 29/0 0/24 BE12 65 0/31 25/0 65 0/27 20/0 CB7 64 0/28 17/0 CE5 58 0/28 35/0 1.2x10' cells were treated with OKT 3. Cells were resuspended in 8 ml 15 culture medium and seeded in an anti-MGG-coated flask. Culture supernatant were concentrated four-fold and absorbed with IgE- Sepharose. Effluents from IgE-Sepharose were then fractionated on
PLA
2 -Sepharose and GIF activity in the effluent and eluate fraction was determined.
o Acid eluate fractions from IgE-Sepharose were assessed for the presence of IgE-BF. The proportion of IgE-RFC in the absence of IgE-BG was 26.3 0.6 c GIF activity was determined by using the 12H5 cells. Numbers represent the percent rosette inhibition by the effluentieluate fractions from lentil lectin Sepharose. Proportion of IgE-RFC in the absence of IgE-BF was 26.0 0.7 indicates the presence of GIF.
As shown in Table iV, acid eluate fraction from IgE-Sepharose of 6 out of 7 clones contained detachable amounts of IgE-BF. The effluent fractions from IgE-Sepharose were then fractionated on PLA 2 -Sepharose, and the effluent and eluate fractions from the immunosorbent were assessed for GIF activity. The results shown in Table IV indicate that the majority of GIF from 5 out of 7 clones bound to PLA,-Sepharose and recovered by elution at acid pH. In order to confirm that cross-linking of CD3 is required for these clones to produce antigen-binding GIF, the 5 clones were cultured in anti-MGG-coated cells without treatment with anti-CD3. As expected, culture supernatants did not contain IgE-BF, and GIF in the supernatant failed to bind to PLA,-Sepharose.
The present invention provides a technique to allow the development of o GIF-producing T-cell populations from PBMC of patients allergic to bee o* venom PLA 2 and to establish GIF-producing hybridomas from the T-cells.
15 Representative hybridomas express CD3 determinants and TCRa3, indicating that they are T-cell hybridomas. Furthermore, the TcR complex on the hybridomas appears to be functional. Both parent T-cells (Carini, et al., J. Immunol. Methods, 127: 221, 1990) and the majority of the GIFproducing hybridomas (Tables III, IV) produced IgE-BF upon cross-linking of CD3. Cross-linking of TcRap on CL3 and AC5 clones by the monoclonal antibody WT31 and anti-MGG also resulted in the formation of IgE-BF (results not shown). Further testing of representative CD3 hybridomas showed that all of the CL3, BE12. AC5 and CB7 clones expressed both CD4 and CD8. Since BUC cells employed for construction 25 of the hybridomas are CD4' CD8' (personal communication from Dr. J.
Stobo), it is not clear whether the parent T-cells of the hybridomas coexpressed both CD4 and CD8.
The present expenments showed that some of the T-cell hybridomas produced antigen(PLA2)-binding GIF upon cross-linking of CD3 on the cells. This finding is in agreement with the fact that representative murine GIF-'orming hybridomas formed antigen-binding GIF upon stimulation with antigen-pulsed syngeneic macrophages or by cross-linking of CD3 on the cells (Iwata Ishizaka, J. Immunol., 141: 3270, 1988, Iwata, et al., J.
Immunol., 143: 3917, 1989), and suggested similarities between the antigen-binding GIFs from the two species. In the murine system, the antigen-binding GIF obtained from the hybridomas suppressed the in vivo antibody response in carrier (antigen)-specific manner. It was also found that the antigen-binding GIF from the hybridomas were composed of antigen-binding polypeptide chain and non-specific GIF (Jardieu and Ishizaka, in Immune Regulation by Characterized Polypeptides,
G.
Goldstein, et al., ed., Alan R. Liss, New York, p.595, 1987), and that the antigen-binding chain shared a common antigenic determinant 14-12 with *99o** those of the effectqr type suppressor T-cell factor (TseF) (Iwata, et al. ibid, 15 1989). Separate experiments have shown that both the monoclonal antio lipomodulin antibody 141-89 and anti-I-J antibodies bound not only GIF, but also non-antigen binding chain chain) of TseF and TsiF (Jardieu, et al., J. Immunol., 138: 1494, 1986, Steele, et al., J. Immunol., 142: 2213, 1989). These findings collectively suggest that the antigen-binding GIF is 20 identical to TseF. Parent T-cells of a representative murine Ts hybridoma 71B4 were obtained by stimulation of OVA-primed spleen cells by homologous antigen, followed by propagation of the antigen-activated
T-
cells in the presence of GIF. (Iwata Ishizaka, J. Immunol., 141: 3270, 1988). The same strategy was employed to obtain the parent cells of the 25 human T-cell hybridomas in the present expenments. Indeed, both nonspecific GIF and PLA 2 -binding GIF from the human hybridomas bound to 141B9-Sepharose which previous studies had shown could also absorb murine TsFs (Steele, et al., J. Immunol., 142: 2213, 1989). It could be that
PLA
2 -binding GIF from the human T-cell hybridomas represents human antigen-specific TseF. However, it is still possible that the antigen-binding GIF may be a counterpart of murine TsiF. Recent experiments in our laboratory have shown that the typical murine helper T-cell clone D10.G4.1 can produce antigen-binding GIF, if the cells were precultured in the presence of a phospholipase A 2 inhibitor, and then stimulated with antigen(conalbumin)-pulsed antigen-presenting cells (Ohno, et al., Interat.
Immunol., 2: 257, 1990). It was also found that this antigen-binding GIF bound to the monoclonal antibody 14-30, which is specific for TsiF (Ferguson and Iverson, J. Immunol., 136: 2896, 1986), rather than the monoclonal antibody 14-12. Green, et al., Mol. Cell Immunol., 3: 1987) also reported that D10.G4.1 clone produced antigen-binding TsF upon antigenic stimulation with UV-irradiated antigen-pulsed macrophages, and that this factor, together with accessory molecules, induced the generation of the effector type, antigen-specific Ts. Since PBMC from allergic patients contain helper T-cells, it is still possible that the antigenbinding GIF from the human hybridomas represents TsiF rather than TseF.
S. 15 Takeuchi, et al., Immunol., 141: 3010, 1988) established Tse clones from PBMC of KLH-primed individuals, who had received repeated injections of a large dose of homologous antigen. Modulin, et al., (Nature, 322: 459, 1986) also established Ts clones from lesions of lepromatous leprosy patients. However, prior to the present invention, effector molecules mediating suppressor activity (TsF) from human Ts cells have not been identified. Similarities between human GIF and mouse GIF suggest that the PLA,-binding GIF from human T-cell hybridomas may represent TsF from human suppressor T-cells. The T-cell hybridomas, which produce antigen-binding GIF, will facilitate biochemical 25 characterization of the molecules. It has been repeatedly shown in the mouse that Ts as well as TsF (antigen-binding GIF) suppressed the in vivo IgE antibody response more effectively than the IgG antibody response (Ishizaka, et al., J. Immunol., 114: 110, 1975). If the allergen-binding GIF from the human T-cell hybricomas actually represent TsF, it is a reasonable expectation that the T-cell factor may suppress the IgE antibody response of the donor of parent T-cells.
EXAMPLE 4 PREPARATION OF HYBRIDOMA CELL LINES PRODUCING CEDAR POLLEN-SPECIFIC
GIF
Japanese cedar pollen is a major allergen in Japan and causes seasonal allergic rhinitis and conjuctivitis in a large percentage of the population. In order to further test the general applicability of the teachings of the invention to other antigens, the methods for generating antigen-specfic GIF-producing T-cells and T-cell hybridomas (described above) were o o/ applied to peripheral blood mononuclear cells from patients allergic to S* Japanese cedar allergen.
The major allergen in Japanese cedar (Sugi, Cryptomena japonica) is a glycoprotein designated cryj-1 (Yasueda, et al., J. Allergy and Clin.
.15 Immunol., 71:77, 1983). For these studies, the allergen was isolated from extracts of cedar pollen by this method with slight modifications. Briefly.
pollen was defatted with ether, and extracted 3 times with 0.125M ammonium bicarbonate. Carbohydrate in the extracts were removed by nexadecyltrimethyl ammonium brorr 7e. Proteins in the extracts were orecipitated with 80% saturated ammonium sulfate, and the precipitate dissolved in 0.05M Tris-HCI buffer, pH 7.8. After extensive dialysis against the Tris-HCI buffer, the protein fraction was applied to a DEAE cellulose column (DE-52, Whatman), and a flow-through fraction was obtained. The fraction was concentrated, dialyzed against 0.01M acetate buffer, pH and applied to a CM cellulose column (CM-52. Whatman), which was ecuilibratec with the buffer. Tne column was washed with the buffer, anc Croteins retained in tne column elutec witn 0. M onosohate buffer containing 0.3M sodium chloride. Proteins in the eluate were further fractioned by gel filtration through a Sephacryl S-200 HR column to obtain a major protein fraction containing cryj-1. The major protein in the fraction was 42kDa as determined by SDS-polycrylamide gel electrophoresis, and N-terminal amino acid sequence of the protein was identical to that of cryj-1. The protein was conjugated to Affigel 10 at 1.5 mg/ml gel.
A synthetic phospholipase
A
2 inhibitor, 2-(p-amylcinnamoyl)-amino-4chlorobenzoic acid, (ONO-RS-082, ONO Pharmaceutical Co.) was used instead of recombinant human lipocortin 1. Previous experiments had shown that ONO-RS-82 is a specific inhibitor of phospholipase
A
2 and facilitates the generation of GIF-producing cells in mouse spleen cell cultures (Ohno, et al., International Immunology, 1:425, 1989). When spleen cells of ovalbumin-primed mice were stimulated with ovalbumin, and antigen-activated T-cells were propagated with IL-2 in the presence of either 2pM ONO-RS-082. or 3pg/ml recombinant human lipocortin I, GIF- S producing, antigen-specific T-cells were generated. Antigen stimulated Tcells and construction of T-cell hybridomas were carried out essentially the same as described above, except that purified cryj-1 was used as antigen, and ONO-RS-082 was employed as a phospholipase A 2 inhibitor. Thus, mononuclear cells were obtained from periheral blood of patients allergic to Jaoanese cedar pollen, and suspended in RPMI 1640 medium containing 10% fetal calf serum (FCS). A suspension of the mononuclear cells (3x106 cells/ml) were cultured for 3 cays in the presence of cryj-1. Non-adherent cells were recovered, resuspended in RPMI medium *25 containing 10% FCS, (3x10 5 cells/ml), and cultured for 4 days in the presence of 60 units/ml human IL-2 and 2pM ONO-RS-082. Cells propagated in this manner were then recovered and fused with BUC cells to construct hybridomas.
Hybridomas were treated with the monoclonal anti-CD3 antibocy SPB-T3b (Spits, et al., Hybridoma 2:423, 1983), and the presence of CD3 on the cells were tested by immunofluorescence. Only CD3+ hybridomas were subcloned by limiting dilution.
The CD3+ hybridoma clones were maintained in complete DME medium containing 10% FCS, and culture supernatant of each clone was assessed for the presence of GIF by using the 12H5 cells. Results obtained with hybridomas from one patient are shown in Table V. GIF activity was detected in culture supernatants of three hybridomas; 31E9, 3187, and 32B4. Supernatants of the other two hybridomas, 31H6 and 31H3, appear to have weak GIF activity. Thus, the GIF-producing hybridomas were treated with anti-CD3 antibody followed by anti-mouse immunoglobulin, and the cells were cultured for 24 hr. Culture supernatants were then fractionated on cryj-1 coupled immunosorbent. The presence of GIF ,15 activity in the flow-through fraction and the acid-eluate fraction from the immunosorbent was assessed by using the 12H5 cells. The results included in Table V indicate that GIF from the 31E9 cells bound to cryj-1- Affigel and could be recovered by elution at acid pH, whereas GIF from the 31 7 cells failed to bind to the antigen-coupled immunosorbent. The results indicate that the 31E9 cells produce GIF having affinity for cryj-1, uoon stimulation with anti-CD3.
a TABLE V PRODUCTION OF HUMAN CEDAR ALLERGEN-SPECIFIC HYBRIDOMASa Hybridoma rosette inhibitionb GIF activity in cryi-1 Sepharose c Clone (effluent/eluate) unbound bound none 0/23 0/29 31H6 20/13(r) 5/20 12/10(t) 31A11 0/25
ND
31E9 0/22 20/0(+) 31H3 23/12( 0/34 38/16(-) 10 31B7 4/24 31 F7 0/26
ND
32B4 22/14(t) 38/22(±) a Hybridomas in this table were derived from two separate experiments.
b Culture supernatants of unstimulated hybridomas were screened for the 15 presence of GIF. Aliquots of 12H5 cells were incubated with culture supernatant of each hybridoma in the presence of mouse IgE. Culture supernatants of the 12H5 cells were filtered through CF50A to remove IgE, and filtrates were fractionatea on lentil lectin Sepharose. IgE-BF in the effluent and eluate fractions was assessed by rosette inhibition.
t 20 Numbers in the column represent the percent rosette inhibition by the effluent/eluate fractions from lentil lectin Sepharose. signs indicate the presence or absence of GIF, respectively.
Representative hybridomas were treated with anti-CD3 antibody and culture supernatants were fractionated on cryj-1 coupled Affigel. The presence of GIF activity in the flow-through (unbound) fraction, and acid eluate (bound) fraction was determined by using 12H5 cells. Culture filtrates of the 12H5 cells were fractionated on lentil lectin Sepharose.
Numbers represent percent rosette inhibition by the effluent/eluate fractions from lentil lectin Sepharose. GIF from the 31 E9 cells bound to cryi-1-Affigel and was recovered by elution at acid pH, while GIF from the 31B7 cells faileo to be retained in the cryj-1-Affigel column.
EXAMPLE PREPARATION AND CHARACTERIZATION
OF
HYBRIDOMA CELL LINES PRODUCING MONOCLONAL
ANTIBODIES
SPECIFIC FOR HUMAN GIF A. CONSTRUCTION AND SCREENING OF HYBRIDOMAS Human GIF in culture supernatant of the T-cell hybridoma CL3 was purified by using anti-lipomodulin (141-B9)-Sepharose. The affinity-purified GIF was mixed in complete Freund's adjuvant, and BALB/c mice were immunized by intraperitoneal injections of the antigen, given 3 times at 2 week intervals. Two weeks after the last immunization, spleen cells of the immunized mice were obtained, and 1x107 spleen cells were transferred into syngeneic BALB/c mice which had been irradiated with 625R y ray.
The recipients were immunized immediately after cell transfer and 2 weeks 15 later with purified GIF included in incomplete Freund's adjuvant. One week after the booster, their spleen cells were fused with HPRT-deficient B cell line SP 2/0-14AG. The cells were cultured in HAT medium with BALB/c macrophages as feeder layer. One hundred and two hybridoma clones obtained in the culture were selected for the formation of mouse 20 immunoglobulin, and Ig-forming hybridomas were selected for anti-GIF antibody production by ELISA assay, followed by bioassay using the 12H5 cells.
In ELISA assay, Immulon I plates (Dynatech) were coated with affinitypurified GIF. Control wells were filled with DPBS. After blocking the wells with 2% BSA, culture supernatants were applied to each well, and the binding of mouse Ig to the wells was determined by using alkalinephosphatase-coupled anti-mouse Ig antibodies. As shown in Table VI, 44.
culture supernatants of 11 hybridoma clones gave a significant ELiSA signal.
TABLE VI Selection of Anti.GIF.Producing Hybridomas 8 Hybridoma Ig Clone jsotvoe Sioal/Cotr GIF Activityc Efftuen/lut none 334 F 3550 338H 318H 388FI 4768 489G 481 F 3350 15 419A 31 2F Medium Control 1gM 1gM 1gM 1gM lgG 2 a 1gM lam 1gM 1gM 1gM 1gM 0/0 0.195/0.003 0.388/0.012 0.31 6/0.050 0.149/0.046 0.892/0.100 0.100/0.020 0.1 74/0.00 0.460/0.092 0.203/0.073 0.542/0-15 0.533/0.029 0/0 29/1(+ 33/1(+ 28/0() 0/29(- 0/31(- 0/28() 0/20(- 7/15(? 18/0(+ 0/27(- 27/1() 14/8() 0/31 aCulture supernatants, which were oositive in the ELISA assay, were assessed for the ability to bind GIF from CL3 clone.
b Binding of mouse lg 'in culture supernatants of the hybridlomas to GIFcoated wells, as compared with nonsoecific binding of Ig in the same supernatants to BSA-coated wells. Optical density at 410 nip.
Mixtures of purified GIF with culture supernatants of hybridomas were filtered through YM100 membranes, and the filtrates were assessed for GIF activity. The 12H5 cells were cultured with mouse IgE in the presence of the filtrate. lgE-BF formed by the cells was fractionated on lentil lectin Sepharose and lgE-BF in the effluent and eluate fractions from the lectin-coupled Sepharose were assessed by rosette inhibition.
Numbers reoresent the percent rosette inhibition by the effluent-eluate fractions. GIF switched the nature of lgE-BG formed by the cells (top column vs. bottom column-). indicates the presence of GIF The presence of anti-GIF in the culture supernatants of the 11 hybridoma clones was then determined by using the 12H5 cells (Iwata, et J.
Immunol., 140: 2534, 1988). The globulin factor of culture supernatant from each clone was obtained by precipitation with 50% saturated ammonium sulfate. After dialysis against phosphate buffered saline, the fraction was adjusted to 1/5 volume of the original culture supernatant.
Aliquots of an affinity-purified GIF prepared from CL3 clone using 141B9 Sepharose. These aliquots were mixed with an equal volume of the globulin fraction from each clone, and the mixtures were incubated overnight at 4'C. The mixtures were then filtered through YM100 membranes, and the presence of GIF in the filtrates was assessed. Then, aliquots of a suspension of the 12H5 cells were mixed with an equal volume of the filtrate, and the cell suspensions were cultured for 24 hours .in the presence of 10 pg/ml mouse IgE. The culture supernatants were 15 filtered through CF50A membranes to remove IgE, and IgE binding factors o in the filtrates were fractionated on lentil lectin Sepharose. The results of the experiments, included in Table VI, indicate that GIF was removed by the culture supernatants of 338H, 318H, 388F,, 476B, and 335C clones, indicating that these hybridomas produce anti-GIF.
20 B. PURIFICATION OF HUMAN GIF WITH MONOCLONAL
ANTI-GIF
Among the six hybridoma clones which produced monoclonal antibodies to GIF, only 388F, produced IgG antibody. This hyoridoma was subcloned and cultured in high glucose Dulbecco's medium supplemented with FCS. Culture supernatants were concentrated by ultra filtration and IgG in the supernatants was recovered by using Protein A-Sepharose. The monoclonal antibody was then coupled to Tresyl-activated Sepharose to prepare immunosorbent. In order to determine whether the monoclonal antibody could bind the same molecules as those bound to antilipomodulin (141B9) SeDharose, GIF in culture supernatant was absorbed 47.
with 141-B9-Sepharose, and was recovered by elution at acid pH. The affinity-purified GIF preparation was then fractionated with anti-GIF (388F,)coupled Sepharose. After the effluent fraction was obtained, the immunosorbent column was washed with 10 column volumes of Dulbecco's phosphate buffered saline (DPBS), and then eluted with glycine HCI buffer, pH 3.0. A serial dilution of the effluent and eluate fractions were assessed for GIF activity by using the 12H5 cells. The results shown in Table VII indicate that GIF in the acid eluate fraction from 14189- Sepharose bound to the anti-GIF (388F,)-Sepharose and was recovered again by elution at acid pH. The results indicate that both anti-lipomodulin and anti-GIF bind human GIF.
48.
TABLE VII Fractionation of Partially Purified Human GIF on the Anti-GIF (388F,) Coupled Sepharosea Fraction from 3 88F,-Sepharose Effluent Dilution 1:10 1:20 GIF Activityb Effluent/Eluate 0/35 0/29 Eluate a..
a. a a *aa.
a a *o 1:20 1:40 1:40 39/0 26/0 27/0 0/27 10 Unfractionated Media Control SGIF in culture supernatants of CL-3 clone was purified by using the anti-lipomodulin Sepharose. The affinity purified GIF (1.5 ml) was fractionated on 0.75 ml of 388F,-coupled Sepharose. After recovering the effluent fraction, the column was washed with 10 column volumes of DPBS, and then eluted with 3 column volumes of giycine HCI, pH SGIF activity was assessed by using the 12H5 cells by the same procedures described in Table IV. Numbers in the column indicate the percentage rosette inhibition by the effluent/eluate fractions from lentil lectin Sepharose. indicates the presence of GIF.
49.
In order to cetermine if the anti-human GIF could bind mouse GIF, mouse GIF from Ts hybridoma, 231F, cells were purified by usina 141B9- Sepharose, and aliquots of the purified mouse GIF were fractionated on either 1 4 1 B9-Sepnarose or 388F,-Sepharose. After the effluent fractions were obtained, immunosorbents were washed with 3 column volumes of DPBS, and then eluted with 3 column volumes of glycine HCI buffer, pH As expected, all GIF activity was absorbed to 141B9 Sepharose, and recovered by elution at acid pH. Neither the effluent nor washing fraction contained GIF activity. When the same GIF preparation was fractionated on 388F,-Sepharose, weak GIF activity was detected in the effluent fraction.
The majority of the activity was detected in washings with DPBS. but the acid eluate fraction did not contain a detectable GIF activity. It apoears that mouse GIF bind to anti-human GIF with extremely low affinity, and disassociate from the immunosorbent by washing at neutral pH. These 15 results indicate that the monoclonal antibody 388F, is specific for human
GIF.
C. PURIFICATION OF HUMAN GIF BY ION EXCHANGE
CHROMATOGRAPHY
AC5 cells were suocloned by limiting dilution and CD3' clones obtained.
These cells were then adjusted to serum-free ABC medium. Expression of CD3 on the subclones cultured in the medium was confirmed bv fluorocytometry. Culture supernatants of CD3' subclones were concentrated 10-30 fold, and GIF activity in serial dilutions of the preDarations was determined. Based on these results, subclone (AC5-23) 25 was selected, since a 1:3 dilution of the 10-fold concentrated supernatant of this subclone could switch the 12H5 cells from the formation of glycosylated IgE-BF to the formation of unglycosylated IgE-BF.
Studies were oone to determine whether human GIF could be purified by ion-exchange column chromatography. Culture supernatant of the subclone in ABC medium was concentrated 25-fold. A 10ml aliquot of the concentrated culture supernatant was adjusted to pH 8.0 with Tris, diluted 8-fold with distilled water, and then applied to a DEAE-Sepharose column.
Proteins bound to the column were eluted with 10mM Tris buffers containing increasing concentrations of NaCI (see Example Each fraction was concentrated to 10ml and assessed for GIF activity.
o• TABLE VIII Distribution of GIF Activity in DEAE-Sepharose Fractions Fraction Tris HCI NaCI (mM) a 100 150 200 Protein Content (ua) b 65.5 35.0 42.5 38.5 41.5 42.0 GIF ACTIVITY 21/0 20/6 7/20 3/19 0/21 0/20 0/22 .r medium control Concentrated culture supernatants of the AC5 cells were diluted 8-fold with distilled water, and applied to DEAE-Sepharose column. Fraction 1 represents passed through fraction combined with washing with Tris HCI pH 8.0 containing 20mM NaCl. The column was eluted stepwise with 10mM Tris hCI containing increasing concentrations of NaCI.
Total protein recovered after concentration of each fraction. After elution with Tris buffer containing 200mM NaCI, much protein retained in the column.
GIF activity was detected by using the 12H5 cells. Numbers represent the percent rosette inhibition by the affluent/eluate fractions from lentil lectin Sepharose. The proportion of RFC in the absence of IgE-BF was 22.6 0.7 indicate the presence or absence of GIF.
52.
As shown in Table VIII, the GIF activity was detected in the passed-through fraction and in the eluate with 50mM NaCI, but not in the other fractions.
Titration of a serial dilutions of the first two fractions indicated that the pass-through fraction had higher GIF activity than the 50mM fraction.
Repeated experiments with a separate culture supernatant confirmed that the majority of GIF in culture supernatants could be recovered from a DEAE-Sepharose column, when culture supernatant of AC5 cells were concentrated 100-fold, diluted 3-fold with distilled water, and then passedthrough the column. The passed-through fraction and washings with 10mM Tris buffer containing 50mM NaCI were combined, and concentrated to the original volume of the sample applied to DEAE-Sepharose. Titration of GIF activity in serial dilutions of the concentrated culture supernatant and the passed-through (50mM NaCI) fraction showed that a 1:30 dilution of both samples could switch the 12H5 cells from the formation of glycosylated IgE-BF to the formation of unglycosylated IgE-BF. It was also found that 75 to 80% of protein in the culture supernatant could be removed by passing through the DEAE-Sepharose.
In order to estimate the molecular mass of GIF, 0.5ml of the concentrated passed-through fraction from the DEAE-Sepharose was applied to a 20 Superose-12 column and proteins were eluted at a flow rate of 1ml/min.
In this experiment, 5ml fractions were collected, and each fraction was assessed for GIF activity by using the 12H5 cells. GIF activity was detected in fraction 9, which was recovered between 70 and 75 min. Since Fractions 6 and 8 may also have a weak activity, GIF activity in the serial dilutions of fractions 6, 8, 9 was assessed. The GIF was detected in a 1:10 dilution of fraction 9, but not in a 1:2 dilution of the other fractions. The results suggested that the molecular mass of the major species of human GIF is in the range of 11 KDa to 18KDa. For better estimation of the size of GIF molecules, gel filtration on a Superose-12 was repeated in the same design, except that 1ml fraction or 2.5ml fractions were collected. Three separate experiments indicated that the majority of GIF was recovered between 68 and 72 min. It appears that the molecular mass of GIF is 12- 18KDa, as estimated by gel filtration.
Studies were also done identifying GIF by SDS PAGE. Two liter culture supernatant of the hybridoma in ABC medium were concentrated 100-fold, and fractionated on a DEAE-Sepharose column. Based on the experiments described above, the concentrated supernatant was diluted 3-fold with deionized water and passed through the DEAE-Sepharose. The passed-through fraction was concentrated, pre-absorbed with human IgGcoupled Sepharose, and GIF in the fraction was purified by affinity chromatography on the 388F1-coupled Affigel (see Example In some experiments, the acid eluate fraction from the immunosorbent was adjusted S: to pH 8.0, and affinity-purification with 388F1-Affigel was repeated. Analysis of the affinity purified GIF preparation by SDS PAGE was performed under i reduced and non-reduced conditions. The major band in the affinity- *purified material has the molecular mass of 14KDa under reduced conditions and 15KDa under non-reduced conditions. In addition, a 67KDa f band was frequently observed. A portion of the affinity-purified preparation was dialyzed against DPBS and the GIF activity in the preparation was titrated. Assuming 100% recovery of GIF during dialysis and lyophilization, the sample applied to SDS-PAGE should have a GIF titer of 1:250.
Experiments were carried out to determine the relationship between the 14KDa protein and GIF. The GIF in 2 liter culture supernatant of AC5 clone was purified by DEAE-Sepharose chromatography followed by affinitypurification using 388F1-Affigel. Acid eluates from the immunosorbent was adjusted to pH 8.0, concentrated to i ml by ultrafiltration and fractionated on a Superose 12 column. Every 2.5ml eluate fractions were assessed for activity by using the 12H5 cells. In this experiment, the majority of GIF activity was detected in the fraction eluated between 67.5 and 70 minutes.
The presence of GIF in the fraction was confirmed by ELISA using biotincoupled mAb 141-B9. Although the ELISA signal was weak, only the GIFcontaining fraction gave ELISA signal. One ml of the GIF-containing fraction was lyophilized and analyzed by SDS PAGE. The results confirmed that the 14KDa peptide is present in the GIF-containing fraction.
Deposit of Materials The following cell lines have been deposited with the American Type Culture Collection, 1301 Parklawn Drive, Rockville, MD, USA (ATCC): Cell Line ATCC Accession No. Deposit Date 388F, HB 10472 May 31, 1990 AC5 HB 10473 May 31, 1990 15 31E9 HB 11052 June 2, 1992 These deposit were made under the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure and the Regulations thereunder (Budapest 20 Treaty). This assures maintenance of viable cultures for 30 years from the date of deposit. The organisms will be made available by ATCC under the terms of the Budapest Treaty which assures permanent and unrestricted availability of the progeny of the culture to the public upon issuance of the pertinent U.S. patent or upon laying open to the public of any U.S. or 25 foreign patent application, whichever comes first, and assures availability of the progeny to one determined by the U.S. Commissioner of Patents and Trademarks to be entitled thereto according to 35 USC §122 and the Commissioners rules Cursuant tnereto (incluoing 37 CFR §1.14 witr particular reference to 886 OG 638).
The assignee of the present application nas agreed that if the culture deposit should cie or be lost or destroyea when cultivated uncer suitable conditions, it will be promptly replaced on notification with a viabie specimen of the same culture. Availability of a deposited strain is not to be construed as a license to practice the invention in contravention of the rights granted under the authority of any government in accordance with its patent laws.
-0 The foregoing written specification is considered to be sufficient to enaoie one skilled in tne art to practice the invention. The present invention is nor to be limited in scope by the cell lines deposited, since the deoosited emboorment is intendeo as a single illustration of one aspect of the nvention ano any cell lines that are functionally equivalent are within the 15 scope of this invention. The deposit of material does not constitute an admission that the written description herein contained is inadequate to enable the practice of any aspect of the invention, including the best mode thereof, nor is it to be construed as limiting the scope of the claims to the spec:fic illustration that it represents. Inoeea, various modifications of the invention in addition to those shown and described nerein will become apparent to those skilled in the art from the foregoing oescription ano fal; ~within ;he scooe of the aopended c;a!ms.
56 The claims defining the invention are as follows: 1. Substantially pure human antigen-specific glycosylation inhibiting factor (GIF) having a molecular mass of 12-18 kDa.
2. The GIF of claim 1, wherein the antigen is an allergen.
3. The GIF of claim 1, wherein the antigen is an autoimmune antigen.
4. The GIF of claim 2, wherein the allergen is a venom.
The GIF of claim 4, wherein the venom is honey bee venom.
6. The GIF of claim 5, wherein the honey bee venom allergen is PLA 2 7. The GIF of claim 6 which has the antigen binding specificity of the human antigen-specific GIF produced by cell line AC5 having ATCC accession number HB 10473.
a 8. The GIF of claim 6 which is produced by cell line AC5 having ATCC accession number HB 10473.
9. The GIF of claim 2, wherein the allergen is pollen.
25 10. The GIF of claim 9, wherein the pollen is tree pollen.
11. The GIF of claim 10, wherein the tree pollen is cedar pollen.
12. The GIF of claim 11 which has the antigen binding specificity of the human antigen-specific GIF produced by cell line 31E9 having ATCC accession number HB 11052.
Claims (39)
13. The GIF of claim 11 which is produced by cell line 31E9 having ATCC accession number HB 11052.
14. A continuous hybridoma cell line capable of producing human antigen-specific glycosylation inhibiting factor (GIF). The hybridoma of claim 14, wherein the antigen is an allergen.
16. The hybridoma of claim 14, wherein the antigen is an autoimmune antigen.
17. The hybridoma of claim 15, wherein the allergen is a venom.
18. The hybridoma of claim 17, wherein the venom is honey bee 15 venom.
19. The hybridoma of claim 18, wherein the honey bee venom allergen is PLA 2 20 20. The hybridoma of claim 19 which is cell line AC5 having ATCC accession number HB 10473. o*
21. The hybridoma of claim 15, wherein the allergen is pollen. 25 22. The hybridoma of claim 21, wherein the pollen is tree pollen.
23. The hybridoma of claim 22, wherein the tree pollen is cedar pollen.
24. The hybridoma of claim 23, which is cell line 31E9 having ATCC accession number HB 11052. A method of producing a continuous hybridoma cell line caoable of producing human antigen-spec;fic glycosylation inhibiting factor (GIF) comprising: obtaining human antigen-primed T-cells activated to the 3 antigen; and combining the activated T-cells by fusion with a fusion partner cell line to produce hybridomas capable of producing human antigen-specific GIF.
26. The methoc of claim 25, wherein the human antigen-primed T-cells of step which are activated to the antigen are produced by: acquiring a sample containing human antigen-primed T-cells; 5 (ii) activating the T-cells by culturing the T-cells in tre presence of the antigen for which the T-cells are primed; and (iii) culturing the activated T-cells in the presence of IL-2 and phospholipase A 2 inhibitor. 27 The method of claim 26, wherein the phospholipase A, inhibitor is Slipocortin.
28. The methoa of claim 26, wherein the antigen for which the T-cells are primed is chemically modified.
29. The metnod of claim 28, wherein the antigen is chemically modifiec with guanidine HCI. The method of claim 28, wherein the antigen is chemically modified witn cyanogen bromide.
31. The method of claim 25. wherein the human antigen-primed T-cells are from the peripheral blood mononuclear cell fraction.
32. The method of claim 25, wnerein the fusion partner cell line is a human lymphoblastoid T-cell line.
33. The method of claim 32. wherein the lymphoblastoid cell line is BUC.
34. A process for producing substantially pure human antigen-specific GIF comprising: culturing a continuous hybridoma cell line capable of producing human antigen-specific GIF such that the cell iene 5 produces human antigen-specific GIF; and isolating substantially pure human antigen-specific GIF from the culture.
35. The process of claim 34, wherein the continuous hybridoma cell line is produced by the method of claim
36. The process of claim 34. wherein during cultunng the hybridoma cell line is exposed to antigen-pulsed syngeneic macrophages or antibodies to the CD3 or T-cell receptor.
37. The process of claim 34. wherein the human antigen-specific GIF is substantially purified by affinity purification using the antigen to S"which the antigen-specific GIF binds. F_
38. The process of claim 34, wherein the step of isolating substantially pure human antigen-specific GIF includes: reacting the hybridoma cell line culture with a monoclonal antibody specifically reactive with human GIF; (ii) eluting the human GIF from the monoclonal antibody; (iii) reacting the eluted GIF with the antigen to which the human antigen-specific GIF binds; (iv) eluting the human antigen-specific GIF from the antigen; and recovering the human antigen-specific GIF.
39. The process of claim 34, wherein the step of isolating substantially pure human antigen-specific GIF includes: reacting the hybridoma cell line culture with the antigen to which the human antigen-specific GIF 5 binds; (ii) eluting the human GIF from the antigen; (iii) reacting the eluted GIF with a monoclonal antibody specifically reactive with human GIF; (iv) eluting the human antigen-specific GIF from the 0 monoclonal antibody; and recovering the human antigen-specific GIF. 61 The process of claim 34. wherein the steo of isolating substantially pure human antigen-specific GIF includes: contacting the hybridoma cell line culture with a anionic exchange resin; (ii) eluting the human GIF from the matrix; (iii) reacting the eluted GIF with a monoclonal antibody specifically reactive with human GIF or with the antigen to which the human antigen-specific GIF binds, or both: (iv) eluting the human antigen-specific GIF; and recovering the human antigen-specific GIF.
41. The process of claim 40, wherein the anion exchange moiety attached to the resin is DEAE.
42. The process of claim 40, wherein the human GIF is eluted using a salt concentration from about 10mM to about 100mM.
43. The process of claim 40, wherein the salt is NaCI.
44. The process of claim 40, wherein the hybridoma cell line culture is concentrated from about 25-fold to about 1000-fold then diluted from about 3-fold to about 10-fold before contacting with the anionic exchange resin. The process as in either of claims 38 or 39, wherein the monoclonal antibody has the specificity of a monoclonal antibody produced by cell line 388F, having ATCC accession number HB 10472.
62. 46. The process as in either of claims 38 or 39, wherein the monoclonal antibody is produced by cell line 388F, having ATCC accession number HB 10472. 47. A method of suppressing a human immune response to an antigen which comprises administering to the human an immunosuppressively effective amount of human GIF having a molecular mass of 12-18 kDa, wherein the human GIF can specifically bind to the antigen. 48. The method of claim 47, wherein the antigen is an allergen. 49. The method of claim 47, wherein the antigen is an autoimmune antigen. 15 50. The method of claim 48, wherein the antigen is a venom. S. 51. The method of claim 50, wherein the venom is honey bee venom. 52. The method of claim 51, wherein the honey bee venom allergen is 20 PLA 2 g *53. The method of claim 52, wherein the human GIF has the antigen binding specificity of the human antigen-specific GIF produced by cell line AC5 having ATCC accession number HB 10473. S 54. The method of claim 52, wherein the human GIF is the human antigen-specific GIF produced by cell line AC5 having ATCC accession number HB 10473. 55. The method of claim 48, wherein the allergen is pollen. TE 63 56. The method of claim 55, wherein the pollen is tree pollen. 57. The method of claim 56, wherein the tree pollen is cedar pollen. 58. The method of claim 57, wherein the human GIF has the antigen binding specificity of the human antigen-specific GIF produced by cell line 31E9 having ATCC accession number HB 11052. 59. The method of claim 57, wherein the human GIF if produced by cell line 31E9 having ATCC accession number HB 11052. The method of claim 47, wherein the administration is parenteral. 61. The method of claim 60, wherein the parental administration is by 15 subcutaneous, intramuscular, intraperitoneal, intracavity, transdermal, or intravenous injection. *99 62. The method of claim 47, wherein the administration is at a dosage of about 0.001 mg/kg/dose to about 2 mg/kg/dose.
63. The method of claim 47, wherein the administration is at a dosage of about 0.001 mg/kg/dose to about 0.2 mg/kg/dose. 9*
64. A pharmaceutical composition comprising immunosuppressive amounts of substantially purified human antigen-specific GIF having a molecular mass of 12-18 kDa and a pharmaceutically inert carrier. 64 The pharmaceutical composition of claim 64, wherein the substantially purified human antigen-specific GIF has the antigen binding specificity of the human antigen-specific GIF produced by the cell line AC5 having ATCC accession number HC 10473 or by cell line 31E9 having ATCC accession number HB 11052.
66. The pharmaceutical composition of claim 64, wherein the substantially purified human antigen-specific GIF is produced by cell line AC5 having ATCC accession number HB 10473 or by cell line 31E9 having ATCC accession number HB 11052.
67. A continuous hybridoma cell line capable of secreting monoclonal antibodies specifically reactive with human GIF, but not reactive with lipomodulin. *.15
68. The hybridoma cell line of claim 67 which is cell line 388F 1 having ATCC accession number HB 10472.
69. A monoclonal antibody specifically reactive with human GIF, but not 20 reactive with lipomodulin.
70. The monoclonal antibody of claim 69, wherein the monoclonal antibody has the specificity of the monoclonal antibody produced by hybridoma cell line 388F, having ATCC accession number HB 25 10472.
71. The monoclonal antibody of claim 70, where the monoclonal antibody is produced by hybridoma cell line 388F, having ATCC accession number HB 10472.
72. Substantially pure human antigen-specific glycosylation inhibiting factor (GIF), which GIF is substantially as hereinbefore described with reference to Example
73. A continuous hybridoma cell line capable of producing human antigen-specific glycosylation inhibiting factor (GIF), which hybridoma cell line is substantially as hereinbefore described with reference to Example 1 or Example 4.
74. A method of producing a continuous hybridoma cell line capable of producing human antigen-specific glycosylation inhibiting factor (GIF), which method is substantially as hereinbefore described with reference to Example 1 or Example 4. 15 75. A process for producing substantially pure human antigen-specific GIF, which process is substantially as hereinbefore described with reference to Example
76. A continuous hybridoma cell line capable of secreting monoclonal 20 antibodies specifically reactive with human GIF, but not reactive with lipomodulin, which hybridoma cell line is substantially as hereinbefore described with reference to Example
77. A monoclonal antibody specifically reactive with human GIF, but not 25 reactive with lipomodulin, which monoclonal antibody is substantially a. a as hereinbefore described with reference to Example DATED THIS 20TH DAY OF SEPTEMBER 1996 LA JOLLA INSTITUTE FOR ALLERGY AND IMMUNOLOGY By their Patent Attorneys CULLEN CO ABSTRACT The invention relates to human antigen-specific glycosylation inhibition factor (GIF) The invention also provides a continuous hybridoma cell line capable of producing GIF, a method of producing such a cell line, a process for producing GIP, and a method of suppressing a human immune response using GIF. 0 00 :00090 0. 0*0 0. *'000 of 0 000 0 *000 ::66 0
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US70937591A | 1991-06-03 | 1991-06-03 | |
US709375 | 1991-06-03 | ||
AU21617/92A AU2161792A (en) | 1991-06-03 | 1992-06-03 | Suppressor t-cell hybridoma |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU21617/92A Division AU2161792A (en) | 1991-06-03 | 1992-06-03 | Suppressor t-cell hybridoma |
Publications (2)
Publication Number | Publication Date |
---|---|
AU6573196A AU6573196A (en) | 1997-01-16 |
AU707739B2 true AU707739B2 (en) | 1999-07-15 |
Family
ID=25618349
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU65731/96A Ceased AU707739B2 (en) | 1991-06-03 | 1996-09-20 | Suppressor T-cell hybridoma |
Country Status (1)
Country | Link |
---|---|
AU (1) | AU707739B2 (en) |
-
1996
- 1996-09-20 AU AU65731/96A patent/AU707739B2/en not_active Ceased
Also Published As
Publication number | Publication date |
---|---|
AU6573196A (en) | 1997-01-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
RU2047177C1 (en) | Method for detection of ige-producing b-lymphocytes | |
Carrel et al. | Subsets of human Ia-like molecules defined by monoclonal antibodies | |
KR100584051B1 (en) | Peptide Immunogens for Vaccination against and Treatment of Allergy | |
WO1992007574A1 (en) | Glycoproteins associated with membrane-bound immunoglobulins as antibody targets on b cells | |
JPH07126300A (en) | Monoclonal antibody that recognizes specific antibody of human b cell receptor conjugate | |
WO1992020373A1 (en) | Heteroconjugate antibodies for treatment of hiv infection | |
JPH04506359A (en) | Antibody antagonist of human interleukin-4 | |
US5298420A (en) | Antibodies specific for isotype specific domains of human IgM and human IgG expressed or the B cell surface | |
Moore et al. | Rodent IgE-binding factor genes are members of an endogenous, retrovirus-like gene family. | |
US5723582A (en) | Antigen-specific human glycosylation inhibiting factor | |
JPH05503687A (en) | Treatment of autoimmune diseases | |
EP0407399A1 (en) | Immunosupression with anti-pan t-cell immunotoxin compositions | |
JPH10509948A (en) | Human antibodies to T cell receptor peptides and methods for their preparation | |
JPH06510790A (en) | Secreted Mac-2 binding glycoprotein | |
JP3578787B2 (en) | Method for recombinant production of biologically active polypeptide | |
AU707739B2 (en) | Suppressor T-cell hybridoma | |
Fehlner et al. | Murine T cell responses to melittin and its analogs. | |
JPH07242566A (en) | Immunosuppressive agent | |
WO1992021240A1 (en) | Suppressor t-cell hybridoma | |
US5786168A (en) | Method for recombinant production of antigen non-specific glycosylation inhibiting factor (GIF) | |
Quan et al. | Purification and partial amino acid sequence of suppressive lymphokine from a CD8+ CD57+ human T hybridoma. | |
Thomas et al. | Glycosylation-inhibiting factor from human T cell hybridomas constructed from peripheral blood lymphocytes of a bee venom-sensitive allergic patient. | |
Mori et al. | Epitope specificity of bee venom phospholipase A2-specific suppressor T cells which produce antigen-binding glycosylation inhibiting factor | |
JPH11509723A (en) | Preparation of antigen-specific glycosylation inhibitor | |
JP2004149507A (en) | Immunosuppressive agent |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
MK14 | Patent ceased section 143(a) (annual fees not paid) or expired |