Monoclonal Antibody Specific to a Novel Epitope of the LFA-1 Antigen of Human T Lymphocytes

FIGS. 1A and 1B are fluorescence graphs showing the reactivities of the monoclonal antibody embodying the invention and known monoclonal antibody 2F12, with certain T lymphocytes.

FIGS. 2A and 2B are graphs developed to identify the lymphocyte sub-population into which the cytotoxic effector cell is classified CD8 control cells are designated by a line drawn between open circles, CD8+S6F1- subset cells are designated by a line drawn between closed squares and CD8+S6F1+ subset cells are designated by a line drawn between closed circles.

FIGS. 3A and 3B are graphs developed to show the relationship between suppressor effector activity and T8 cell populations bearing the S6F1 antigen.

FIGS. 4A, 4B and 4C depict graphs developed to show immunoprecipitation analysis data of the S6F1 and LFA-1 antigens.

Peripheral blood lymphocytes were separated from blood samples obtained from healthy donors by Ficoll-Hypaque density gradient centrifugation. The lymphocytes were separated into T and non-T cell populations by E rosette formation with sheep erythrocytes as described in Meuer et al., supra. Initially, monoclonal antibodies were screened for reactivity with T cells to discern any pattern of specific cell binding. T4 and T8 lymphocyte cells were obtained by complement-mediated lysis with T4 monoclonal antibody and T8 antibody respectively and then reactivity with monoclonal antibody analyzed by cell sorting prodcedures using an EPICS.RTM. instrument marketed by Coulter Corporation of Hialeah, Fla. The monoclonal antibody embodying the invention was determined after analysis for reactivity with unfractionated T, T4 and T8 cells and characterized as reactive preferentially with T8 cells and binding specifically to the antigen identified herein as "S6F1". The monoclonal antibody, anti-S6F1, can distinguish killer effector and suppressor effector cells in T8 lymphocyte populations. The monoclonal antibody defines a cell surface structure comprised of a 180,000 dalton and 95,000 dalton molecular weight glycoprotein. Sequential immunoprecipitation studies and two dimensional gel analysis, as will be fully discussed, indicate that the S6F1 monoclonal antibody recognizes a novel epitope on the LFA-1 antigen.

Referring to FIGS. 1A and 1B, data is provided to confirm the reactivity of the S6F1 monoclonal antibody with certain T lymphocytes. FIG. 1A represents a cytofluorographic analysis of unfractionated T lymphocytes, CD4+ lymphocytes and CD8+ lymphocytes derived from normal human blood from ten donors. 1.times.10.sup.6 cells of each T lymphocyte population was analyzed for their reactivity with the S6F1 monoclonal antibody at a ratio of 1:1000 by indirect immunofluorescence on an EPICS.RTM.C flow cytometer commercially available from Coulter Corporation of Hialeah, Florida. The cell preparation and staining procedures employed were conventional and well known in the art of cell sorting. Each histogram displays the number of cells along the ordinate or Y-axis versus fluorescence intensity along the abscissa or X-axis thereof.

In this representative study showing the fluorescence profile of the expression of the S6F1 antigen on the referenced T lymphocytes, the S6F1 antibody was reactive with 17% of unfractionated T cells, 14% CD4+ lymphocytes and 58% of CD8+ cells. Thus, the S6F1 antigen was expressed preferentially on the CD8+ population of lymphocytes. The S6F1 antibody reacted with greater than 70% of null cells, granulocytes and the T cell line HSB in this study. However, it reacted with only 10% of cells in population of macrophages, B cells, B cell lines, i.e., Raji, Ramos, Nalms-1, EB virus transformed cell lines, hematopoietic lines U-937, K-562, KG-1 or the T cell lines Molt 4, CEM and JM derived from other conducted testing.

FIG. 1B represents a study of the reactivity of CD4 and CD8 lymphocytes from normal donors with varied dilutions of the known monoclonal antibody 2F12 which recognizes the LFA-1 antigen. Analysis was by indirect immunofluorescence using an EPICS.RTM.C cell sorter. Each histogram displays cell numbers versus fluorescence intensity. In histograms a and d using anti-LFA-1 antibody at a dilution of 1:100, a cell reactivity of 98% resulted. In histograms b and e using the monoclonal antibody at a dilution of 1:1000, a cell reactivity of 98% resulted. In histograms c and f using the antibody at a dilution of 1:10000, a CD4 cell reactivity of 52% and a CD8 reactivity of 58% resulted. This analysis reflects the variation in staining patterns of both CD4 and CD8 lymphocyte populations with the monoclonal antibody 2F12 specific to the LFA-1 antigen whereas the S6F1 monoclonal antibody showed preferential binding to CD8+ lymphocytes.

Since the S6F1 antibody was seen to react with a sizeable population of CD8+ lymphocytes, the antibody was utilized to subdivide the CD8+ population of peripheral blood lymphocytes to assess the functional heterogeneity of these cells. A test protocol was implemented to study whether alloreactive cytotoxic T lymphocytes, either precursor or effector cells, could be separated from CD8+ suppressor cells with the S6F1 monoclonal antibody. To determine whether the precursor of alloantigen-specific cytotoxic T cells belonged either to the S6F1+ or S6F1-population of CD8+ cells, the specific cell mediated lympholysis (CML) by subsets of CD8+ cells was examined.

CD8+ cells were freshly isolated using known procedures and labelled with the S6F1 antibody and developed with fluorescein-conjugated F(ab')2 goat anti-mouse F(ab')2 (Tago). The labelled CD8 cells were sorted using the EPICS.RTM.C cell sorter into CD8+S6F1+ and CD8+S6F1-populations as described in Morimoto et al., J. Immunol 134:1508-1515 (1985). Then, 1.times.10.sup.6 /ml of CD4+ cells obtained by complement-mediated lysis added to generate killer inducer function in this system were cultured with 1.times.10.sup.6 /ml of unfractionated CD8+ cells, sorted CD8+S6F1+ or CD8+S6F1-cells in the presence of equal numbers of irradiated allogenic stimulator cells, i.e., E-cells in multiple well culture plates in 5% CO.sub.2 humidified at 37.degree. C. After six (6) days of culture, CD4 cells were removed from culture by complement-mediated lysis of CD4+ cells. .sup.51 Cr labelled release cell mediated lympholysis by subsets of CD8 cells was assayed after a four (4) hour cell incubation as described in Meuer et al., Proc. Natl. Acad. Sci, U.S.A. 79:4395-4399 (1982).

Referring to FIG. 2A, the graph details % Specific Lysis along the ordinate and E/T Ratio Target along the abscissa. A pair of experimental tests were conducted as depicted in FIG. 2A. It appeared that the majority of cytoxic T cells against specific target cells resided in the CD8+S6F1-subset, while a small portion was found in the CD8+S6F1+ subset.

Referring to FIG. 2B, the data displayed establishes that the cytotoxic effector T cells belong to mixed lymphocyte reaction (MLR) activated CD8+S6F1+ cells and not CD8+S6F1-cells. The protocol followed to achieve this determination was as follows: 1.times.10.sup.6 ml unfractionated T cells were sensitized with equal numbers of irradiated (5000 Rad) allogenic (E-) stimulator cells in final RPMI 1640 medium (10% pooled human AB serum-Pel Freeze), 4 mML-glutamine, 25 mM Hepes buffer (Microbiological Associates), 0.5% sodium bicarbonate and 10% penicillin-streptomycin in a 25 cm.sup.2 flask in 5% CO.sub.2 humidified atmosphere at 37.degree. C. for six (6) days. Unfractionated T cells thus sensitized with allogenic E-cells were separated into CD8+S6F1+ and CD8+S6F1-subsets of cells by means of an EPICS.RTM.C cell sorter. Both unfractionated and fractionated CD8+ cells were analyzed in cell mediated lympholysis against .sup.51 Cr labelled target cells comprising thawed cryopreserved allogenic (E-) cells cultured over an approximate 10-12 hour period and incubated for one (1) hour at 37.degree. C. in a 0.2 ml of sodium chromate in saline solution (1 mCi/ml, New England Nuclear). The target cells, i.e., .sup.51 Cr labelled cells, were washed and resuspended at 10.sup.5 /ml in final media. A standard four (4) cytotoxity assay was performed using V-bottom wells of a culture plate. A pair of experiments was conducted as depicted in FIG. 2B.

In contrast to the data displayed in FIG. 2A, the CD8+S6F1+ lymphocyte population demonstrated greater allospecific killer effector activity than did the unfractionated CD8+ population. Virtually no killer effector activity was observed in the CD8+S6F1- lymphocyte population. Other data observed from these experiments, but not included herein, showed that the S6F1 monoclonal antibody did not block the effector function although the anti-MHC class I and anti-LFA-1 monoclonal antibodies blocked the killer effector function as described in Meuer et al., supra; Sanchez-Madrid et al., J. Exp. Med. 158:1785-1803 (1983); Sanchez-Madrid et al., Proc, Natl. Acad. Sci. U.S.A., 79:7489-7493 (1982). This data shows that killer precursor T cells are found in the CD8+S6F1- subset of cells, but the killer effector activity belongs to the CD8+S6F1+ subset.

The data developed by these tests suggest that the S6F1 antigen may become expressed on a S6F1 negative CD8+ cell population during a primary MLR reaction. A primary MLR was initiated using CD8+S6F1- lymphocytes combined with CD4+ cells and the T lymphocytes evaluated for expression of the S6F1 antigen. Three (3) separate experiments were conducted in which data developed showed 45-55% of CD8+S6F1- cells expressed the S6F1 antigen on their cell surfaces after MLR activation. These results indicate that the S6F1 antigen expression is induced on a S6F1 negative population of CD8+ cells.

Data was developed to determine which subset of CD8+ cells was activated to become suppressor effector cells as shown in FIG. 3A. Freshly isolated unfractionated T cells were separated into CD8+S6F1+ and CD8+S6F1- subsets using previously discussed technology and an EPICS.RTM.C cell sorter. Varying numbers of CD8 cells were added to 5.times.10.sup.4 B cells and 2.times.10.sup.4 CD4+ cells stimulated with pokeweed mitogen (PWM) in round bottom, multiple well, micro-culture plates and total IgG immunoglobulin production was measured after seven (7) days in culture. Culture supernatants for IgG secretion were determined by radio-immunoassay as described in Meuer et al., supra.

As shown in FIG. 3A, the addition of increasing numbers of unfractionated CD8+ cells resulted in the suppression of PWM driven IgG secretion by B cells in a dose dependent fashion. When increasing numbers of CD8+S6F1- cells were added to CD4+ and B cells, an even greater degree of suppression of IgG secretion was observed. In contrast, the addition of CD8+S6F1+ cells to the mixture of B and CD4+ cells resulted in only a slight decrease in IgG secretion. These results show that the precursor of CD8 suppressor cells belonged to the CD8+S6F1- subset of cells.

Data was developed to determine whether T cell suppressor activity was found in the CD8+S6F1+ population by examining the suppressor function of the subsets of CD8+ cells after MLR activation in an alloantigen-stimulated IgG synthesis system. Prepared T cells were activated in MLR against alloantigen for six (6) days and alloactivated T cells were separated into CD8+S6F1+ and CD8+S6F1- cells on an EPICS.RTM.C cell sorter. Unseparated CD8 and subsets of CD8+ cells were added to autologous peripheral blood lymphocytes (PBL) in the presence of irradiated alloantigen used MLR and the effects of alloantigen driven IgG synthesis was calculated as follows: ##EQU1##

As shown in FIG. 3B, alloactivated unfractionated CD8+ cells suppressed alloantigen driven IgG synthesis in a dose dependent manner. The CD8+S6F1- population activated with alloantigen showed greater suppressor activity in this system than did the unfractionated CD8+ cells. However, alloantigen activated CD8+S6F1+ lymphocytes exhibited a minimal suppressor activity only when very large numbers of cells were added to these cultures. These results indicate that both suppressor precursor and suppressor effector cells belong to the CD8+S6F1- lymphocyte population. More importantly, the above results imply that suppression is probably not merely a manifestation of cytotoxicity.

In view of the efficacy of the S6F1 antigen in the identification of cytotoxic effector T cells, the structure of the antigen was analysed by immunoprecipitation procedures of the S6F1 and LFA-1 antigen using anti-2F12. The analysis data is presented in FIGS. 4A, 4B and 4C.

Cells from T cell-enriched cells (95% E+ cells) were activated with Concanavalin A (ConA) (20 .mu.g/ml) for three (3) days and labeled with .sup.125 I by lactoperoxidase-catalysed iodination [per Hubbard et al., Biochemical Analysis of Membrane) (ed. Maddy, AH) P-427-501 (Chapman and Hall, 1976)], washed and then solubilized in lysis buffer (1% W/V Nonidet P-40 in 20 mM Tris-HCl buffer pH 8.0 containing 150 mM NaCl, 1 mM EDTA and 1 mM phenylmethyl (sulphony) fluoride). The lysates (5.times.10.sup.7 cell equivalents) were then incubated overnight at 4.degree. C. with 5 microliters of ascitic fluid of the individual antibodies before precipitating with 50 microliters of 10% w/v Protein A: Sepharose. The immunoprecipitates were then washed once in lysis buffer with 0.1% (w/v) SDS and twice with lysis buffer alone, prior to analysis by SDS-PAGE as described in Laemmli, Nature (London) 227:680-684 (1970). Lysates from labeled peripheral blood lymphocytes were precleared with the TS1/18 antibody for four immunoprecipitations prior to precipitation with the S6F1 antibody. Lysates were labelled and prepared as described, after which the immune complexes were eluted from the beads by incubation in isoelectric focusing sample buffer (9.2M urea, 2% Nonidet P-40, 2% ampholines pH range 2-11 (Serva) at 50.degree. C. for 30 min. Two dimension NEGPHE/SDS-PAGE analysis was done as previously described in O'Farrell, et al., Cell 12:1133-1142 (1977); Rudd et al., J. Biol. Chem. 200:1927-1936 (1985).

The FIG. 4 data depicts a composite of patterns of polypeptides precipitated by the S6F1 monoclonal antibody and the 2F12 monoclonal antibody which recognizes the LFA-1 antigen from Con A activated T cells. FIG. 4A displays two columns or lanes 1 and 2 in which the data of lane 1 relates to the S6F1 and lane 2 relates to the 2F12 monoclonal antibodies, respectively. Lane 1 shows that the S6F1 antibody precipitated two major bands at 180,000 daltons (Mr) and 95,000 daltons (Mr). Lane 2 shows that the 2F12 antibody precipitated a band pattern which appears to correspond to the alpha and beta subunits of the LFA-1 antigen, respectively.

The data of FIG. 4B is derived from a direct demonstration of the identity of the S6F1 antigen made by sequential immuno-depletion and two dimensional non-equilibrium gel analyses (NEPHGE/SDS-PAGE). Lane 1 relates to the S6F1 antibody and lane 2 relates to the 2F12 antibody. The data of FIG. 4B shows that the 2F12 antibody to LFA-1 antigen completely deleted material with which the S6F1 reacted. It is noted that the monoclonal antibody TS1/18 had been shown to react with the beta subunit of the LFA-1 antigen. See Sanchez-Madrid et al., supra, both citations.

FIG. 4C presents data developed from analysis by 2-D NEPHGE/SDS-PAGE of the S6F1 and LFA-1 antibodies. This data shows that antibodies against the S6F1 and LFA-1 antigens precipitated two polypeptides at 180,000 daltons and 95,000 daltons which focused at an approximate isolelectric position of pH 5.3 and 5.7, respectively. The data of FIG. 4C supports the observation that the S6F1 monoclonal antibody binds the LFA-1 antigen.

Although the S6F1 and 2F12 monoclonal antibodies immunoprecipitate the same structure, their reactivity with functionally distinct cell subsets appears to differ. The data graphically displayed in FIGS. 1A and 1B showed that the S6F1 monoclonal antibody stained CD8+ lymphocytes preferentially whereas the 2F12 antibody exhibited a reactivity with CD4+ lymphocytes also. It is believed that the reactivity of CD8+ lymphocytes with the S6F1 antibody does not reflect simply a low affinity of the antibody for the LFA-1 antigen structure itself.

We determined from analyses conducted with the S6F1 monoclonal antibody that the actual epitope it recognizes appears not to be involved in the process of cytotoxicity because the antibody was unable to block this function. However, other known monoclonal antibodies directed against the LFA-1 antigen were able to effectively block this cytotoxicity process. See Sanchez-Madrid et al., supra. Further, the S6F1 antigen was determined to be expressed primarily on CD8+ cells, null cells and a small fraction of macrophages and B cells. The data developed does indicate that the epitope on the LFA-1 antigen recognized by the S6F1 antibody may be formed distantly on the molecule by the development of a binding site on the LFA-1 antigen involved in the adherence function to target cells.

Thus, the S6F1 monoclonal antibody can distinguish between cytotoxic effector and suppressor effector cells in CD8 lymphocyte populations by recognizing a novel epitope on the LFA-1 antigen. This unique characteristic of the S6F1 antibody is useful in diagnosing and defining the CD8 cytotoxic effector cell and to distinghish it from the suppressor effector cell especially by flow cytometric techniques and by other known assay techniques as well. For instance, it is known that CD8 cytotoxic effector cells play a key role in human graft rejection, such as, in rejection of renal and other grafts. The S6F1 antibody to CD8+S6F1 cytotoxic effector cells could identify these effector cells selectively and thereby be useful to enable the functioning of the suppressor cell population to be of therapeutic value in reversing allograft rejection. Thus, the specificity of the S6F1 monoclonal antibody might be useful in the treatment of ongoing graft rejection.

This specificity of the S6F1 monoclonal antibody within the class of CD8 cells is particularly unique and useful in view of the heretofore unavailability of an antibody which could provide such a positive and more precise phenotypic means to distinguish suppressor effector from cytotoxic effector cells within the CD8 class of lymphocytes. Monoclonal antibodies to define the CD8 precursor of the cytotoxic cells in the human immune system is discussed in Clement et al., J. Immunol. 133:2461-2468 (1984) but without achieving the results achievable with the S6F1 monoclonal antibody embodying this invention.

A cell line which produces the S6F1 monoclonal antibody corresponding to this invention has been deposited in the American Type Culture Collection, Rockville, Md. 20852 on Oct. 30, 1987, prior to the filing of this application. The cell line was assigned ATCC Accession No. HB 9579.

The complete diagnostic and possible therapeutic applications of the monoclonal antibody of the invention have not been determined. The administration of an effective amount of the antibody, either alone or coupled to a radioisotope, drug or toxin may have therapeutic benefit in human organ transplant procedures consistent with the characteristics of the monoclonal antibody described herein.