A human suppressor T-cell factor that inhibits T-cell replication by interaction with the IgM-Fc receptor (CD7)

A Human Suppressor T-Cell Factor That Inhibits T-Cell Replication by Interaction with the IgM-Fc Receptor (CD7) Mohamed Emara, William M. Baldwin III, Olivera J. Finn, and Fred Sanfilippo

A B S T R A C T : We have previously described the induction of human suppe.,sor T cells from /resh peripheral blood lymphocyte~ of a kidney transplant recipient by in ritro stimulation with an autologous irradiated antidonor CTL hne (EE-1) grou,n.from a biopsy of the patient's own renal allograft. The induced T cells (designated TsEE) were shown to inhibit the in ritro generation of proliferative and cytotoxic responses of autologous T cells and nonautologous T celL that shared HLA-B7 with TsEE cells. Stimulation of TsEE cells hv the autologous irradiated inducer line (EE-I) produced soluble factors (designated TsEEF) that similarl}, inhibited auto/ogou.~ and nonautologous T-cell responses to al/oantigens and mitogens, but in a non-HLA-re.*tricted manner. In this study, u,e examined the functional interaction of TsEEF with ~ariou.~ ~ells surface receptors. TsEEF specifically inhibited the proliferation of stimulated and tran.*jbrmed T cells expressing CD7, a putative receptorfor IgM-Fc (FcRtx). Blocking or capping of CD7-FcR# determinants on responder T cells by pretreatment with IgM or anti-CD7 monoclonal antibodies (3A1. HuLy-m2) abrogated TsEEF activiLy. Conversel.y. pretreatment of T cell with TsEEF significantly reduced their binding of IgM and HuLy-m2. TsEEF was demonstrated not to be IgM or lgG, and its activi(y u,as not removed b~' preabsorption with IgM or lgG; however, it.~ activi(y could be competitively inhibited by coculture with IgM. By cocapping experiments and studies utilizing CD7 (Hut-78) and CD7 + (HSB, Molt-4) T-cell lines, TsEEF actit'i(~, did not appear to involve interactions with other T-cell or non-T-cell surface receptor.,. These findings suggest a novel rolefor FcRgt-CD7 T-cell surface receptors in binding certain soluble T-cell facto~:~ that result in the inhibition of T-cell replication.

ABBREVIATIONS AMLR CM CTL EBV ELISA

FcR

autologous mixed lymphocyte reaction complete medium cytotoxic T lymphocyte Epstein-Barr virus enzyme-linked immunosorbent assay receptor for immunoglobulin Fc fragment

IL-1 MHC MLR MoAb PBL Ts

interleukin 1 major histocompatibility complex mixed lymphocyte reaction monoclonal antibody peripheral blood lymphocyte human suppressor T cells

From the Departments of Pathology, Microbiolog) and Immunology, and Surger> Duke Unirer.~i(yand VA Medical Centers, Durham, North Carilona. Address reprint requests to Dr. F. Sanfilippo, Box 3712, Duke University Medical Cenzer, Durham. NC 277•0. Received September26, 1988; revised December 14, 1988.

Human Immunology25, 87-102 (1989) © AmericanSocietyfor Histocompatibilityand Immunogenetics,1989

87 0198-8859/89/$3,50

88

M. Emara et ai

INTRODUCTION Fc receptors (FcR) are known to be present on a variety of immunocompetent cells [1-3] and appear to play a role in regulating their activity. For example, FcR are involved in antigen uptake and processing by macrophages, proliferation of B cells, release of lysosomal enzymes by neutrophils, and release of mediators by basophils, mast cells, and platelets [2,3]. These FcR-dependent functions can be inhibited by various factors, including soluble Fc fragments [2]. In contrast, the role of FcR and their tigands in regulating T-cell responses remains unclear despite substantial work in this area [4-12]. In most studies of FcR-mediated effects on T-cell function, the regulation of humoral responses has been examined. For example, the suppressive activity of certain T lymphocytes was initially associated with the presence of lgG-FcR (FcR-/), whereas T cells possessing IgM-FcR (FcR/x) were found to enhance in vitro B-cell proliferation and immunoglobulin (Ig) synthesis [4,7-9]. In addition, the nature of FcR/x and whether it exists as a separate entity or as a part of other T-cell surface receptors such as CD7 remain controversial [13-15]. We have recently examined human suppressor T-cell lines (TsEE) induced from the peripheral blood lymphocytes (PBL) of a renal allograft recipient by autologous mixed lymphocyte reaction (AMLR) with an antidonor alloreactive line derived from the patient's own allograft [16]. These Ts lines exhibit major histocompatibility complex (MHC)-restricted, but nonspecific suppression of in vitro allogeneic and mitogen T-cell responses [16]. We have also functionally characterized a soluble factor (TsEEF) derived from these suppressor T-cell lines and shown that this factor inhibits the generation of in vitro allogeneic mixed lymphocyte reaction (MLR) and cytotoxic T lymphocyte (CTL) responses in a non-MHC-restricted manner. TsEE cells and TsEEF have been shown to be nontoxic, not to shift MLR kinetics or interfere with the utilizations of interleukin 1 (IL-1) or interleukin 2 (IL-2) by T cells, and to be effective only during the early phase of proliferative T-cell responses [ 16,17 ]. TsEEF activity has also been found to be dose-dependent, adsorbed by PBL, T-cell-specific, free of tumor necrosis factor (TNF) and interferon-'y (IFN-~) activity [17], and effective only on PBL and T-cell lines that express high levels of CD7 [ 18], as identified by the 3A1 anti-CD7 monoclonal antibody (MoAb). In the study presented here, we have used human T-cell lines in addition to PBL to further elucidate the suppressor mechanism of TsEEF. The suppressive activity of TsEEF was found to be mediated by binding to FcR~ determinants expressed on CD7 + T cells, although TsEEF itself was not an IgM or IgG molecule. The suppressive activity of TsEEF was inhibited by coculture with IgM, while pretreatment with TsEEE blocked the binding of IgM and anti-CD7 (HuLy-m2) MoAb to suppressible T cells. In addition, pretreatment of T-cell lines with anti-CD7 antibodies or IgM resulted in the capping or blocking of FcR/z-CD7 determinants and significantly decreased their susceptibility to TsEEF suppression.

MATERIALS A N D METHODS Media. Cells were cultured in complete medium (CM) consisting of RPMI 1640 (Grand Island Biological Co., Grand Island, NY) supplemented with 2.5c~ sodium bicarbonate, 100 U/ml penicillin, and 100 tzg/ml streptomycin, 2.0 mM L-glutamine, 0.02 M Hepes, and 10% heat-inactivated fetal calf serum (FCS). CM supplemented with 20% FCS was used for growth of Epstein-Barr Virus (EBV)-transformed B-cell cultures.

Suppressor T-Cell Factor Interaction with FcR~

89

TsEEF production. TsEEF was prepared as previously described [17]. Briefly, TsEE cells (5 x 104) were cocultured in vitro with an equal number of the autologous irradiated (3000 rad) inducer cells (the allograft-derived T-cell line EE-1) in 200 ~l CM per well of 96-well round-bottomed microtiter plates. After 48 hr of incubation at 37°C in 5% CO2, cell-free supernatants were filter-sterilized and stored at -70°C until use. Control supernatants were obtained from each of the cells used to generate TsEE cells in AMLR (i.e., EE-PBL and EE-1 T-cell line) or from a 48-hr coculture of EE-PBL and irradiated EE-1 cells. These control T-cell supernatants showed no suppressive activity when compared with control medium [17]. Human cell culture and suppression. Suppression of human T-cell lines HSB and Molt-4 and B-cell lines SB [19,20] and EBV-transformed JC lymphocytes (JC-EBV) was measured by culturing 5 x l04 cells/well in 0.3 ml CM in 96-well round-bottomed plates with or without TsEEF added at 50% by volume and incubating for 4 days at 37°C in 5% CO2. Culture wells were pulsed with 1 /zCi/well of [3H]thymidine 18 hr before harvesting. For measuring suppression of mitogen-induced proliferation, responder PBL (2 x 105 per well unless indicated otherwise) were stimulated in 96-well round-bottomed plates with phytohemagglutinin (PHA) at 2~g/106 cells in 0.3 ml CM with or without TsEEF added (33% by volume). Proliferation of triplicate cultures was measured by [3H]thymidine uptake on day 4. The proliferation of cell cultures containing TsEEF was compared with those cultured in the absence of TsEEF (controls), and the percentage of suppression was calculated by the formula: [ 1 - (experimental response/control response)] x 100%.

Immunofluorescence labeling andflow cytometry. MoAbs of Leu series 4, 3a, 1, 2a, 11, 15, and 18 (Becton-Dickinson, Oxnard, CA) were used to identify T-cell antigens CD3, CD4, CD5, CD8, CD16, CD11, and CD45R, respectively. Other MoAbs used included T l l (anti-CD2, Ortho Diagnostics, Raritan, NJ), Tac (anti-IL-2R provided by Dr. T. Waldmann, N.I.H.), Leu7 (anti-human NK1 marker) and anti-transferrin receptor (Becton-Dickinson), T29/33 (anti-human leukocyte common antigen, CD45, Hybritech, Inc., LaJolla, CA), 4F2, a MoAb against a cell marker found on T- and B-cell lines and hematopoietic and nonhematopoietic cells [13], and W6/32 ~anti-HLA class I, Bioproducts for Science, Inc., Indianapolis, IN). The anti-CD7 MoAbs 3A1 [13] and HuLy-m2 [21] of IgG2 subclass were provided by Drs. B. Haynes of Duke University and I.F. McKenzie of University of Melbourne, respectively. DUPAN 3 (anti-pancreatic tumor antigen, IgG2 subclass) was provided by Dr. R. Metzgar, Duke University, and used as a negative control MoAb [22]. Fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse or rabbit anti-human IgG were obtained from Tago and Assoc., Burlingame, CA, and Behring Diagnostics, Somerville, NJ, respectively. Cell populations were stained by indirect immunofluorescence as described elsewhere [23]. Briefly, 1 x 105 cells were incubated on ice for 45 min with 100/zl of MoAb at previously determined optimal final concentrations in phosphate buffered saline (PBS) supplemented with 10% FCS and 0.02% sodium azide. After three washes, cells were incubated with 50/~l FITC-conjugated goat anti-mouse secondary antibody (1 : 100) for an additional 45 min at 4°C in the dark. Labeled cells were then washed twice with PBS and subjected to flow cytometric analysis with a cytofluorograf 50H (Ortho Diagnostic Systems). Results were expressed as relative mean fluorescence intensity, or as the percent positive cell staining relative to the negative control MoAb (DUPAN 3).

90

M. Emaza eta],

Ig preparation and purification. Two euglobulin precipitates were prepared from normal human sera (5 ml) and sera pooled from three patients with demonstrated IgM paraproteins. The euglobulin precipitate was prepared by dialysis of sera against distilled water for 18 hr at 4°C, separated by centrifugation (final concentration approximately 1 mg/ml), and was used as such or subjected co further purification by resuspension in PBS and separation on a 1.5 .~ 65 cn-~ sephacryl 300 (Pharmacia) column. Two-ml fractions were collected and tested for IgM and IgG by a "capture" enzyme-linked immunosorbent assay (ELISA~ [24], in which the fractions were incubated in wells that had been coated with monoclonal anti-human Fc-/x or Fc-y (clones HB 57 and HB 43, respectively, American Type Culture Collection [ATCC], Rockville, MD). The bound lgs were detected with biotin-conjugated anti-K and anti-~ MoAbs (Atlantic Antibodies, ME) followed by streptavidin-coupled horseradish peroxidase (Zymed, CA } and O-phenyl diamine. The first two fractions of the IgM peak, which contained 0.2 mg/ml IgM (estimated by ELISA using paraprotein standards) with nc) detectable amounts of IgG, were pooled for use as IgM. Two fractions following the IgM peak that contained 0.2 mg/ml IgG and no detectable IgM were also pooled and used as IgG controls.

Preparation oflg-coated bead;. Human anti-Fc-/~, Fc-~/and Fab-/~ MoAbs of IgG~ subclass (HB 57, 43, and 138 respectively, ATCC), as well as purified lgM and IgG fractions, were coupled to Affi-Gel 10 (Bio-Rad, Rockville Centre, NY) using the techniques recommended by the manufacturer. Successful coupling of MoAb and Ig fractions to the Affi-Gel beads was confirmed by immunofluorescence labeling. The ability of anti-Ig coupled beads to adsorb Ig from control preparations was confirmed by ELISA using the above MoAbs in addition to two different polyvatent antibodies to human sera as controls.

Blocking and capping of cell surface receptors. CD7 molecules were blocked and capped by adding an excessive amount (1.0 ml, 1 : 50 dilution of ascites) ot anti-CD7 (3A1 or HuLy-m2 MoAb), or human IgM (1.0 mg/ml), to T-cell lines (HSB or Molt-4) or to PBL (1 x 106/ml) for 18 hr at 37°C and 5~¢ CO2. DUPAN 3 was used as a negative control MoAb. For appropriate assays, TsEEF or medium was added (50% by volume), and the incubation was continued for an additional 2 hr. Cell proliferation was tested by cell culture assays, or cells wer~* washed twice, stained by the indicated MoAb, and subjected to flow cytometric assay. Capping efficiency was judged by the high level of reduction in cell surface expression (60%-80% reduction) due to capping. Other cell surface markers (CD2, IL-2R) were similarly blocked and capped using predetermined optimum concentration (1:10 dilutions) of T l l and Tac MoAbs, respectively.

Statistical analysis. Comparisons between groups included computer programs for analysis of variance (ANOVA), Student's t-test, and Bonferroni (DUNN) t-test from SAS (Cary, NC). RESULTS

TsEEF Interacts with CD7 + T Cells Staining PBL and HSB or Molt-4 T-cell lines [19,20], which are suppressible by TsEEF [17], using a panel of MoAbs against T-cell markers demonstrated that these cells express high levels of CD7 and that the HSB line in particular expressed significant CD7, but low levels of CD2, and virtually no CD3, CD4.

Suppressor T-Cell Factor Interaction with FcR/.t

Preincubotion

91

3H- Thymidine Incorporation CPM ~_SEM(x10 -3) HSB line

Hours

mAb

TsEEF

2 i

_

m

--

+

18

DUPAN3

18

Tll

18

3A1

4 ~

_

6

8

10

12

2

4

i

J

L

J

i

i

Molt- 4 line 6 8 i___

t

10

12

L

h

14(74) ~-

(52)

*

) +

)(59) *

)(68)*

) -F

l(48)*

~1(64) *

)

m

+

}4(OO)

l-i(2]) Ns

FIGURE 1 TsEEF activity is inhibited by capping of CD7 but not CD2 receptors. T-cell lines (HSB and Molt-4) were preincubated in CM (uncapped), T l l (anti-CD2), 3A1 (anti-CD7), or DUPAN 3 (negative control MoAb) for 18 hr (capped cells) prior to 4 days of in vitro culturing (5 × 104/200 td) in the absence (controls) or the presence of TsEEF added at 50% by volume. Values in parentheses represent percent suppression of T-cell proliferation with TsEEF treatment relative to controls. *p < 0.05 and not significant (NS), p > 0.05.

CDS, CD8, CD11, or CD16 [18,20]. Furthermore, capping T-cell lines with anti-CD7 (3A1) MoAb abolished their susceptibility to the suppessive activity of TsEEF against their proliferation, suggesting a possible involvement of CD7 determinants in TsEEF-mediated suppression [18]. However, T-cell lines that had been incubated with or without TsEEF (1 × 106/ml) for 4 hr in the cold exhibited similar high fluorescence intensity with 3A1 (anti-CD7), indicating that TsEEF did not appear to block 3A1 binding to CD7 cell surface marker [18]. To examine possible secondary effects of pretreatment with 3A1 MoAb, cocapping experiments were performed using 3A1 and T l l (anti-CD2) MoAb. Preincubation of HSB and Molt-4 cells with 3A1 substantially blocked the suppressive effect of TsEEF, while preincubation with T11 or a negative control MoAb ( D U P A N 3) had no significant effect on TsEEF mediated suppression (Figure 1). In parallel experiments the possibility of CD7-CD2 cocapping was examined by pretreating Molt-4 cells (CD2 +, CD7 +) with either D U P A N 3 (control) or 3A1 MoAb and then performing cytofluorimetric analysis with either 3A1 or T l l MoAb. As shown in Figure 2, pretreatment with 3A1 MoAb significantly reduced binding of 3A1, but had no effect on the binding of T11 MoAb when compared to control cells that were pretreated with D U P A N 3 MoAb. Similar capping experiments with HSB cells using 3A1 and the W6/32 MoAb against a monomorphic HLA class I antigen, which is strongly expressed on HSB [13,20], gave similar results (data not shown). T o further investigate possible cocapping effects by 3A1 (anti-CD7) MoAb with other cell surface markers expressed on HSB cells, HSB cells were preincubated for 18 hr at 37 ° and 5% CO2 with D U P A N 3 (negative control MoAb) or 3A1 prior to staining with a panel of MoAb (see Materials and Methods) having various specificities. As shown in Table 1, the pretreatment of H S B cells with 3A1 (anti-CDT) MoAb had no effect on cell staining with any of the MoAb tested other than itself (3A1) and T29/33 (anti-leukocyte common

92

M: Emara et at.

Effect of3A1 Pretreatment on CD7 and CD2 Expression

2001

3AI (66.6%)

t

Li ._

Dupon 3 - Pretreated

E z -$

I

200

TI I (95.4%)

l!

--

Dupon 3 - Pretreoted

-

TII (97. 1%)

3A~ (~7 8 % )

\

I000 100

l oo

3AI-Capped

100o

5AI-Capped

Relative Fluorescence Intensity FIGURE 2 Molt-4 T-cell lines pretreated with 3A1 or D U P A N 3 were stained with anti-CD2 (Tll) and CD7 (3A1) MoAb (solid lines). Values in parentheses represen~ percent positive cell staining relative to D U P A N 3 as controls (dashed lines).

TABLE 1

T h e effect o f capping H S B T cells with anti-CD7 M o A b on expression o f o t h e r cell markers Mean fluorescence intensity after preincubation with

Cell staining with MoAb

Specificity

DUPAN 3

5AI

DUPAN 3 3A1 T29/33 Anti-transferrin R 4F2 Leu 7 W6/32 Leu 11 Leu 4

(Negative control) CD7 CD45 Transferrin R Leukocyte HNK1 H L A class I NK CD3

48 213 197 208 722 287 574 46 45

28 46 178y 48 (76~ 167 (20) 734 (0) 238 ([~) 651 ~0) 2(~ 25

" Values in parentheses indicate the percent decrease in staining relative to DUPAN 3 pretreatment,

Suppressor T-Cell Factor Interaction with FcR/.~ TABLE 2

93

Expression of CD7 and CD45 on test cells Relative mean fluorescence intensity ~ of cells pretreated with the indicated MoAb HSB (CD7 +)

Hut-78 ( C D 7 )

Cells stained with MoAb

DUPAN 3

3AI

DUPAN 3

DUPAN 3 3AI T29/33 Anti-transferrin R

20 118 190 208

11 13 92 167

9 27 70 ND

JC-EBV ( C D 7 ) DUPAN 20 28 260 93

a Cell lines preincubated with DUPAN 3 (negative control) or 3AI MoAb (10c"cells/ml MoAb) I8 hr at 37°C and 5~ CO2 prior to staining.

antigen, CD45), which were both significantly reduced. A minor reduction in anti-transferrin receptor staining was also seen, although this reduction was not significant on repeat testing (Table 2). The possible involvement of CD45 in TsEEF-mediated suppression was examined by four different approaches. First, TsEEF was unable to suppress the proliferation of the CD7 Hut-78 T-cell line [13] in each of three different experiments (data not shown). Second, preincubation of the CD7 + HSB T-cell line with T29/33 MoAb for 18 hr at 37°C prior to culturing with or without TsEEF for 4 days did not affect the level of suppression (71% suppression) observed relative to ceils treated with D U P A N 3 control MoAb. Third, pretreatment of the HSB T-cell line with negative control MoAb D U P A N 3 or T29/33 MoAb for 18 hr at 37°C prior to staining with anti-CD7 3A1 MoAb did not appear to reduce the level of CD7 expression as measured by flow cytometry (data not shown). Finally, the JC-EBV transformed cell line [17] and Hut-78 T-cell line that were not suppressed by TsEEF were found to be C D 7 - and CD45 + as shown in Table 2. Thus, the inhibitory activity of TsEEF on T cells appeared to be associated with the expression of CD7 and not CD45. Similarly, the possible involvement, of CD45R (Leul8) in TsEEFmediated suppression was excluded by the repeated demonstration that HSB T-cell lines were Leu 18-negative (data not shown). T o examine the possibility that low levels of surface IL-2R expressed on HSB or Molt-4 might be of importance in the action of TsEEF, both lines were preincubated with D U P A N 3 (control), 3A1, or anti-IL-2R (Tac) MoAb, and ceils were then stained for CD7 cell marker expression or cultured in the presence or absence of TsEEF for an additional 3 days. As shown in Table 3, cells that were preincubated (capped) with 3A1 exhibited significantly less staining ( 4 3 % - 6 0 % reduction, p < 0.05) by 3A 1, while preincubation of either cell line with Tac did not affect the high expression of CD7 detected by 3A1 MoAb. Proliferation of both D U P A N 3 (control) and Tac pretreated cell lines was significantly suppressed by TsEEF ( 5 4 % - 6 7 % suppression, p < 0.05), whereas capping o f either cell line with 3A1 MoAb abolished the suppressive activity of TsEEF on cell proliferation.

TsEEF Activity Involves Interactions with IgM-Fc Receptor T o identify potential interactions between TsEEF and CD7-FcR/,, 3A 1 MoAb or human IgM fractions (at 1.0 or 0.2 mg/ml) were added to HSB cells at day 0 with

94

M. Emara et at.

TABLE 3

The effect of pretreatment with anti-CD7 (3A1) or anti-IL2 receptor M o A b on 3A1 expression and TsEEF activity

M o A b used in pretreatment DUPAN 3A1 Tac DUPAN 3A1 Tac

3

3

T-cell lines"

Relative mean fluorescence intensity o f 3 A I

Molt-4 Molt-4 Molt-4 HSB HSB HSB

208.1 65.1 161.2 392.5 222.9 393.1

~ H - T h y m i d i n e u p t a k e ( C P M ± SEMI upon culture with Medium 10,997 674v 5792 0417 6860 12,572

± ± ± ± ± ±

TsEEF ~

2966 84 206 744 994 673

5066 532~i 2840 2167 5810 4184

:t: { I 0 3 (5 <<) ~ I i 7 121) "~:~ ± 271 ( 5 1 ) L 2 2 6 (66~, ~ ± 9 v l {[S~:':: ~- ~3 (67i'

Background staining of Molt-4 and HSB T-cell lines with D U P A N 3 MoAb (negative controls,~ was equal m m(, ~ fluorescence of 75 and 701 respectively b Values in parentheses represent percent suppression of cell proliferation by TsEEF relative to ~(mm)[s (me,tium~ ' p < 0.05; NS = not significant (p > 0.05).

or without TsEEF (1 : 1 volume), and then cultured for 4 days. As shown in Figur~~ 3A, PBL stimulated with P H A in the presence of TsEEF alone showed significant suppression in all groups ( 5 1 % - 6 7 % suppression, p < 0.05) relative to thos~ that were stimulated in the absence of TsEEF Icontrol cultures), whereas neither IgM (1 mg/ml) nor 3A1 alone suppressed PHA-induced proliferation. However. the suppression of PBL responsiveness to mitogen by TsEEF was completely abrogated by the simultaneous addition to IgM. In contrast, the simultaneous addition of 3A1 M o A b did not affect TsEEF-mediated suppression of P H A induced proliferation ( 7 5 ~ suppression, p < 0.05), suggesting that 3A1 does no~ bind to the same determinant on CD7 as TsEEF (see below). As shown in Figure 3B, similar results were obtained when the HSB cell line was tested with TsEEF or IgM added alone or together (1 : 1 ratio) for 4 days. Parallel experiments demonstrated that lower concentrations (0.2 mg/ml) of IgM resulted in a 50C,: decrease of TsEEF-mediated suppression of T-cell proliferation Mata not s hown~

FIGURE 3 TsEEF activity is diminished by coculture with IgM. IgM or 3Ai MoAb was either added at 33% by volume alone or together with TsEEF (1 : 1 v/v) to PBL (1 > 105/well) stimulated with PHA (A) or added at 40% by volume with or without TsEEF (1 : 1 v/v) to HSB cell line cultures (5 × 104/well), as shown in (B). Cell cu Itures in the Presence of: TsEEF IgM 3AI

3 H-Thymidine Incorporotion CPM't_SEM

A. PHA-ActivatedPBL 50

I00

J

hC51) * _.

44+

I

+

+

150

200

i

J

(x10"3) IB HSB-Line

E

4

6

8

10

12

I

=

I

I

i

I

~(s2). ~(01)NS

H

~(14)NS

~(67)~

~1(79) * ~(IO)NS

+

4-

~(75)*

~(73),

14 _J

Suppressor T-Cell Factor Interaction with FcR/_t

TsEEF added (50% volume toT cell lines

95

5Hqhymidine incorporation CPM*_.SEM(xlO5 ) 2 5 4 5 6 I

None (medium) Unabsorbed Anti-Fc#-absorbed Anti-Fc Z,- absorbed Anti-Fab#-obsorbed None(medium) Unabsorbed Anti-Fc ju- o bsor bed Anti-Fc ~ - absorbed Anti-Fab#-absorbed

r

I

~ ( 2 (91)

)~

I

HSB

(89), 88).

__,~(52) * ~(58). [--I (46) -~

t4 Molt. 4

I--4(43)*

FIGURE 4 TsEEF is not IgM or IgG. TsEEF was preincubated 4 hr in the cold with beads coated with anti-Fc-/*, Fc-% or Fab-/, MoAbs prior to addition at day 0 (50~ by volume' to HSB and Molt-4 T-cell lines (5 x 1 0 4 cells/well) cultured for 4 days.

To investigate whether TsEEF was in fact IgM, beads coated with MoAbs against various human Ig fragments were prepared and used in absorption studies of TsEEF. As shown in Figure 4, TsEEF unabsorbed or preabsorbed with beads coated with anti-Fc-/z, Fc-y, or Fab-/* yielded significant and comparable suppression of HSB or Molt-4 T-cell line proliferation. The absence of Ig in TsEEF was further confirmed by coating plates with unabsorbed TsEEF or TsEEF absorbed with Fc-/z, Fc-y, and Fab-/z MoAbs, and testing by ELISA for the presence of Ig using MoAbs to human Fc-/z and Fab-~. N o detectable binding of anti-human Fc-/~ and Fab-~ to absorbed or unabsorbed TsEEF was seen, whereas control plates coated with Ig demonstrated appropriate binding (data not shown). To determine whether the observed inhibition of TsEEF activity by IgM was the result o f a direct interaction between these molecules or their competition for cell binding, TsEEF was preabsorbed with IgM- or IgG-coated beads and tested for its suppressive capacity. As shown in Figure 5, T cells from the Molt-4 line that were cultured for 4 days in CM, or with either IgG or IgM alone showed comparable levels of proliferation. Adding unabsorbed TsEEF either alone or together with IgG resulted in comparable suppression (50% and 54@, respectively, p < 0.05), while the addition of IgM with unabsorbed TsEEF resulted in marked reduction of TsEEF-mediated suppression to a nonsignificant level ( 2 6 ~ , p > 0.05). The addition of TsEEF that had been preabsorbed with IgG or IgM resulted in suppression (53% and 48% respectively) comparable to that seen with unabsorbed TsEEF (50%). These results indicate that the inhibition of TsEEF activity by IgM was not due to the binding of TsEEF to IgM, but rather suggest that TsEEF and IgM (but not IgG) compete for binding to the same determinants (i.e., FcRbt) on suppressible target T cells, or that pretreatment with IgM results in a conformational change that reduces binding of TsEEF. CD7 (FcR~t) i n t e r a c t i o n s w i t h T s E E F To further examine the involvement of CD7 and FcR/z in TsEEF-mediated suppression, we tested the effect of pretreating PBL with either anti-CD7

96

M. Emara et ai,

Molt 4 cultured 5H.Thymidine incorporotion CPM t SEM(xlO-3) 2 4 6 8 10 12 14 4days with: I

Meclium (control) TsEEF IgG (control) IgG +TsEEF IgG-Abs.TsEEF IgM (control) IgM+ TsEEF IgM -Abs.TsEEF

I

I

I

1

I

I

)(5o), (54), (53). b (z6) NS 1)(ae).

FIGURE 5 TsEEF activity is not absorbed by IgG or IgM. Molt-4 cells (5 x 10' weU) were cultured in 0.3 ml CM with or without TsEEF (1 : 2 dilution v/v) alone, with IgG or IgM added alone at 33% by volume, with a mixture of TsEEF and either IgG or IgM (66%, 1 : 1 ratio), or with TsEEF that had been preabsorbed (Abs) with beads coated with IgG or IgM. Values in parentheses represent percent suppression of cell proliferation in culture containing TsEEF relative to their appropriate control (without TsEEF).

(HuLy-m2) or IgM prior to in vitro stimulation with P H A in the absence (control) or the presence o f TsEEF. As shown in Figure 6, the addition o f TsEEF to either untreated or control M o A b ( D U P A N 3) pretreated PBL resulted m significant suppression o f P H A proliferative responses relative to those cultured in the absence o f TsEEF (medium control). In contrast, pretreating PBL with either IgM or the anti-CD7 M o A b (HuLy-m2) prior to addition of TsEEF in culture significantly reduced the suppressive activity of TsEEF. Control testing of PBL and T-cell lines showed no effect of treatment with 3A1, HuLy-m2, or IgM alone on cell proliferation. Although pretreatment by HuLy-m2 and human IgM only caused a 53 % reduction in IgM binding as measured by flow cytometry Idata not shown), this was sufficient to completely block the suppressive acnvlty of TsEEF. To evaluate whether the abrogation of TsEEF suppressive actiwty by p r e t r e a t m e n t with HuLy-m2 M o A b or IgM was due to their binding to the TsEEF binding site, the ability o f TsEEF to block the binding of anti-CD7 M o A b or IgM to PBL was examined. For this purpose, PBL were incubated with or without TsEEF (1 x 106 ceU/1 ml TsEEF) for 4 hr in the cold prior to staaning with anti-CD7 (3A1 or HuLy-m2), IgM, or D U P A N 3 (neganve control MoAb~. As shown in Figure 7, PBL that were incubated with TsEEF for 4 hr in the cold had a decreased capacity to bind either IgM or HuLy-m2 M o A b ~60~: and 71¢7/~reduction, respectively) when compared to PBL that were preincubated with m e d i u m control. In addition. TsEEF was not capable o f blocking the binding of 3A1 (anti-CD7) to PBL, suggesting that 3A1 and TsEEF appear to bind to different epitopes, confirming our previous observation [18]. Moreover, cell labeling studies demonstrated that preincubation of H S B and Molt-4 T-cell lines with TsEEF substantially blocked the binding of IgM ( 4 2 % - 4 4 % reduction~ but

Suppressor T-Cell Factor Interaction with FcR/~

3H-Thymidine Incorporation CPM +_.SEM (x I0 -3)

PHA-Stimulation of PBL Pretreated with: Antibody

TsEEF

None

97

2

4

6

8

I

I

|

I

(63)

]

16 H

l

4

*

m

Dupon

+

(55) *

H (NS)

w

Huly- M2

-t-

D

IgM

+

H

(NS)

FIGURE 6 TsEEF interacts with FcRbt (CD7) receptors. EE-PBL (2 × 104/well) were preincubated (1 × 106/ml) with medium (control), DUPAN 3 (negative control MoAb), HuLy-m2 (anti-CD7) MoAb, or human IgM for 18 hr prior to in vitro stimulation by PHA (2 /,g/106) with or without (control) TsEEF added (50% by volume). Values in parentheses represent percent suppression at day 4 of cultures treated with TsEEF.

had no effect on IgG binding (data not shown). These results suggest that TsEEF interacts with CD7 epitopes recognized by HuLy-m2 MoAb and IgM but not 3A1. DISCUSSION We have previously propagated human suppressor T-cell lines (TsEE) from fresh PBL of a kidney transplant recipient by coculturing PBL with an autologous alloreactive T-cell line (EE-1) grown from a needle biopsy of the patient's own renal allograft [ 16]. These TsEE cells were capable of inhibiting the generation of MLR and CTL responses of autologous or nonautologous responder T cells that shared class I HLA-B7 with TsEE cells regardless of the stimulatory phenotypes [16]. In additional studies [17], we found that in vitro activation of TsEE cells with the inducer CTL line (EE-1) yielded soluble T-cell factors (TsEEF) that were inhibitory for MLR, CTL, as well as mitogen-induced T-cell responses. Although TsEEF activity was non-HLA-restricted, it was similar to that of TsEE cells in that it was nontoxic and did not appear to shift MLR kinetics or interfere with IL-1 or IL-2 utilization, and was effective only when added during the early phase of T-cell responses. In addition, TsEEF activity was shown to be dose-dependent, T-cell-specific, and able to be absorbed by PBL [ 17]. Labeling of PBL and T-cell clones that were suppressible with TsEEF using a panel of MoAb against T-cell surface markers showed that these cells all expressed high levels of CD7 [ 18]. In agreement with other reports [19,20], CD7 was the only T-cell marker significantly expressed on the HSB T-cell line, raising the possibility that this determinant might be involved in TsEEF-mediated suppression [ 18]. In this report, we examined the mechanism by which TsEEF inhibits T-cell proliferation utilizing T lymphoblastoid cell lines (HSB and Molt-4) in addition

98

M. Emara et al.

PBL etreo d with Medium TsEEF 100

3AI(45.9%)

0k l .Q 100 E

I M2 (45.9%)

Z U

r~lY i M2.(13.4%)

I 0 100 lgM(44:8%) ~ ( i 7.9%) ~ I

0~ "

I

400

I

400

Re Iotive Fluorescence Intensity FIGURE 7 TsEEF effects on the binding of lgM and anti-CD7 antibodies, ] x 10" SG-PBL were preincubated with 1 ml of medium or TsEEF for 4 hr in the cold, washed twice, and stained with either IgM or anti-CD7 (solid lines) MoAb. Values in parentheses represent percent of test MoAb-positive staining relative to control MoAb DUPAN (dashed lines).

to PBL that are suppressed by TsEEF [17]. The CD7 antigen is a human cell surface glycoprotein ( 3 7 - 4 0 kd) expressed on early thymocytes and mature T lymphocytes and is recognized [13,21,25,26] by several M o A b (e.g., 3A1, 4A WT1, HuLy-m2, RFT2), some of which bind to the same epitopes (3A1 and WT1) or different epitopes (3A1 and HuLy-m2) on CD7. Furthermore. a recent report [14] has suggested that CD7 may be part of the FcR/z on T cells. These findings raised the possibility that TsEEF-mediated suppression of T-celt proliferative responses may also involve interactions with the FcR/z. Results from competitive inhibition studies using TsEEF and human IgM preparations or 3A 1 M o A b indicate that suppression o f CD7 * T lymphpoblastoid cell lines can be significantly diminished in the presence of IgM but not 3A1. Moreover, p r e t r e a t m e n t of cells with anti-CD7 (3A1 or HuLy-m2) M o A b or IgM, respectively, abrogated the suppressive acuvlty of TsEEF on T-cell proliferation. Although the exposure of T-cell lines to TsEEF prior to staining with 3Al revealed no blocking of 3A1 anti-CD7 antibody binding by TsEEF [18], it ,s

Suppressor T-Cell Factor Interaction with FcR>

99

possible that TsEEF binds to a CD7 epitope other than that recognized by 3A 1, and may deliver its inhibitory signal without blocking the binding of 3A1 MoAb. This possibility is strengthened by our observation that HuLy-m2, another MoAb that has been shown to block IgM binding to T cells [21], is blocked from binding to T cells that were pretreated with TsEEF. In contrast, 3A1 binding to TsEE-pretreated cells was not significantly reduced. These findings suggest that TsEEF binds to FcR/2, which functionally represents a determinant on the CD7 receptor identified by HuLy-m2 but not 3A1, and that capping of the CD7 receptor by pretreatment with 3A1 or HuLy-m2 abrogates the suppressive effect of TsEEF. Although TsEEF is a product of functionally active T cells [ 17], the possibility of its being an IgM molecule was ruled out using two different techniques. Results from these experiments indicated that TsEEF activity was not absorbed by anti-Fc-/~ or Fab-/, MoAb-coated beads, nor could IgM-specific determinants be detected in TsEEF by direct ELISA. Since the T11 (CD2) T-cell surface receptor has been shown to be involved in T-cell activation [27,28], the role of this marker in mediating TsEEF-dependent suppression was also examined. Studies of in vitro proliferation of the Molt-4 T-cell line that was pretreated with anti-CD2 MoAb (T11) and then cultured with or without TsEEF revealed no effect on the suppressive activity of TsEEF. Furthermore, in agreement with other reports [21], immunofluorescence studies of Molt-4 T cells that were pretreated with either D U P A N 3 or anti-CD7 (3A 1) and stained with 3A1 or T11 MoAb demonstrate that CD2 does not cocap with CD7. These results indicate that the CD2 receptor does not appear to be involved in suppression mediated by TsEEF, and that the abrogation of TsEEFmediated suppression by capping off CD7 is a specific rather than a nonspecific consequence o f capping. Although recent reports indicate that CD7 and CD3 may [29] or may not [21] cocap, the absence of CD3 on the HSB line [ 1 8 - 2 0 ] indicates that the suppressive effect of TsEEF does not involve a coordinate interaction with CD3. Further examination of the possible cocapping of CD7 with other cell surface receptors revealed that with the exception of T29/33 (leukocyte common antigen, CD45), no other marker of the tested panel was significantly affected by 3A 1 pretreatment. Although CD45 appeared to cocap with CD7, its involvement in TsEEF-mediated suppression was considered unlikely since: (1) T29/33 is expressed on B- and T-cell lines that were not suppressed by the factor; (2/ preincubation of the HSB CD7 + T-cell line with T29/33 for 18 hr prior to culturing with TsEEF did not abrogate TsEEF-mediated suppression, and (3) repeated testing of TsEEF activity against the Hut-78 T-cell line which is CD7 but express CD45 showed no suppressive activity against the proliferation of this line. While previous reports [ 4 - 8 ] have defined a functional role of FcR/z in regulating B-cell responses, our findings are the first to implicate FcR/, (CD7) cell surface receptors in the down-regulation of T-cell responses. In this regard, TsEEF appears to function differently than many other T suppressor factors that mediate suppression of T-cell responses [30-35]. These findings also suggest that T-cell factors that mediate suppression via interactions with CD7 may be the product of only certain types of Ts cells, or are common Ts-cell products whose specificity is masked by the concomitant production of other nonspecific suppressor factors. Our own testing of unfractionated supernatants from other human Ts cells derived in MLR as controls for TsEE [17], as well as from primary Ts-cell lines derived from human liver allograft biopsies (Emara et al., manuscript in preparation), has not demonstrated CD7 functional specificity. Studies exam-

100

M. Emara et ai. ining fractionated supernatants from these and other Ts cells are in progress. Nevertheless, the suppressor T cells (TsEE) identified in this study appear to inhibit target cell proliferation via soluble factors (TsEEF) that bind directly to FcR/z on CD7 ÷ T lymphocytes, suggesting a novel role for this T-cell receptor molecule.

ACKNOWLEDGMENTS

We thank Dr. B. Haynes (Duke University) for providing 3A1 MoAb and Drs. M.,.',. San&in and I.F.C. McKenzie {University of Melbourne) for providing HuLy-m2 MoAb The authors express thanks to Drs. P. Cresswell and D.N. Howell (Duke University> fi)r helpful comments, Mr. R.J. Scibilia for help in the statistical analysis, and S. Gornto for assistance in manuscript preparation.

REFERENCES 1. Diamond B, Boccumini L, Birshtein BK: Site of binding of IgG2b and lgG2a by mouse macrophage Fc receptors by using cyanogen bromide fragments. J lmmuno! 134:1080, 1985. 2. Morgan EL, Weigte WO: Biological activities residing in the Fc region of immunoglobulin. Adv Immunol 40:61, 1987. 3. Mathur A, Lynch RG, Kohter G: The contribution of constant region domains to the binding of murine IgM to Fcbt-receptors on T cells. J Immunol 140:143, 1988. 4. Moretta L, Webb SR, Grossi CE, Lydyard PM, Cooper MD: Functional analysis ot two human T cell subpopulations: Help and suppression of B cell responses by T cells bearing receptors for IgM or IgG. J Exp Med 146:184, 1977. 5. Haynes BF, Fauci AS: Activation of human B lymphocytes. X. Heterogeneity ol concanavalin A generated suppressor cells of pokeweed mitogen-induced plaquvforming cell response of human peripheral blood B lymphocytes. J Immunol 121 :559, 1978. 6. Kolsch E, Oberbarnscheidt J, Bruner K, Heuer J: The Fc receptor: Its role in the transmission of differentiation signals. Immunol Rev 49:61, 1980. 7. Andersson B, Skoglund AC, Ronholm M, Lindstein T, Lamon EW, Collisson EW, Walia AS: Functional aspects of IgM and IgG Fc receptors on murine T lymphocytes. Immunol Rev 56:6, 1981. 8. Moretta L, Mingari MC, Moretta A, Fauci AS: Human lymphocyte surface markers. Semin Hematol 19:273, 1982. 9. Uher F, Puskas E, Gergely J, Bazin H: An IgM-producing immunocytoma induces large numbers of splenic T lymphocytes with Fc~ receptors. Immunology 61:32T. 1987. 10. Fridman WH, Guimenzanes A, Gisler RH: Characterization of a suppressive immunoglobulin-binding factor (IBF). I. Production of IBF by a ()-positive lymphoma (L-5178-Y). J Immunol 119:I266, 1977. 11. Fridman WH, Rabouridin-Combe C, Neauport-Sautes C, Gisler RH: Characterization and function o f T cell Fc receptor; Immunol Rev 56:51, 1981. 12. Daley MJ, Nakamura M, Gefter ML: Functional and biochemical characterization of secreted I-J (+) suppressor factor that binds to immunoglobulin. J Exp Med 163:1415, 1986. 13. Haynes BF: Human T lymphocyte antigens as defined by monoclonal antibodies. Immunol Rev 57:127, 1981.

Suppressor T-Cell Factor Interaction with FcR~t

101

14. Sandrin MS, Vaughan HA, Lovering KE, Curnow K, Deacon NJ, McKenzi IFC: Characterization of the gene coding for the human lymphocyte surface antigen HuLy-m2 (CD7). In McMichael AJ et al (eds): Leukocyte Typing IIl. Oxford, Oxford University Press, 1987. 15. Arnffo A, Seed B: Molecular cloning of two CD7 (T cell leukemia antigen) cDNAs by a Cos cell expression system. Embo J 6(11):3313, 1987. 16. Emara M, Miceli MC, Finn OJ, Sanfilippo F: Human suppressor T cells induced in vitro with an autologous renal allograft derived T cell line. I. suppressor cell induction, function and specificity. Hum Immunol 23:223, 1988. 17. Emara M, Sanfilippo F: Human suppressor T cells induced in vitro with an autologous renal allograft derived T cell line. II. Activity and specificity of a soluble suppressor factor. Transplantation 47:364, 1989. 18. Emara M, Finn OJ, Baldwin WM III, Sanfilippo F: Human suppressor T cells induced in vitro with an autologous renal allograft derived T cell line. III. Specificity and functional mechanisms of suppressor T cell factors derived from allograft-induced human T suppressor cells. Transplant Proc, in press. 19. Haynes BF: Summary of T cell studies performed during the second international workshop and conference on human leukocyte differentiation antigens. In Reinherz EL, et al (eds): Leukocyte Typing II. New York, Springer-Verlag, 1986. 20. Howell DN, Salter RD, Koppelman BD, Cresswell P: Expression of differentiation antigens by hybrids of human lymphoblastoid cells. Hum Immunol 17:443, 1986. 21. Thurlow PJ, Lovering KE, McKenzie IFC: A monoclonal antibody detecting a new human T cell antigen HuLy-m2. Transplantation 38:143, 1984. 22. Metzgar RS, Gaillard MT, Levine SJ, Tuck FL, Bossen EH, Borowitz MJ: Antigens of human pancreatic adenocarcinoma, cells defined by murine monoclonal antibodies. Cancer Res 42:601, 1982. 23. Beverley PCL, Callard RE: Distinctive functional characteristics of human T lymphocytes defined by E-rosetting or monoclonal anti-T cell antibody. Eur J Immunol 11:329, 1981. 24. Hiemstra PS, Baldwin WM, Van der Voort EAM, Paul LC, Van Es LA, Daha MR: Polymeric IgA antibody response to rabbit antithymocyte globulin in renal transplant recipients. Transplantation 45:701-705, 1988. 25. Tax WJM, Tidman N, Janossy G, Trejdosiewicz L, Willems R, Leeuwenberg J, DeWitte TJM, Capel PJA, Koene RAP: Monoclonal antibody (WTI) directed against a T cell surface glycoprotein: Characteristics and immunosuppressive activity. Clin Exp Immunol 55:427, 1984. 26. Raftery MJ, Lang CJ, Ivory K, Tidman N, Sweny P, Fernando ON, Moorhead JF, Janossy G: RFT2 (CD7) monoclonal antibody as prophylaxis against renal allograft rejection. Transplant Proc 17:2737, 1985. 27. Meuer SC, Hussey RE, Fabbi M, Fox D, Acuto O, Fitzgerald KA, Hodgdon JC, Protentis JP, Schlosman F, Reinherz EL: An alternative pathway ofT cell activation: A functional role for the 50kd T11 sheep erythrocyte receptor protein. Cell 36:897, 1984. 28. O'Flynn K, Krensky AM, Beverley PCL, Burakoff SJ, Linch DC: Phytohaemagglutinin activation ofT cells through the sheep red blood cell receptor. Nature (London), 313:686, 1985. 29. Sandrin MS, Vaughan HA, McKenzie IFC: XlI International Congress of the Transplantation Society, abstract, 1988, p 256. 30. Crosier PS, Simpson EJ, MacDonald PR: Regulation of human cytotoxic T lymphocyte responses by a soluble suppressor factor. Cell Immunol 95:54, 1985.

102

M. Emara et al

31. Maki T, Satomi S, Gotoh M, Monaco AP: Conrra-IL2; a suppressor lymphokine that inhibits IL2 activity. J Immunol 136:3298, 1986. 32. Chu W-S, Carpino MR, Dent A, Rich S: Suppressor T cell growth and differentiation. Production of a suppressor T cell differentiation factor by the routine thymom~ BW5147, 1987..] Immunol 138:78, 1987. 33. Sawada SG, Suzuki G, Kawase Y, Takaku F: Novel immunosuppressive agent FKS06. In vitro effects on the cloned T cell activation. J Immunol 139:1797, 1987~ 34. Germain RN, Benacerraf B: Helper and suppressor T cell factors. Semin lmmunopa.. thol 3:93, 1980. 35. Zembala M, Romano GC, Colizzi V, Little JA, Asherson GL: Nonspecihc 1 suppressor factor (nsTsF) cascade in contact sensitivity: nsTsF-1 causes and Ly 1 ' 2 I-A+ immune T cell to produce a second, genetically restricted nsTsF-2. J tmmunol 137:1138, 1986.