Eur. J . Immunol. 1975. 5: 27-32
M. Schlesinger and M. Chaouat Department of Experimental Medicine and Cancer Research, The Hebrew University - Hadassah Medical School, Jerusalem
Association of H-2 antigens and EAC receptors
27
The association of H-2 antigens and EAC receptors on the surface of peritoneal cells+ The exposure of murine peritoneal cells to anti-H-2 sera results in a diminished expression of H-2 antigen o n the cell surface. Concomitant with this “H-2 modulation” the capacity of macrophages t o bind sheep red blood cells coated with antibody and complement (EAC) was markedly diminished. In contrast, there was no change in the capacity of modulated macrophages to bind sheep red blood cells coated with antibody alone (EA). Antibodies to K end H-2 specificities were more effective in reducing the binding of EAC than antibodies to D end H-2 specificities. Exposure of peritoneal cells to 0 or Ly antisera had n o effect on the formation of EAC rosettes. Exposure o f peritoneal cells to anti-H-2 sera, under conditions which would not allow modulation of H-2 antigens, also prevented the reduction of EAC binding. Thus, the EAC receptors and H-2 antigenic specificities seem to be closely related on the surface of peritoneal cells, but constitute distinct cell surface structures. Preliminary evidence indicates that vinblastine, a microtubule depolymerizing agent, may disrupt the close association of EAC receptors and H-2 antigens. It is suggested that the association of EAC receptors and K end H-2 determinants on the membrane of macrophages may have implications for the regulation of the immune response by H-2-linked Ir genes. ~
1. Introduction
2. Materials and methods
The exposure of murine peritoneal cells to anti-H-2 sera, in the absence of complement (C), leads t o the disappearance of H-2 antigenicity from the cell surface of peritoneal macrophages [ 1-31. Following such “antigenic modulation” [ 3 , 41 murine macrophages are n o longer susceptible to the cytotoxic effect of H-2 antibodies and complement. Evidence has been presented that H-2 modulation probably results from the active pinocytosis of complexes of H-2 antibody along with cell surface H-2 antigenic determinants [ 1-31.
2.1. Mice
The cell surface of macrophages displays several distinctive receptors, which probably play an important role in the immunological functions of macrophages [ 51. Macrophages possess receptors for the Fc component of immunoglobulins [ 61. These receptors can be assayed by the formation of rosettes with sheep red blood cells (SRBC) coated with 7 S antibodies to SRBC (EA-rosettes) [7]. The membrane of macrophages also possesses receptors for activated C3, which can be assayed by testing the formation of rosettes with sensitized SRBC and C (EAC-rosettes) [S]. In the present study, modulation o f the H-2 antigenicity o f peritoneal macrophages was found to be accompanied by a concomitant disappearance of the EAC receptors from the surface of peritoneal macrophages. The nature of the association of H-2 antigenicity and EAC receptors was analyzed.
+ This study was supported by USPHS Research Agreement No. 06-048-1 and by a Lady Davis Endowment. Correspondence: Michael Schlesinger, Department of Experimental Medicine and Cancer Research, The Hebrew University - Hadassah Medical School, Jerusalem, Israel Abbreviations: C Complement EA: Sheep red blood cells coated with antibody EAC: Sheep red blood cells coated with antibody and complement NMS: Normal mouse serum SRBC Sheep red blood cells PBS: Phosphate buffered saline PC: Peritoneal cell(s) LNC: Lymph node cell(s)
Mice of the A/J, BALB/c, C3H and C57BL/6 inbred strains were used.
2.2. Antisera Alloantisera were elicited by repeated i.p. injections of spleen cell suspensions as previously described [8]. The following alloantisera were used: A/J anti-C57BL, A/J anti-BALB/c, DBA/2 anti-C3H, C3H anti-DBA/2, and C57BL anti-A sera. The cytotoxic titers obtained with various anti-H-2 sera in the present study are shown in Table 1 . AKR/J anti-C3H spleen serum was used as anti-O-C3H serum, whereas (AKR/l x C57BL)Fl anti-C3H thymus serum was used as anti-Ly A . 1 serum. Table 1. The cytotoxic effect of various antisera on allogeneic peritoneal and lymph node cells
A
Test cells BALB/c C3H C57BL/6 LNC PC LNC PC LNC PC LNC
Antiserum
PC
C57BLanti-A C3Hanti-DBAI2 DBA/Zanti-C3H A antiieALBlc A anti-CS7BL
64a) 128
16 64
32 128
32 32 128 128 8
32
32
64
64
32 64
64
a) Cytotoxic titers are expressed as the reciprocal of the highest serum dilution which still killed 50 % or more of the test cells Rabbit antisera against SRBC were elicited by i.p. injections of 1 ml of 25 % suspensions (v/v) of washed SRBC. The rabbits were bled from the marginal ear vein 5 days after the first inoculation in order t o obtain antisera containing mainly 19 S antibodies. Sera rich in 7 S antibodies were obtained from rabbits bled a week after the third or fourth injection o f SRBC. Purified 19 S antibody used in some experi-
28
M. Schlesinger and M. Chaouat
ments was obtained by collecting the excluded fraction of antisera filtered through a Sephadex G-200 column (Pharmacia, Uppsala) [ 91. 2.3. Cell suspensions Peritoneal cells were obtained from normal, unstimulated mice by repeated washings of their peritoneal cavities with saline [ 1 1 . The cell concentration was adjusted to 2 x 106/ml in saline. In some experiments, adherent cells were removed from the peritoneal cell suspension. Peritoneal cells were suspended in Hanks' basic salt solution, supplemented with 10 % fetal calf serum, and 1 .O ml of this suspension was delivered into Falcon plastic petri dishes, 45 mm in diameter. The petri dishes were incubated for 3 h at 37 "C in a humidified incubator containing 5 % CO,, and the nonadherent cells were collected carefully with a pipette. The concentration of the nonadherent cell population was adjusted to 1 .O x 1 o6 cells/mI. 2.4. Modulation of H-2 The induction o f modulation was carried out as described previously [ I ] . Briefly, H-2 modulation was induced by adding 0.05 ml H-2 alloantiserum to 0.5 ml of peritoneal cell suspension, and incubation of the mixture for 2 h at 37 "C. 2.5. Inhibition of modulation Peritoneal cell suspensions were exposed to various inhibitors for 1 h at 37 "C, prior to the addition of H-2 antisera, as described previously [ 21. The following compounds were used: colchicine (Sigma, Chemical Co., St. Louis,Mo.), vinblastine sulfate (Velban) (Lilly & Co., Indianapolis, Indiana), and sodium azide (Merck, Darmstadt). In some experiments modulation was inhibited by the exposure of peritoneal cell suspensions to 4 "c.
2.6. The sensitization of SRBC SRBC were coated with antibody and complement (EAC) according to the procedure described by Lay and Nussenzweig [ 71 and by Stjernsward et al. [ 101. SRBC were washed 3 times in phosphate buffered saline (PBS) (pH 7.4), containing Mg* and Ca++ ions at a final concentration of 1 0 - 3 ~ and 0.5 x 1 0 - 3 concentration, ~ respectively. A 5 % suspension of SRBC in PBS was prepared and a volume of 2 ml of the suspension was incubated for 3 0 min at 37 "C with rabbit anti-SRBC serum. Early rabbit anti-SRBC serum was used at a concentration of 1 : 10, whereas hyperimmune serum was used at a concentration o f 1 :50. Following their coating with antibody, the SRBC were washed twice in PBS and resuspended in 2 ml of normal mouse serum (NMS) diluted 1 : 10 in PBS, which had been absorbed with an equal volume of packed, washed SRBC. After incubation for 30 min at 37 "C the cells were washed again in PBS, and the concentration of SRBC was adjusted to 1 %. The precedure used for the preparation of SRBC coated with antibody alone (EA) was similar to the procedure used for the preparation of EAC with the following exceptions: the medium used was PBS devoid of Mg++ and Ca* ions, and no complement was added.
Eur. J. Immunol. 1975. 5: 27-32 suspensions were thoroughly mixed on a cyclomixer (ClayAdams), spun at 200 x g for 2 min, incubated for 15 min at 37 " C , and then for an additional 15 min at room temperature. Just prior t o counting of the rosettes the suspension was gently drawn up and down a pasteur pipette. Rosettes were counted in a hemocytometer. Rosettes were defined as lymphoid cells to which at least 5 red blood cells adhered.
3. Results 3.1. The effect of anti-H-2 sera on the formation of EAC and EA rosettes
The exposure of murine peritoneal cells to various anti-H-2 sera for 2 h at 37 "C diminished the proportion of cells forming rosettes with SRBC coated with early anti-SRBC antibody and C ( 19s EAC rosettes). Exposure of strain A peritoneal cells to DBA/2 anti-C3H serum (Table 2) or C57BL anti-A serum (Table 3 ) reduced the proportion o f EAC rosettes formed by an average o f about 80 %. The formation of EAC rosettes by C57BL peritoneal cells was inhibited t o a similar striking extent by exposure o f the peritoneal cells to A antiC57BL serum (Table 3). C3H anti-DBA/2 serum inhibited the formation of EAC rosettes by an average of only about 40 % (Table 2). In contrast to the marked inhibitory effect of anti-H-2 sera on the formation of EAC rosettes, these sera did not decrease the percentage of EA rosettes. None of the anti-H-2 sera tested affected the capacity of peritoneal cells to bind SRBC coated with hyperimmune, 7 S antibody, either in the presence or in the absence of C (Tables 2 and 3). Table 2. The effect of C3H anti-DBA/2and DBA/2 antLC3H sera on the formation of rosettes by strain A peritoneal cells Serum to which the cells were exposed NMS NMS NMS NMS C3H anti-DBA/Z C3H antkDBA/2 C3H anti-DBA/Z C3H anti-DBA12 DBA/2 antX3H DBA/2 antX3H DBA/2 antX3H DBA/2 antiC3H
Type of rosette tested 19 Sb) EAC EA 19s 7 Sc) EAC IS EA 19s EAC 19s EA I S EAC IS EA 19s EAC 19s EA IS EAC IS EA
9i of a l l s forming rosette& 20.2 5.8 22.4 24.6 4.6 4.6 21.0 19.8 4.8 5.8 24.2 23.6
13.2 31.6 28.8 1.6 4.6 1.2 32.6 16.4 8.0 29.8 14.2 2.0 4.6 4.4 39.0 32.2 3.2 6.0 6.0 4.2 5.0 5.2 36.0 20.4
Each vertical column represents results obtained in a single experiment. 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. 7 S antibody: rabbit serum bled 7 days after 3 injections of SRBC.
2.7. Rosette formation
The exposure of peritoneal cells to anti-Hi-2 alloantisera, which had been inactivated for 30 min at 56 OC, eliminated the binding of EAC as effectively as exposure to fresh sera. Therefore, H-2 antisera did not inhibit the formation of EAC rosettes by forming antibody-antigen-C complexes which in turn blocked the cell surface C receptors.
To lo6 peritoneal cells suspended in 0.25 ml PBS wasadded an equal volume either of EAC or EA suspension. The cell
Anti-H-2 sera seem to exert their effect on the EAC receptors of adherent cells only. Thus, when the adherent cells were
Eur. J. Immunol. 1975. 5: 27-32
Association of H-2 antigens and EAC receptors
Table 3. The effect of anti-H-2 sera on the formation of rosettes by peritoneal cells of mice of the A and C57BL strains Serum to which the cells expod
Source of periton4 % Of cells 19SEAC
CS7BL anti-A/J
A/1 All
NMS
AIJ
NMS
A/J NMS C57BL A/J antiC57BL C57BL C57BL anti-A/J
15.7 f 2.1a) 2.9 f 0 . 3 13.1b) 3.6 23.p) 3.9
19SEA
3.0 f 0 . 7 2.8 f 0 . 6 2.4 1.4 3.9 5.1
rwtter 7SEAC7SEA
27d) 28.1
25.7 22.0
a) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. Mean of 12 experiments SE. b) 19 S antibody: first fraction of antiserum obtained by filtration through a Sephadex G200 column. Mean of two experiments. c) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. Mean of two experiments d ) 7 S antibody: rabbit serum bled 7 days after 3 injections of SRBC.
*
Table 4. The effect of anti-H-2 serum on the formation of EAC rosettes by either unfractionated suspensions of strain A peritoneal cells or by nonadherent peritoneal cells
Serum kited
UnfrlCtkMtCd psritonul cells suspendon
NMS
9.0“)
C57BLantbA
2.3
121 21
29
rosettes by BALB/c peritoneal cells. When C57BL anti-A serum was absorbed with C3H spleen cells, its activity on C3H peritoneal cells was completely eliminated, yet maintained its capacity of reducing the formation of EAC rosettes by BALB/c cells. Conversely, absorption of the serum with BALB/c spleen cells eliminated its activity on BALB/c peritoneal cells, without affecting its activity on C3H peritoneal cells. A/J anti-BALB/c serum was found t o have a strong inhibitory effect on the formation of 19 S EAC rosettes by BALB/c peritoneal cells (Table.5). This serum contains antibodies against K end H-2 specificities, and against “Ia antigens” [ 121. Antibodies to Ia antigens have recently been shown to block EAC receptors specifically (Schlesinger and Chaouat, unpublished data). A/J anti-BALB/c serum absorbed with C57BL spleen to remove antibodies to Ia antigens was found to retain its high efficiency in modulating the EAC receptors on BALB/c peritoneal cells. These results indicate again that EAC receptors on the surface of peritoneal cells are associated with K end H-2 antigenic specificities. Exposure of peritoneal cells t o either @or Ly antisera had no effect on the formation of EAC rosettes (Table 5 ) . Table 5. The effect of various antisera on the formation of 19 S EAC rosettes by peritoneal cells
Nodhecent puitoneal
ceh 5.5 5.8
9.0 7.4
NMS
NMS
BALBlc BALBIc BALB/c C3H C3H A1J
C57BL antl-A
UJ
NMS
C3H C3 H C3H
A antl-BALB/c
a) Percentage of cells forming rosettes Each vertical column represents results obtained in a single experiment.
C57BL anti-A
NMS C57BL antM
removed from the peritoneal cell suspension the remaining, nonadherent cells were not affected by exposure to anti-H-2 sera (Table 4).
Anti+ serum
3.2. Specificity of the sites linked to EAC receptors Exposure of peritoneal cells of the A/J strain to the DBA/2 anti-C3H serum inhibited the formation of EAC rosettes t o a much greater extent than exposure to the C3H anti-DBA/2 serum. This difference could not result from any difference in the antibody activity of these sera, as reflected in their cytotoxic titers with cells of the A strain (Table 1). The different activity of the sera could result from the fact that DBA/2 anti-C3H serum reacts with K end H-2 specificities on strain A cells, whereas the C3H anti-DBA/2 serum reacts with D end H-2 specificities on strain A cells [ 111. If this interpretation is correct, it would seem that the EAC binding site on peritoneal cells is associated more intimately with K end H-2 specificities than with D end H-2 specificities. Further support for this assumption was gained from experiments with the C57BL anti-A serum, a serum containing antibodies against multiple specificities o f both the D end and K end H-2 regions. While C57BL anti-A serum had a similar cytotoxic effect on C3H and on BALB/c target cells(Tab1e l ) , it had a different effect on the formation of EAC rosettes by cells of both strains. C57BL anti-A serum inhibited the formation of 19 S EAC rosettes as effectively in C3H peritoneal cells as in strain A peritoneal cells (Table 3), but had only a weak inhibitory effect on the formation of EAC
Anti-Ly m m
18.6 2.1 11.0 18.5 3.8 14.2 4.4 18.2 15.3 16.6
9.0 2.1 8.0 13.8 2.3
a) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. Each vertical column represents results obtained in a
single experiment. 3.3. Nature of the association o f H-2 antigens and EAC receptors The concomitant disappearance of H-2 antigens and EAC receptors from the surface of peritoneal cells suggests that H-2 specificities, particularly of the K region, are closely associated with EAC receptors on the cell membrane. The nature o f this association was explored by studying the effect of various inhibitors on the suppression o f EAC rosette formation by anti-H-2 sera. Exposure of peritoneal cells to anti-H-2 sera, under conditions which would not allow modulation of H-2 antigens [ 1 , 21 also prevented the disappearance of EAC receptors from the cell surface. Thus, keeping peritoneal cells on ice during their exposure to anti-H-2 sera, or exposing them to anti-H-2 sera M NaN3, completely preat 37 OC in the presence of vented the inhibitory effect of anti-H-2 sera on the formation
Eur. J. Immunol. 1975. 5: 27-32
M. Schlesinger and M. Chaouat
30
Table 6. The effect of exposure of strain A peritoneal cells to antiH-2 sera at low temperature or in the presence of NaN3 on their capacity to form rosettes Exp. Serum to which Type of rosette no. the cells wen: exposed tested NMS
1
NMS NMS NMS
C57BL anti-A C57BI anti-A C57BL anti-A C57BL anti-A
70 of cells forming
3 7 W
19 s E A C ~ ) 20.0 5.2 19 S EA 19.8 7 S EAC 7SEA 17.0 19 S EAC 4.2 3.6 19 S EA 25.2 7 S EAC 7 S EA 29.8
NMS 19 S EAC NMS 19 S E A C3H anti-DBA/2 19 S EAC C3H anti-DBA/?19 S EA DBA/2 anttC3H 19 S EAC DBAI2 antLC3H 19 S EA C57BL anti-A 19 S EAC C57BL anti-A 19 SEA
I1
rosettes 4’0) N~N~C) 17.4
28.8
26.2 4.8 23.8 6.4 21.2 2.4 19.0 4.8
7.2
14.2 4.4 6.0 5.2
3.8 5.4
28.4 6.6 27.0 5.4 32.8 5.8 27.3 6.8
of I9 S EAC rosettes (Table 6). These experiments indicate that although H-2 antigens and EAC receptors may be in close topographical proximity on the membrane of peritoneal cells, they are not carried by identical cell surface structures. They seem to be located far enough from each other, so that H-2 antibody does not interfere by steric hinderance with the formation of EAC rosettes. Microtubule depolymerizing agents, such as vinblastine, are relatively weak inhibitors of H-2 modulation [2, 31. I t was previously found [2], and confirmed in the present study, that at a concentration of 1 0 - 5 M , vinblastine does not affect the modulation of H-2 antigens on peritoneal cells by specific antiserum. However, this concentration of vinblastine completely prevented the disappearance of EAC receptors from Table 7. The effect of cxposure of strain A peritoneal cells to antiH-2 serum in the presence of vinblastine on their capacity to form rosettes
v;,of cells forming rosettcsa’ No inhibitor Vinblastinr added addcd
Serum to u hich the cells \vcrc rxpocd
Typc of
NMS
19 S EACb’ 23.0‘)
NMS
19 S E A
5.8
5.0
3.8
5.8
C57B1 anti-A
19 S EAC
3.4
C 5 7 B L anti-A
19 S I.:A
3.6 5.4
26.4 2.6
18.4 3.8
21.2
3.0
21.6‘)
15.6
a) Prior to the rosette test the peritoneal cells were exposed to various sera for 2 h at 3 7 O C in the presence or absence of 10-4M
vinblastine. b) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. c) Each vertical column reprcsents results obtained in a single experi-
ment.
4 Discussion The present study demonstrates that exposure of peritoneal cells to anti-H-2 sera, a procedure leading to the disappearance of H-2 antigenicity from the surface of the cells [ 1-31, results in the concomitant disappearance of EAC receptors from the membrane of peritoneal cells. EA receptors were unaffected by modulation of the H-2 antigenicity. Antisera reactive with K end H-2 specificites had a stronger effect on EAC receptors than antisera reactive with D end H-2 specificities. The different activity o f various H-2 antisera may result from three factors: ( I . ) different concentration of anti-H-2 antibodies; (2.) the presence of either K end or D end H-2 antibodies; and (3.) contamination by antibodies to non-H-2 antigens and particularly by anti-la antibodies [ 121.
3.2
26.0 24.0 15.8 2.3 21.6 24.4
a) Prior to the rosette test the peritoneal cells were exposed to various sera for 2 h at 37 OC. b) Prior to the rosette test the pcritoneal cells were exposed to various sera for 2 h at 4 O C . c) Prior to the rosette test the peritoneal cells were exposed to various sera for 2 h at 37 O C in the presence of 10-4M NaN,. d) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. 7 S antibody: rabbit serum bled 7 days after 3 injections of SRBC.
roscttr tcstcd
the membrane of peritoneal cells upon exposure to anti-H-2 serum (Table 7). In contrast to vinblastine, 10- ’M colchicine failed to prevent the inhibitory effect of anti-H-2 serum on the formation of EAC rosettes.
Antisera against Ia antigens have recently been shown to react specifically with EAC receptors (Schlesinger and Chaouat, unpublished data). In contrast to the effect of H-2 antisera, the effect of anti-la antibodies could not be inhibited by exposure of the cells t o various metabolic inhibitors (Schlesinger and Chaouat, unpublished data). While the antisera used in the present study may contain antibodies to Ia antigens, these antibodies were not required for the effect of H-2 antisera on EAC receptors. A/J antiBALB/c serum, which in addition t o antibodies to K end H-2 antigens also contains la antibodies, retained its strong activity on EAC receptors even after absorption with C57BL spleen presumably removed all la antibodies. There was no correlation between the cytotoxic titer obtained with various sera and their effect on the formation of EAC rosettes. Thus, C57BL anti-A serum had a similar cytotoxic effect on C3H and BALB/c target cells, but inhibited the formation of EAC rosettes by C3H cells to a greater extent than by BALB/c cells. C3H anti-DBA/2 serum had a weaker cytotoxic effect on strain A cells than DBA/2 antiC3H serum, yet the former serum was more effective in suppressing the formation of EAC rosettes by strain A cells. The results of the present study support the conclusion that EAC receptors on the membrane of peritoneal cells are closely linked to H-2 antigenic specificities, particularly those of the K end. Previous studies have shown that K end and D end H-2 specificities constitute distinct entities on the cells’ membranes. Exposure of lymphocytes t o antisera against either K end or D end H-2 determinants leads t o redistribution and “capping” o f only the corresponding antigenic specificities on the cells’ membranes, while the distribution of the other H-2 specificities remains unaltered [ 131. Similarly, exposure of peritoneal cells to antisera against some H-2 specificities leads to disappearance of the corresponding H-2 specificities, without affecting the expression of other H-2 specificities (Schlesinger and Chaouat, unpublished data, [ 21). In studies of the modulation of the TL antigen of thymus and leukemic cells [4] it was shown that several TL specificities modulate concomitantly upon exposure of the cells to antibodies against only one of the TL specificities [ 141. On the other hand, TL antigens constitute distinct entities, since antibodies to, say TL-I, will not inhibit the capacity of cells possessing the TL-I and TL-3 antigenic specificities
Eur. J. Immunol. 1975. 5: 27-32
t o absorb antibodies to TL-3 determinants [ 141. There are indications in the present study that although EAC receptors and H-2 antigenic determinants are closely linked on the cell surface, they constitute distinct entities. Exposure of peritoneal cells to anti-H-2 sera at 4 OC or in the presence of NaN3, procedures which prevent the modulation of H-2 antigens [ 1, 21 but d o not prevent the coating of the cells with H-2 antibodies, did not lead to any reduction o f the capacity of the cells to form EAC rosettes. Moreover, exposure of peritoneal cells to vinblastine at a concentration of 1 0 - 5 M , which does not inhibit H-2 modulation [ 1, 21, was found t o prevent the disappearance of EAC receptors from the surface of the cells. Although colchicine at a similar concentration was without effect it seems possible that the close association of EAC receptors and H-2 antigens is maintained by the inte. 'ity of the microtubule system. Studies by Berlin and coworkers [ 151 have shown that the cell surface o f macrophages operates as a functional mosaic, in which membrane transport sites are separated from membrane parts which are internalized during phagocytosis. This functional heterogeneity of the cell membrane is broken down by microtubule depolimerizing agents [ 151. Similarly, microtubules seem t o affect the cell surface arrangement of membrane constituents which determine the agglutinability of the cells by concanavalin A [16-181. Mantovani et al. [ 191 presented evidence indicating that the EAC and EA receptors on the surface of macrophages may play different, and even opposing, roles during the immune reaction. In early stages of the immune response, complexes of antigen, IgM antibody, and complement bind t o EAC receptors on the surface of macrophages. These complexes remain bound to the cell surface of macrophages so that the antigen is presented as a continuous stimulus to other cells involved in the immune response. In this way, the EAC receptors on the cell membrane of macrophages may amplify the immune reaction in the early stages o f the immune reaction. In contrast, during the secondary response, EA receptors bind complexes of antigen and IgC antibody which are then rapidly interiorized and degraded. Thus, EA receptors may serve to eliminate any excess antigenic stimulus during the secondary immune response. The extreme efficiency of peritoneal macrophages in antibody-induced suppression of the immune response has been amply documented [20, 2 I]. Recent studies have suggested additional amplifying functions for EAC receptors. Feldmann has shown that monomeric IgM sequestered from the cell surface of T lymphocytes becomes attached t o the surface of macrophages, and that macrophages armed with T cell-derived immunoglobulin then cooperate with B cells in the production of antibody to thymuodependent antigens [22, 231. It now seems that the macrophage receptor for T-derived immunoglobulin is identical with the EAC receptor, since antisera against mouse C3 abolish the in vitro immune response to thymusdependent antigens (Feldmann, M., personal communication). Dukor and Altman provided evidence that while B lymphocytes become triggered by antigen binding, an additional signal for the response of B cell may be the attachment o f activated C3 t o the EAC receptors [24]. The present demonstration of an association between histocompatibility antigens and EAC receptors may have implications on the genetic regulation of the immune response. Many Ir genes in the mouse are linked t o the K end o f the H-2 locus [25]. Mice genetically deficient in their response to certain antigens produce normal levels of IgM antibody during the primary immune response, but show an impaired
Association of H-2 antigens and EAC receptors
31
production o f IgC antibodies in the secondary response [25 1. It is assumed that Ir genes affect the formation of specific antibodies through their effect on the immune function of T lymphocytes [ 261. Shevach and Rosenthal[27] recently showed that macrophage-bound antigens can activate T lymphocytes only when the macrophages are derived from animals possessing Ir genes determining an immune response t o these antigens. They suggested that antigen recognition sites on T lymphocytes are physically related t o sites on these cells which determine their binding t o macrophages, and that both are linked to histocompatibility antigens. The results of the present study suggest an additional mechanism for the regulation o f immune response by Ir genes It is postulated that Ir genes may affect the functional capacity of EAC receptors on macrophages. This hypothesis is based on the assumptions that EAC receptors are important for the cooperation o f T lymphocytes with B lymphocytes and for the switch-over from IgM to IgG antibody produotion. It is possible, then, that K end H-2 antigens or antigenic determinants specified by H-2 linked Ir genes, which are located in close proximity to EAC receptors on the cell surface, may affect the attachment of antigen-antibody-C complexes to macrophages Only when the EAC receptors are adjacent t o appropriate cell surface structures determined by Ir or H-2 genes, will the complexes be bound in a way suitable for triggering the reaction of immunocompetent B lymphocytes and possibly also of T lymphocytes. Received February 9, 1974; in revised form August 31, 1974. 5. References 1 Schlesinger, M. and Chaouat, M., Tissue Antigens 1972.2: 427. 2 Schlesinger, M. and Chaouat, M., Transplant. Proc. 1973. 5: 105. 3 Schlesinger, M., Biomedicine 1973. 18: 437. 4 Boyse, E.A., Stockert, E. and Old, L.J., Proc. Nut. Acad. Sci US 1967. 58: 954. 5 Nussenzweig, V., Bianco, C., Dukor, P. and Eden, A., in Amos,
D.B. (Ed.) Progress in Immunology, Academic Press, New York 1971, p. 73. 6 Tizard, I.R., Bacteriol. Rev. 1972. 35: 365. 7 Nussenzweig, V. and Lay, W.H.,J. Exp. Med. 1968. 128: 991. 8 Schlesinger, M. and Hurvitz, D., Transplantation 1969. 7 132. 9 Schlesinger, M., Cohen, A. and Hurvitz, D., Isr. J. Med. Sci 1969. 5: 235. 10 Stjernsward, J., Jondal, M., Vanky, F., Wigzell, H. and Sealy, R., Lancer 1972. i: 352. 11 Klein, J. and Shreffler, D.C., Transplant. Rev. 1971. 6: 3. 12 Sachs, D.H. and Cone, J.L.,J. Exp. Med. 1973. 138: 1289. 13 Neauport-Sautes, C., Lilly, F., Silvestre, D. and Kourilsky, F.M., J. Exp. Med 1973.137: 511. 14 Old, L.J., Stockert, E., Boyse, E.A. and Kim, J.H., J. Exp. M d . 1968. 127: 523. 15 Ukena, T.E. and Berlin, R.D., J. Exp. Med. 1972. 136: 1. 16 Inbar, M., Ben-Bassat, H. and Sachs, L., Proc. Nut. Acad Sci US 1971. 68: 2748. 17 Berlin, R.T. and Ukena, T.E., Nature-New Biol. 1972.238: 120. 18 de Petris, S., Raff, M.C. and Malucci, L., Nature-New Biol. 1973. 244: 275. 19 Mantovani, B., Rabinovitch, M. and Nussenzweig, V., J. Exp. Med. 1972. 135: 780. 20 Schwartz, R.S., Ryder, R.J.W. and Cottlicb, A.A., Progress in Allergy 1970. 14: 81.
A. Polliack, U. Hammerling, N. Lampen and E. de Harven
Eur. J. Immunol. 1975. 5: 32-39
21 Haughton, G., Cell Immund. 1971. 2 : 567. 22 Feldmann, M., J. Exp. Med. 1972. 134: 103. 23 Feldmann. M. and Nossal, G.J.V., Transplant. Rev. 1972. 13: 3. 24 Dukor, P. and Hartmann, K.U., CelL Immund. 1973. 7: 349.
25 Grumet, F.C., Mitchell, G.F. and McDevitt, H.O.,Ann. New York Acod. Sci 1971. 190: 170.
32
A. PoIIiack,+ u. Hammerling, N. Lampen and E. de Harven
26 Benacerraf, B. and McDevitt, H.O., Science 1972. 175: 233. 27 Shevach, E.M. and Roscnthal, A.S., J. Exp. Med. 1973. 138: 121 3.
Surface morphology of murine B and T lymphocytes: A comparative study by scanning electron microscopy"
Memorial Sloan-KetteringCancer Center, New and Department Of Hematology'A variety of murine lymphocytes of known B or T derivation obtained from and Hebrew University different lymphoid organs were prepared for scanning electron microscopy University Medical School, Jerusalem+ (SEM) by the critical point drying method after collecting the cells by aspiration onto silver membranes. Comparison of SEM appearances of cells prepared by this technique and serological classification according to surface antigens showed that most T cells had smooth surfaces with few microvilli, while many B lymphocytes were moderately to markedly villous. Further evidence for the above correlation was obtained by examining thymic cells and enriched B or T cell populations. Thymic cell suspensions containing less than 5 % B cells showed over 8 0 5% generally smooth cells by SEM. Enriched T cell populations, obtained by mass cytolysis of lymph node preparations with anti-la or anti-Ig sera or by purification through nylon fiber columns, contained over 85 % T cells, and more than 75 % of them were of the smooth cell type. A similar correlation was noted for enriched B cell populations obtained by
cytolysis of lymph node cells with anti-Thy-I serum, and by lysis of EAC-rosettes. Over 90 % of these cells were identified as B cells by immunologic methods and approximately 7 5 % had moderate to markedly villous surfaces. The 1 5 5% difference can be accounted for by the existence of a subpopulation of smooth B cells. Direct observation of EAC-rosettes confirmed that most B cells had moderate t o large numbers of surface microvilli and that less than 10 % were smooth. It is possible that some of the smooth cells seen in enriched B cell populations may represent precursors or B lymphocytes at different stages of differentiation. These results indicate that murine T and B lymphocytes, like their human counterparts, can be recognized in many cases under the SEM on the basis of their surface morphology. Smoother B and more villous T cells are difficult to classify by SEM without parallel immunologic identification.
1. Introduction Thymus-derived (T) and bone marrow-derived (B) lymphocytes can be distinguished by their different surface antigens and receptors [ I , 21. Murine T cells carry on their surface the alloantigens Thy-1 (formerly named 0)[3, 41, the Ly series of alloantigens [ 5 , 61, and TL which is expressed only on thymocytes [ 7 ] . B cells have the alloantigens Ia and Pc [8-111, the heteroantigens, mouse-specific B lymphocyte antigen [I 8091 * ) Supported by Grant CA-08748 from the National Cancer Institute, American Cancer Society Grant DT-31, The Frank Strick Foundation, National Science Foundation Grant GB-34299, and USPHS International Postdoctoral Research Fellowship awarded to Dr. A. Polliack.
Correspondence: E . de Harvcn, Memorial Sloan-KetteringCancer Center, New York, N.Y. 10021, USA Abbreviations: slg: Surface immunoglobulin MBLA: Mouse-specific B lymphocyte antigen MSPCA: Mouse-specific plasma cell antigen TEM: Transmission electron microscopy SEM: Scanning electron microscopy SRBC: Sheep erythrocytes EAC: Sheep erythrocytes coated with rabbit anti-SRBC antibody and complement TDL: Rat thoracic duct lymphocytes LN: Lymph node Anti-lgM/FI: Fluoresceinated rabbit antibody to mouse IgM
(MBLA [ 121) and mouse-specific plasma cell antigen (MSPCA [ 131 1. Furthermore, B cells have readily detectable surface immunoglobulin (slg) [ 141 and also bear receptors for the Fc fragment of IgG [ 151 and for the C 3 component of complement [ 161. Despite these differences in antigen patterns the two types of lymphocytes have not been consistently distinguished by routine light or electron microscopy unless cells were first labeled by their respective surface markers [ 171. However, recently differences in human B and T lymphocytes have been recognized by scanning electron microscopy (SEM) and a spectrum of lymphocyte surface architecture has been described [ 18-20]. Many T cells display smooth surfaces with only small numbers of surface microvilli, whereas the majority of B cells have larger numbers of microvilli. However, some B and T lymphocytes have similar surface features and cannot be assigned to either class accurately [ 18-20]. The present investigation was undertaken to define the surface architecture of murine lymphocyte populations of known derivation, and to establish whether previous findings on human lymphocytes [ 18-20] could be extended to murine cells.
M. Schlesinger and M. Chaouat Department of Experimental Medicine and Cancer Research, The Hebrew University - Hadassah Medical School, Jerusalem
Association of H-2 antigens and EAC receptors
27
The association of H-2 antigens and EAC receptors on the surface of peritoneal cells+ The exposure of murine peritoneal cells to anti-H-2 sera results in a diminished expression of H-2 antigen o n the cell surface. Concomitant with this “H-2 modulation” the capacity of macrophages t o bind sheep red blood cells coated with antibody and complement (EAC) was markedly diminished. In contrast, there was no change in the capacity of modulated macrophages to bind sheep red blood cells coated with antibody alone (EA). Antibodies to K end H-2 specificities were more effective in reducing the binding of EAC than antibodies to D end H-2 specificities. Exposure of peritoneal cells to 0 or Ly antisera had n o effect on the formation of EAC rosettes. Exposure o f peritoneal cells to anti-H-2 sera, under conditions which would not allow modulation of H-2 antigens, also prevented the reduction of EAC binding. Thus, the EAC receptors and H-2 antigenic specificities seem to be closely related on the surface of peritoneal cells, but constitute distinct cell surface structures. Preliminary evidence indicates that vinblastine, a microtubule depolymerizing agent, may disrupt the close association of EAC receptors and H-2 antigens. It is suggested that the association of EAC receptors and K end H-2 determinants on the membrane of macrophages may have implications for the regulation of the immune response by H-2-linked Ir genes. ~
1. Introduction
2. Materials and methods
The exposure of murine peritoneal cells to anti-H-2 sera, in the absence of complement (C), leads t o the disappearance of H-2 antigenicity from the cell surface of peritoneal macrophages [ 1-31. Following such “antigenic modulation” [ 3 , 41 murine macrophages are n o longer susceptible to the cytotoxic effect of H-2 antibodies and complement. Evidence has been presented that H-2 modulation probably results from the active pinocytosis of complexes of H-2 antibody along with cell surface H-2 antigenic determinants [ 1-31.
2.1. Mice
The cell surface of macrophages displays several distinctive receptors, which probably play an important role in the immunological functions of macrophages [ 51. Macrophages possess receptors for the Fc component of immunoglobulins [ 61. These receptors can be assayed by the formation of rosettes with sheep red blood cells (SRBC) coated with 7 S antibodies to SRBC (EA-rosettes) [7]. The membrane of macrophages also possesses receptors for activated C3, which can be assayed by testing the formation of rosettes with sensitized SRBC and C (EAC-rosettes) [S]. In the present study, modulation o f the H-2 antigenicity o f peritoneal macrophages was found to be accompanied by a concomitant disappearance of the EAC receptors from the surface of peritoneal macrophages. The nature of the association of H-2 antigenicity and EAC receptors was analyzed.
+ This study was supported by USPHS Research Agreement No. 06-048-1 and by a Lady Davis Endowment. Correspondence: Michael Schlesinger, Department of Experimental Medicine and Cancer Research, The Hebrew University - Hadassah Medical School, Jerusalem, Israel Abbreviations: C Complement EA: Sheep red blood cells coated with antibody EAC: Sheep red blood cells coated with antibody and complement NMS: Normal mouse serum SRBC Sheep red blood cells PBS: Phosphate buffered saline PC: Peritoneal cell(s) LNC: Lymph node cell(s)
Mice of the A/J, BALB/c, C3H and C57BL/6 inbred strains were used.
2.2. Antisera Alloantisera were elicited by repeated i.p. injections of spleen cell suspensions as previously described [8]. The following alloantisera were used: A/J anti-C57BL, A/J anti-BALB/c, DBA/2 anti-C3H, C3H anti-DBA/2, and C57BL anti-A sera. The cytotoxic titers obtained with various anti-H-2 sera in the present study are shown in Table 1 . AKR/J anti-C3H spleen serum was used as anti-O-C3H serum, whereas (AKR/l x C57BL)Fl anti-C3H thymus serum was used as anti-Ly A . 1 serum. Table 1. The cytotoxic effect of various antisera on allogeneic peritoneal and lymph node cells
A
Test cells BALB/c C3H C57BL/6 LNC PC LNC PC LNC PC LNC
Antiserum
PC
C57BLanti-A C3Hanti-DBAI2 DBA/Zanti-C3H A antiieALBlc A anti-CS7BL
64a) 128
16 64
32 128
32 32 128 128 8
32
32
64
64
32 64
64
a) Cytotoxic titers are expressed as the reciprocal of the highest serum dilution which still killed 50 % or more of the test cells Rabbit antisera against SRBC were elicited by i.p. injections of 1 ml of 25 % suspensions (v/v) of washed SRBC. The rabbits were bled from the marginal ear vein 5 days after the first inoculation in order t o obtain antisera containing mainly 19 S antibodies. Sera rich in 7 S antibodies were obtained from rabbits bled a week after the third or fourth injection o f SRBC. Purified 19 S antibody used in some experi-
28
M. Schlesinger and M. Chaouat
ments was obtained by collecting the excluded fraction of antisera filtered through a Sephadex G-200 column (Pharmacia, Uppsala) [ 91. 2.3. Cell suspensions Peritoneal cells were obtained from normal, unstimulated mice by repeated washings of their peritoneal cavities with saline [ 1 1 . The cell concentration was adjusted to 2 x 106/ml in saline. In some experiments, adherent cells were removed from the peritoneal cell suspension. Peritoneal cells were suspended in Hanks' basic salt solution, supplemented with 10 % fetal calf serum, and 1 .O ml of this suspension was delivered into Falcon plastic petri dishes, 45 mm in diameter. The petri dishes were incubated for 3 h at 37 "C in a humidified incubator containing 5 % CO,, and the nonadherent cells were collected carefully with a pipette. The concentration of the nonadherent cell population was adjusted to 1 .O x 1 o6 cells/mI. 2.4. Modulation of H-2 The induction o f modulation was carried out as described previously [ I ] . Briefly, H-2 modulation was induced by adding 0.05 ml H-2 alloantiserum to 0.5 ml of peritoneal cell suspension, and incubation of the mixture for 2 h at 37 "C. 2.5. Inhibition of modulation Peritoneal cell suspensions were exposed to various inhibitors for 1 h at 37 "C, prior to the addition of H-2 antisera, as described previously [ 21. The following compounds were used: colchicine (Sigma, Chemical Co., St. Louis,Mo.), vinblastine sulfate (Velban) (Lilly & Co., Indianapolis, Indiana), and sodium azide (Merck, Darmstadt). In some experiments modulation was inhibited by the exposure of peritoneal cell suspensions to 4 "c.
2.6. The sensitization of SRBC SRBC were coated with antibody and complement (EAC) according to the procedure described by Lay and Nussenzweig [ 71 and by Stjernsward et al. [ 101. SRBC were washed 3 times in phosphate buffered saline (PBS) (pH 7.4), containing Mg* and Ca++ ions at a final concentration of 1 0 - 3 ~ and 0.5 x 1 0 - 3 concentration, ~ respectively. A 5 % suspension of SRBC in PBS was prepared and a volume of 2 ml of the suspension was incubated for 3 0 min at 37 "C with rabbit anti-SRBC serum. Early rabbit anti-SRBC serum was used at a concentration of 1 : 10, whereas hyperimmune serum was used at a concentration o f 1 :50. Following their coating with antibody, the SRBC were washed twice in PBS and resuspended in 2 ml of normal mouse serum (NMS) diluted 1 : 10 in PBS, which had been absorbed with an equal volume of packed, washed SRBC. After incubation for 30 min at 37 "C the cells were washed again in PBS, and the concentration of SRBC was adjusted to 1 %. The precedure used for the preparation of SRBC coated with antibody alone (EA) was similar to the procedure used for the preparation of EAC with the following exceptions: the medium used was PBS devoid of Mg++ and Ca* ions, and no complement was added.
Eur. J. Immunol. 1975. 5: 27-32 suspensions were thoroughly mixed on a cyclomixer (ClayAdams), spun at 200 x g for 2 min, incubated for 15 min at 37 " C , and then for an additional 15 min at room temperature. Just prior t o counting of the rosettes the suspension was gently drawn up and down a pasteur pipette. Rosettes were counted in a hemocytometer. Rosettes were defined as lymphoid cells to which at least 5 red blood cells adhered.
3. Results 3.1. The effect of anti-H-2 sera on the formation of EAC and EA rosettes
The exposure of murine peritoneal cells to various anti-H-2 sera for 2 h at 37 "C diminished the proportion of cells forming rosettes with SRBC coated with early anti-SRBC antibody and C ( 19s EAC rosettes). Exposure of strain A peritoneal cells to DBA/2 anti-C3H serum (Table 2) or C57BL anti-A serum (Table 3 ) reduced the proportion o f EAC rosettes formed by an average o f about 80 %. The formation of EAC rosettes by C57BL peritoneal cells was inhibited t o a similar striking extent by exposure o f the peritoneal cells to A antiC57BL serum (Table 3). C3H anti-DBA/2 serum inhibited the formation of EAC rosettes by an average of only about 40 % (Table 2). In contrast to the marked inhibitory effect of anti-H-2 sera on the formation of EAC rosettes, these sera did not decrease the percentage of EA rosettes. None of the anti-H-2 sera tested affected the capacity of peritoneal cells to bind SRBC coated with hyperimmune, 7 S antibody, either in the presence or in the absence of C (Tables 2 and 3). Table 2. The effect of C3H anti-DBA/2and DBA/2 antLC3H sera on the formation of rosettes by strain A peritoneal cells Serum to which the cells were exposed NMS NMS NMS NMS C3H anti-DBA/Z C3H antkDBA/2 C3H anti-DBA/Z C3H anti-DBA12 DBA/2 antX3H DBA/2 antX3H DBA/2 antX3H DBA/2 antiC3H
Type of rosette tested 19 Sb) EAC EA 19s 7 Sc) EAC IS EA 19s EAC 19s EA I S EAC IS EA 19s EAC 19s EA IS EAC IS EA
9i of a l l s forming rosette& 20.2 5.8 22.4 24.6 4.6 4.6 21.0 19.8 4.8 5.8 24.2 23.6
13.2 31.6 28.8 1.6 4.6 1.2 32.6 16.4 8.0 29.8 14.2 2.0 4.6 4.4 39.0 32.2 3.2 6.0 6.0 4.2 5.0 5.2 36.0 20.4
Each vertical column represents results obtained in a single experiment. 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. 7 S antibody: rabbit serum bled 7 days after 3 injections of SRBC.
2.7. Rosette formation
The exposure of peritoneal cells to anti-Hi-2 alloantisera, which had been inactivated for 30 min at 56 OC, eliminated the binding of EAC as effectively as exposure to fresh sera. Therefore, H-2 antisera did not inhibit the formation of EAC rosettes by forming antibody-antigen-C complexes which in turn blocked the cell surface C receptors.
To lo6 peritoneal cells suspended in 0.25 ml PBS wasadded an equal volume either of EAC or EA suspension. The cell
Anti-H-2 sera seem to exert their effect on the EAC receptors of adherent cells only. Thus, when the adherent cells were
Eur. J. Immunol. 1975. 5: 27-32
Association of H-2 antigens and EAC receptors
Table 3. The effect of anti-H-2 sera on the formation of rosettes by peritoneal cells of mice of the A and C57BL strains Serum to which the cells expod
Source of periton4 % Of cells 19SEAC
CS7BL anti-A/J
A/1 All
NMS
AIJ
NMS
A/J NMS C57BL A/J antiC57BL C57BL C57BL anti-A/J
15.7 f 2.1a) 2.9 f 0 . 3 13.1b) 3.6 23.p) 3.9
19SEA
3.0 f 0 . 7 2.8 f 0 . 6 2.4 1.4 3.9 5.1
rwtter 7SEAC7SEA
27d) 28.1
25.7 22.0
a) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. Mean of 12 experiments SE. b) 19 S antibody: first fraction of antiserum obtained by filtration through a Sephadex G200 column. Mean of two experiments. c) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. Mean of two experiments d ) 7 S antibody: rabbit serum bled 7 days after 3 injections of SRBC.
*
Table 4. The effect of anti-H-2 serum on the formation of EAC rosettes by either unfractionated suspensions of strain A peritoneal cells or by nonadherent peritoneal cells
Serum kited
UnfrlCtkMtCd psritonul cells suspendon
NMS
9.0“)
C57BLantbA
2.3
121 21
29
rosettes by BALB/c peritoneal cells. When C57BL anti-A serum was absorbed with C3H spleen cells, its activity on C3H peritoneal cells was completely eliminated, yet maintained its capacity of reducing the formation of EAC rosettes by BALB/c cells. Conversely, absorption of the serum with BALB/c spleen cells eliminated its activity on BALB/c peritoneal cells, without affecting its activity on C3H peritoneal cells. A/J anti-BALB/c serum was found t o have a strong inhibitory effect on the formation of 19 S EAC rosettes by BALB/c peritoneal cells (Table.5). This serum contains antibodies against K end H-2 specificities, and against “Ia antigens” [ 121. Antibodies to Ia antigens have recently been shown to block EAC receptors specifically (Schlesinger and Chaouat, unpublished data). A/J anti-BALB/c serum absorbed with C57BL spleen to remove antibodies to Ia antigens was found to retain its high efficiency in modulating the EAC receptors on BALB/c peritoneal cells. These results indicate again that EAC receptors on the surface of peritoneal cells are associated with K end H-2 antigenic specificities. Exposure of peritoneal cells t o either @or Ly antisera had no effect on the formation of EAC rosettes (Table 5 ) . Table 5. The effect of various antisera on the formation of 19 S EAC rosettes by peritoneal cells
Nodhecent puitoneal
ceh 5.5 5.8
9.0 7.4
NMS
NMS
BALBlc BALBIc BALB/c C3H C3H A1J
C57BL antl-A
UJ
NMS
C3H C3 H C3H
A antl-BALB/c
a) Percentage of cells forming rosettes Each vertical column represents results obtained in a single experiment.
C57BL anti-A
NMS C57BL antM
removed from the peritoneal cell suspension the remaining, nonadherent cells were not affected by exposure to anti-H-2 sera (Table 4).
Anti+ serum
3.2. Specificity of the sites linked to EAC receptors Exposure of peritoneal cells of the A/J strain to the DBA/2 anti-C3H serum inhibited the formation of EAC rosettes t o a much greater extent than exposure to the C3H anti-DBA/2 serum. This difference could not result from any difference in the antibody activity of these sera, as reflected in their cytotoxic titers with cells of the A strain (Table 1). The different activity of the sera could result from the fact that DBA/2 anti-C3H serum reacts with K end H-2 specificities on strain A cells, whereas the C3H anti-DBA/2 serum reacts with D end H-2 specificities on strain A cells [ 111. If this interpretation is correct, it would seem that the EAC binding site on peritoneal cells is associated more intimately with K end H-2 specificities than with D end H-2 specificities. Further support for this assumption was gained from experiments with the C57BL anti-A serum, a serum containing antibodies against multiple specificities o f both the D end and K end H-2 regions. While C57BL anti-A serum had a similar cytotoxic effect on C3H and on BALB/c target cells(Tab1e l ) , it had a different effect on the formation of EAC rosettes by cells of both strains. C57BL anti-A serum inhibited the formation of 19 S EAC rosettes as effectively in C3H peritoneal cells as in strain A peritoneal cells (Table 3), but had only a weak inhibitory effect on the formation of EAC
Anti-Ly m m
18.6 2.1 11.0 18.5 3.8 14.2 4.4 18.2 15.3 16.6
9.0 2.1 8.0 13.8 2.3
a) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. Each vertical column represents results obtained in a
single experiment. 3.3. Nature of the association o f H-2 antigens and EAC receptors The concomitant disappearance of H-2 antigens and EAC receptors from the surface of peritoneal cells suggests that H-2 specificities, particularly of the K region, are closely associated with EAC receptors on the cell membrane. The nature o f this association was explored by studying the effect of various inhibitors on the suppression o f EAC rosette formation by anti-H-2 sera. Exposure of peritoneal cells to anti-H-2 sera, under conditions which would not allow modulation of H-2 antigens [ 1 , 21 also prevented the disappearance of EAC receptors from the cell surface. Thus, keeping peritoneal cells on ice during their exposure to anti-H-2 sera, or exposing them to anti-H-2 sera M NaN3, completely preat 37 OC in the presence of vented the inhibitory effect of anti-H-2 sera on the formation
Eur. J. Immunol. 1975. 5: 27-32
M. Schlesinger and M. Chaouat
30
Table 6. The effect of exposure of strain A peritoneal cells to antiH-2 sera at low temperature or in the presence of NaN3 on their capacity to form rosettes Exp. Serum to which Type of rosette no. the cells wen: exposed tested NMS
1
NMS NMS NMS
C57BL anti-A C57BI anti-A C57BL anti-A C57BL anti-A
70 of cells forming
3 7 W
19 s E A C ~ ) 20.0 5.2 19 S EA 19.8 7 S EAC 7SEA 17.0 19 S EAC 4.2 3.6 19 S EA 25.2 7 S EAC 7 S EA 29.8
NMS 19 S EAC NMS 19 S E A C3H anti-DBA/2 19 S EAC C3H anti-DBA/?19 S EA DBA/2 anttC3H 19 S EAC DBAI2 antLC3H 19 S EA C57BL anti-A 19 S EAC C57BL anti-A 19 SEA
I1
rosettes 4’0) N~N~C) 17.4
28.8
26.2 4.8 23.8 6.4 21.2 2.4 19.0 4.8
7.2
14.2 4.4 6.0 5.2
3.8 5.4
28.4 6.6 27.0 5.4 32.8 5.8 27.3 6.8
of I9 S EAC rosettes (Table 6). These experiments indicate that although H-2 antigens and EAC receptors may be in close topographical proximity on the membrane of peritoneal cells, they are not carried by identical cell surface structures. They seem to be located far enough from each other, so that H-2 antibody does not interfere by steric hinderance with the formation of EAC rosettes. Microtubule depolymerizing agents, such as vinblastine, are relatively weak inhibitors of H-2 modulation [2, 31. I t was previously found [2], and confirmed in the present study, that at a concentration of 1 0 - 5 M , vinblastine does not affect the modulation of H-2 antigens on peritoneal cells by specific antiserum. However, this concentration of vinblastine completely prevented the disappearance of EAC receptors from Table 7. The effect of cxposure of strain A peritoneal cells to antiH-2 serum in the presence of vinblastine on their capacity to form rosettes
v;,of cells forming rosettcsa’ No inhibitor Vinblastinr added addcd
Serum to u hich the cells \vcrc rxpocd
Typc of
NMS
19 S EACb’ 23.0‘)
NMS
19 S E A
5.8
5.0
3.8
5.8
C57B1 anti-A
19 S EAC
3.4
C 5 7 B L anti-A
19 S I.:A
3.6 5.4
26.4 2.6
18.4 3.8
21.2
3.0
21.6‘)
15.6
a) Prior to the rosette test the peritoneal cells were exposed to various sera for 2 h at 3 7 O C in the presence or absence of 10-4M
vinblastine. b) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. c) Each vertical column reprcsents results obtained in a single experi-
ment.
4 Discussion The present study demonstrates that exposure of peritoneal cells to anti-H-2 sera, a procedure leading to the disappearance of H-2 antigenicity from the surface of the cells [ 1-31, results in the concomitant disappearance of EAC receptors from the membrane of peritoneal cells. EA receptors were unaffected by modulation of the H-2 antigenicity. Antisera reactive with K end H-2 specificites had a stronger effect on EAC receptors than antisera reactive with D end H-2 specificities. The different activity o f various H-2 antisera may result from three factors: ( I . ) different concentration of anti-H-2 antibodies; (2.) the presence of either K end or D end H-2 antibodies; and (3.) contamination by antibodies to non-H-2 antigens and particularly by anti-la antibodies [ 121.
3.2
26.0 24.0 15.8 2.3 21.6 24.4
a) Prior to the rosette test the peritoneal cells were exposed to various sera for 2 h at 37 OC. b) Prior to the rosette test the pcritoneal cells were exposed to various sera for 2 h at 4 O C . c) Prior to the rosette test the peritoneal cells were exposed to various sera for 2 h at 37 O C in the presence of 10-4M NaN,. d) 19 S antibody: rabbit serum bled 5 days after a single injection of SRBC. 7 S antibody: rabbit serum bled 7 days after 3 injections of SRBC.
roscttr tcstcd
the membrane of peritoneal cells upon exposure to anti-H-2 serum (Table 7). In contrast to vinblastine, 10- ’M colchicine failed to prevent the inhibitory effect of anti-H-2 serum on the formation of EAC rosettes.
Antisera against Ia antigens have recently been shown to react specifically with EAC receptors (Schlesinger and Chaouat, unpublished data). In contrast to the effect of H-2 antisera, the effect of anti-la antibodies could not be inhibited by exposure of the cells t o various metabolic inhibitors (Schlesinger and Chaouat, unpublished data). While the antisera used in the present study may contain antibodies to Ia antigens, these antibodies were not required for the effect of H-2 antisera on EAC receptors. A/J antiBALB/c serum, which in addition t o antibodies to K end H-2 antigens also contains la antibodies, retained its strong activity on EAC receptors even after absorption with C57BL spleen presumably removed all la antibodies. There was no correlation between the cytotoxic titer obtained with various sera and their effect on the formation of EAC rosettes. Thus, C57BL anti-A serum had a similar cytotoxic effect on C3H and BALB/c target cells, but inhibited the formation of EAC rosettes by C3H cells to a greater extent than by BALB/c cells. C3H anti-DBA/2 serum had a weaker cytotoxic effect on strain A cells than DBA/2 antiC3H serum, yet the former serum was more effective in suppressing the formation of EAC rosettes by strain A cells. The results of the present study support the conclusion that EAC receptors on the membrane of peritoneal cells are closely linked to H-2 antigenic specificities, particularly those of the K end. Previous studies have shown that K end and D end H-2 specificities constitute distinct entities on the cells’ membranes. Exposure of lymphocytes t o antisera against either K end or D end H-2 determinants leads t o redistribution and “capping” o f only the corresponding antigenic specificities on the cells’ membranes, while the distribution of the other H-2 specificities remains unaltered [ 131. Similarly, exposure of peritoneal cells to antisera against some H-2 specificities leads to disappearance of the corresponding H-2 specificities, without affecting the expression of other H-2 specificities (Schlesinger and Chaouat, unpublished data, [ 21). In studies of the modulation of the TL antigen of thymus and leukemic cells [4] it was shown that several TL specificities modulate concomitantly upon exposure of the cells to antibodies against only one of the TL specificities [ 141. On the other hand, TL antigens constitute distinct entities, since antibodies to, say TL-I, will not inhibit the capacity of cells possessing the TL-I and TL-3 antigenic specificities
Eur. J. Immunol. 1975. 5: 27-32
t o absorb antibodies to TL-3 determinants [ 141. There are indications in the present study that although EAC receptors and H-2 antigenic determinants are closely linked on the cell surface, they constitute distinct entities. Exposure of peritoneal cells to anti-H-2 sera at 4 OC or in the presence of NaN3, procedures which prevent the modulation of H-2 antigens [ 1, 21 but d o not prevent the coating of the cells with H-2 antibodies, did not lead to any reduction o f the capacity of the cells to form EAC rosettes. Moreover, exposure of peritoneal cells to vinblastine at a concentration of 1 0 - 5 M , which does not inhibit H-2 modulation [ 1, 21, was found t o prevent the disappearance of EAC receptors from the surface of the cells. Although colchicine at a similar concentration was without effect it seems possible that the close association of EAC receptors and H-2 antigens is maintained by the inte. 'ity of the microtubule system. Studies by Berlin and coworkers [ 151 have shown that the cell surface o f macrophages operates as a functional mosaic, in which membrane transport sites are separated from membrane parts which are internalized during phagocytosis. This functional heterogeneity of the cell membrane is broken down by microtubule depolimerizing agents [ 151. Similarly, microtubules seem t o affect the cell surface arrangement of membrane constituents which determine the agglutinability of the cells by concanavalin A [16-181. Mantovani et al. [ 191 presented evidence indicating that the EAC and EA receptors on the surface of macrophages may play different, and even opposing, roles during the immune reaction. In early stages of the immune response, complexes of antigen, IgM antibody, and complement bind t o EAC receptors on the surface of macrophages. These complexes remain bound to the cell surface of macrophages so that the antigen is presented as a continuous stimulus to other cells involved in the immune response. In this way, the EAC receptors on the cell membrane of macrophages may amplify the immune reaction in the early stages o f the immune reaction. In contrast, during the secondary response, EA receptors bind complexes of antigen and IgC antibody which are then rapidly interiorized and degraded. Thus, EA receptors may serve to eliminate any excess antigenic stimulus during the secondary immune response. The extreme efficiency of peritoneal macrophages in antibody-induced suppression of the immune response has been amply documented [20, 2 I]. Recent studies have suggested additional amplifying functions for EAC receptors. Feldmann has shown that monomeric IgM sequestered from the cell surface of T lymphocytes becomes attached t o the surface of macrophages, and that macrophages armed with T cell-derived immunoglobulin then cooperate with B cells in the production of antibody to thymuodependent antigens [22, 231. It now seems that the macrophage receptor for T-derived immunoglobulin is identical with the EAC receptor, since antisera against mouse C3 abolish the in vitro immune response to thymusdependent antigens (Feldmann, M., personal communication). Dukor and Altman provided evidence that while B lymphocytes become triggered by antigen binding, an additional signal for the response of B cell may be the attachment o f activated C3 t o the EAC receptors [24]. The present demonstration of an association between histocompatibility antigens and EAC receptors may have implications on the genetic regulation of the immune response. Many Ir genes in the mouse are linked t o the K end o f the H-2 locus [25]. Mice genetically deficient in their response to certain antigens produce normal levels of IgM antibody during the primary immune response, but show an impaired
Association of H-2 antigens and EAC receptors
31
production o f IgC antibodies in the secondary response [25 1. It is assumed that Ir genes affect the formation of specific antibodies through their effect on the immune function of T lymphocytes [ 261. Shevach and Rosenthal[27] recently showed that macrophage-bound antigens can activate T lymphocytes only when the macrophages are derived from animals possessing Ir genes determining an immune response t o these antigens. They suggested that antigen recognition sites on T lymphocytes are physically related t o sites on these cells which determine their binding t o macrophages, and that both are linked to histocompatibility antigens. The results of the present study suggest an additional mechanism for the regulation o f immune response by Ir genes It is postulated that Ir genes may affect the functional capacity of EAC receptors on macrophages. This hypothesis is based on the assumptions that EAC receptors are important for the cooperation o f T lymphocytes with B lymphocytes and for the switch-over from IgM to IgG antibody produotion. It is possible, then, that K end H-2 antigens or antigenic determinants specified by H-2 linked Ir genes, which are located in close proximity to EAC receptors on the cell surface, may affect the attachment of antigen-antibody-C complexes to macrophages Only when the EAC receptors are adjacent t o appropriate cell surface structures determined by Ir or H-2 genes, will the complexes be bound in a way suitable for triggering the reaction of immunocompetent B lymphocytes and possibly also of T lymphocytes. Received February 9, 1974; in revised form August 31, 1974. 5. References 1 Schlesinger, M. and Chaouat, M., Tissue Antigens 1972.2: 427. 2 Schlesinger, M. and Chaouat, M., Transplant. Proc. 1973. 5: 105. 3 Schlesinger, M., Biomedicine 1973. 18: 437. 4 Boyse, E.A., Stockert, E. and Old, L.J., Proc. Nut. Acad. Sci US 1967. 58: 954. 5 Nussenzweig, V., Bianco, C., Dukor, P. and Eden, A., in Amos,
D.B. (Ed.) Progress in Immunology, Academic Press, New York 1971, p. 73. 6 Tizard, I.R., Bacteriol. Rev. 1972. 35: 365. 7 Nussenzweig, V. and Lay, W.H.,J. Exp. Med. 1968. 128: 991. 8 Schlesinger, M. and Hurvitz, D., Transplantation 1969. 7 132. 9 Schlesinger, M., Cohen, A. and Hurvitz, D., Isr. J. Med. Sci 1969. 5: 235. 10 Stjernsward, J., Jondal, M., Vanky, F., Wigzell, H. and Sealy, R., Lancer 1972. i: 352. 11 Klein, J. and Shreffler, D.C., Transplant. Rev. 1971. 6: 3. 12 Sachs, D.H. and Cone, J.L.,J. Exp. Med. 1973. 138: 1289. 13 Neauport-Sautes, C., Lilly, F., Silvestre, D. and Kourilsky, F.M., J. Exp. Med 1973.137: 511. 14 Old, L.J., Stockert, E., Boyse, E.A. and Kim, J.H., J. Exp. M d . 1968. 127: 523. 15 Ukena, T.E. and Berlin, R.D., J. Exp. Med. 1972. 136: 1. 16 Inbar, M., Ben-Bassat, H. and Sachs, L., Proc. Nut. Acad Sci US 1971. 68: 2748. 17 Berlin, R.T. and Ukena, T.E., Nature-New Biol. 1972.238: 120. 18 de Petris, S., Raff, M.C. and Malucci, L., Nature-New Biol. 1973. 244: 275. 19 Mantovani, B., Rabinovitch, M. and Nussenzweig, V., J. Exp. Med. 1972. 135: 780. 20 Schwartz, R.S., Ryder, R.J.W. and Cottlicb, A.A., Progress in Allergy 1970. 14: 81.
A. Polliack, U. Hammerling, N. Lampen and E. de Harven
Eur. J. Immunol. 1975. 5: 32-39
21 Haughton, G., Cell Immund. 1971. 2 : 567. 22 Feldmann, M., J. Exp. Med. 1972. 134: 103. 23 Feldmann. M. and Nossal, G.J.V., Transplant. Rev. 1972. 13: 3. 24 Dukor, P. and Hartmann, K.U., CelL Immund. 1973. 7: 349.
25 Grumet, F.C., Mitchell, G.F. and McDevitt, H.O.,Ann. New York Acod. Sci 1971. 190: 170.
32
A. PoIIiack,+ u. Hammerling, N. Lampen and E. de Harven
26 Benacerraf, B. and McDevitt, H.O., Science 1972. 175: 233. 27 Shevach, E.M. and Roscnthal, A.S., J. Exp. Med. 1973. 138: 121 3.
Surface morphology of murine B and T lymphocytes: A comparative study by scanning electron microscopy"
Memorial Sloan-KetteringCancer Center, New and Department Of Hematology'A variety of murine lymphocytes of known B or T derivation obtained from and Hebrew University different lymphoid organs were prepared for scanning electron microscopy University Medical School, Jerusalem+ (SEM) by the critical point drying method after collecting the cells by aspiration onto silver membranes. Comparison of SEM appearances of cells prepared by this technique and serological classification according to surface antigens showed that most T cells had smooth surfaces with few microvilli, while many B lymphocytes were moderately to markedly villous. Further evidence for the above correlation was obtained by examining thymic cells and enriched B or T cell populations. Thymic cell suspensions containing less than 5 % B cells showed over 8 0 5% generally smooth cells by SEM. Enriched T cell populations, obtained by mass cytolysis of lymph node preparations with anti-la or anti-Ig sera or by purification through nylon fiber columns, contained over 85 % T cells, and more than 75 % of them were of the smooth cell type. A similar correlation was noted for enriched B cell populations obtained by
cytolysis of lymph node cells with anti-Thy-I serum, and by lysis of EAC-rosettes. Over 90 % of these cells were identified as B cells by immunologic methods and approximately 7 5 % had moderate to markedly villous surfaces. The 1 5 5% difference can be accounted for by the existence of a subpopulation of smooth B cells. Direct observation of EAC-rosettes confirmed that most B cells had moderate t o large numbers of surface microvilli and that less than 10 % were smooth. It is possible that some of the smooth cells seen in enriched B cell populations may represent precursors or B lymphocytes at different stages of differentiation. These results indicate that murine T and B lymphocytes, like their human counterparts, can be recognized in many cases under the SEM on the basis of their surface morphology. Smoother B and more villous T cells are difficult to classify by SEM without parallel immunologic identification.
1. Introduction Thymus-derived (T) and bone marrow-derived (B) lymphocytes can be distinguished by their different surface antigens and receptors [ I , 21. Murine T cells carry on their surface the alloantigens Thy-1 (formerly named 0)[3, 41, the Ly series of alloantigens [ 5 , 61, and TL which is expressed only on thymocytes [ 7 ] . B cells have the alloantigens Ia and Pc [8-111, the heteroantigens, mouse-specific B lymphocyte antigen [I 8091 * ) Supported by Grant CA-08748 from the National Cancer Institute, American Cancer Society Grant DT-31, The Frank Strick Foundation, National Science Foundation Grant GB-34299, and USPHS International Postdoctoral Research Fellowship awarded to Dr. A. Polliack.
Correspondence: E . de Harvcn, Memorial Sloan-KetteringCancer Center, New York, N.Y. 10021, USA Abbreviations: slg: Surface immunoglobulin MBLA: Mouse-specific B lymphocyte antigen MSPCA: Mouse-specific plasma cell antigen TEM: Transmission electron microscopy SEM: Scanning electron microscopy SRBC: Sheep erythrocytes EAC: Sheep erythrocytes coated with rabbit anti-SRBC antibody and complement TDL: Rat thoracic duct lymphocytes LN: Lymph node Anti-lgM/FI: Fluoresceinated rabbit antibody to mouse IgM
(MBLA [ 121) and mouse-specific plasma cell antigen (MSPCA [ 131 1. Furthermore, B cells have readily detectable surface immunoglobulin (slg) [ 141 and also bear receptors for the Fc fragment of IgG [ 151 and for the C 3 component of complement [ 161. Despite these differences in antigen patterns the two types of lymphocytes have not been consistently distinguished by routine light or electron microscopy unless cells were first labeled by their respective surface markers [ 171. However, recently differences in human B and T lymphocytes have been recognized by scanning electron microscopy (SEM) and a spectrum of lymphocyte surface architecture has been described [ 18-20]. Many T cells display smooth surfaces with only small numbers of surface microvilli, whereas the majority of B cells have larger numbers of microvilli. However, some B and T lymphocytes have similar surface features and cannot be assigned to either class accurately [ 18-20]. The present investigation was undertaken to define the surface architecture of murine lymphocyte populations of known derivation, and to establish whether previous findings on human lymphocytes [ 18-20] could be extended to murine cells.
