Role of Autorosette Forming Cells in Antibody Synthesis in vitro: Suppressive Activity of ARFC in Humoral Immune Response

Immunobiol., vol. 166, pp. 1-11 (1984)

Original Papers

Department of Basic and Clinical Immunology and Microbiology, Medical University of South Carolina, Charleston, South Carolina, U.S.A.

Role of Autorosette Forming Cells in Antibody Synthesis in vitro: Suppressive Activity of ARFC in Humoral Immune Response'} N. KHANSARI, M. PETRINI, F. AMBROGI" P. GOLDSCHMIDT-CLERMONT, and H. H. FUDENBERG Received May 20, 1983 . Accepted July 12, 1983

Abstract The role of autologous rosette forming cells (ARFC) in humoral immune responses was studied using an in vitro system. While depletion of ARFCs from PBL resulted in a significant increase of either tOtal IgG or anti-IT IgG, addition of these cells to the system decreased the production of immun"oglobulin to a level comparable to that of unfractionated PBL. The majority of the ARFCs reacted with anti-Leu2a and anti-LeuS. In contrast, the majority of non-ARFCs reacted with Leu3a and only 10 % with Leu8 monoclonal antibodies. Stimulation of unfractionated PBL with concanavalin A (ConA) resulted in an increase of the ARFC population. ConA stimulation also increased the number of cells reactive with anti-Leu2 and/ or anti-LeuS. The autorosette population had a higher purine nucleoside phosphorylase (PNP) content than the non-ARFC population. Although the ARFC suppressed synthesis of antibody by B cell in vitro when they were mixed with either autologous or allogeneic B cells, a marked proliferation of non-B cells was evident. We conclude that at least two different subpopulations of T cells are capable of forming rosettes with autologous red blood cells.

Introduction Binding of a large population of peripheral blood lymphocytes (T cells) to sheep red blood cells (SRBC) has been known since the early 1970s (1).

*Publication no. 600 from the Department of Basic and Clinical Immunology and Mi­ crobiology, Medical University of South Carolina. Research supported in part by USPHS Grant 5T32AI07063 and by S.c. State Appropriations Grant CR-13. 1 Patologia Medica, Universitl di Pisa, Italy Abbreviations: ARFC = autologous rosette forming cell; Con A = concanavalin A; PNP = purine nucleoside phosphorylase; SRBC = sheep red blood cells; PWM = pokeweed mitogen; E-RFC = sheep erythrocyte rosette forming cell; HBSS = Hank's balanced salt solution; PBS = phosphate-buffered saline.

2 . N.

KHANSARI

et a1.

During studies of lymphocyte surface marker and binding capabilities to erythrocytes of the other species (2), it was noted that a small population of human peripheral lymphocytes also formed rosettes with autologous eryth­ rocytes (3-6). This phenomenon has been also reported in mice (7) and bovine (8). Unlike in man, autorosette formation in mice is found in both T and B cells (9). These findings generated a great deal of controversy as a result of attempts to characterize autorosette forming cells (ARFC) mor­ phologically (10) and functionally (11-14). SUGAWARA and PALACIOS (5) reported that ARFCs were precursors of cytotoxic cells against alloanti­ gens, and also helper cells capable of assisting B·lymphocyte differentiation into plasma cells following a signal from interleukin-2. Similarly WEKSLER et aJ. (16) claimed that a soluble factor released during autologous mixed lymphocyte reactions (MLR) facilitated generation of cytotoxic lympho­ cytes. In contrast, KUMAGAI et aJ. (14) found that ARFCs blocked the development of cytotoxic cells in MLR. Furthermore, LAMBERMONT et aJ. (17) showed that human autologous and allogeneic rosette forming cells belong to one T cell subpopulation. SCHEFFLE and SWARTZ (18) found that ARFC enriched populations responded vigorously to autologous stimula­ tion, but poorly to allogeneic stimuli. They also found that approximately 16 % of ARFC expressed receptors for IgG and an average of 66 % expressed receptors for IgM after overnight culture. The biological significance of ARFCs is related to their ability to proliferate in reaction with autologous B cells (19-21), such proliferation depends on immunological specificity and memory (22), both of which are considered to represent a measure of ARFC immunoregulatory potential.

This became more significant after it was found that ARFC numbers are decreased in patients with cancer (23), chronic lymphocytic leukemia (24), with selective IgA deficiency (25), and in several inbred strains of mice genetically predisposed to autoimmune disease and excessive Iymphoprolif­ eration (2r.-:28). However, most of the investigators studied the prolifera­ tive response of ARFCs in reaction with autologous and allogeneic non-T

cells but very little attention was paid to their effect on antibody synthesis of autologous or allogeneic B cells. In this report, we studied the effect of aurorosecte forming cells on B cells stimulated with pokeweed mitogen (PWM) and antigen (tetanus toxoid). We have also compared the purine nucleoside phosphorylase content of ARFCs and non-ARFCs since this enzyme is found to be associated with lymphocyte function (29, 30).

Materials and Methods IsoJatjo~l

of peripheral blood lymphocytes

Heparinized peripheral blood was collected from healthy volunteers. Lymphocytes were separated by Ficoll-Hypaque density centrifugation. Sheep red blood cell rosette forming cells (E-RFC) and non-E-RFC were separated as described previously by GRAYSON et al. ( 31 ).

Role of Autorosette Forming Cells in Humoral Immune Response

3

Separation ARFC E-rosette forming cells (f cells) were resuspended in Hank's balanced salt solution (HBSS) containing 10% autologous serum in a density of 5 x 106/ ml. One volume of a 10% (three times washed) autologous RBC blood cell suspension in HBSS was added to 3 volumes of autologous T cells (T cell to red cell ratio 1 :50), then followed by 5 min centrifugation at 4 O( and 150 X g. The mixture was incubated overnight at 4°C undisturbed. After incubation, the pellet was resuspended very gently by end-over mixing and ARPCs were enumerated using a hemocytometer grid. The cell suspensions were then layered over Ficoll-Hypaque in a 1:1 ratio and centrifuged at 850 X g at 4°C for 45 min. The interface layer (non-ARFC) was recovered and washed 3 times with HBSS and resuspended in RPMI-1640 medium. The rosetting cells (ARFC) in the pellet were treated with ammonium chloride Tris buffer (Trimza base 0.17 M plus ammonium chloride 0.16 M) to disrupt red cells. Following this treatment, cells were washed 3 times with HBSS and resuspended in RPMI-164D medium.

Culture conditions for in vitro antibody synthesis Isolated lymphocytes were suspended in RPMI-164D buffer~d with NaHCO) and sup­ plemented with L-glutamine (2 mM/ml), penicillin (100 U/ ml), streptomycin (100 ug/m l), HEPES buffer (25 mM), and 10 % heat inactivated FeS, at a concentration of 2 x 106 cells per ml. One half ml of the cell suspension was placed in each well of a 24-well flat bottom microtiter plate (Costar, Cambridge, MA) and incubated at 37°C in a 5 % CO 2 incubator for 8 days. Cultures were supplemented daily with 80 ul of nutritional cocktail as described by MISHFJ,J. and DUTION (16), consisting of 10 % Fes. Pokeweed mitogen (Gibco, Grand Island Biological, Grand Island, New York) was added to the culture at a final concentration of 1:50 at day zero of incubation. Tetanus toxoid-Sepharose 4B was added on day zero to the cultures at 70 ng/10 6 cells based on protein content. The culture supernates were recovered at the end of incubation and assayed for antibody content by the enzyme linked immunosorbent assay (EI.ISA) as described earlier (32).

Mitogen stimulation of lymphocytes Peripheral blood lymphocytes (PBL) were resuspended in RPMI to make a 2 X 106 cells/ml cell suspension. Concanavalin A (ConA) was added to the cell suspensions in a tissue culture flask at a concentration of a 6 ugl ml of the cell suspension, followed by incubation at 37°C and 5 % CO 2 for 72 hrs. Following incubation, cells were washed 3 times with HBSS and AR.FCs were separated.

Mixed lymphocyte reactions Both autologous and allogeneic MLR were performed as described by FOURNIER and CHARREIRE (33 ). Briefly, 100 u1 of responder cells suspensions (1 x let) was mixed with an

equal volume and number of irradiated (2000 R) stimulating cells (E-rosette and adherent cell depleted PBL) in RPM!. Each determination was performed in triplicate and consisted of unfractionated T celis, ARFC depleted T cells and ARFC enriched T cells as responder cells stimulated by either autologous or nonrelated donor non-E-RFC. After a 5 day incubation at 3rC in humidified 5 % CO2, the cultures were pulsed with 1.2 uCi of lH-thymidine (New England Nuclear, Boston, MA) and incubated for an additional 18 hrs. The cells were harvested with an automated cell harvester (Dynatech Instruments, Inc., Santa Monica, CA) and the radioactivity incorporated into the cells counted by a liquid scintillation counter. Results were expressed as the difference between COunts from cultures stimulated by non-E­ RFC and counts from background cultures consisting of 2 x 10(, irradiated non-E-RFC only.

Indirecr immunofluorescence staining Monoclona1 antibodies were purchased from Becton Dickinson (Sunnyvale, CA). Cells were resuspended in phosphate-buffered saline plus HBSS containing 2 % fetal calf serum and 0.01 % sodium azide (PBS-I-IBSS) to make a 2 X 107 cclls/ml ccll.~uspension. The monoclonaf

4

N. KHANSART et a1.

antibodies were adjusted to a 1 uglml solution in PBS-HBSS. 50 ul of cell suspension was mixed with 50 ul of diluted antibodies in a round microtiter plate and incubated for 30 min at 4°C. Cells were washed 3 times with PBS-HBSS and then 50 ul of fluorescent labeled goat anti­ mouse IgG diluted I :30 were added to the cells. Cells were incubated for another 30 min at 4°C and followed by 3 washes with PBS-HBSS. After the final wash, cells were resuspended in 50 ul of 50 % glycerine in PBS and stained cells enumerated microscopically.

Deterrm'nation of PNP activity Purine nucleoside phosphorylase activity was assessed as described by VAN LAARHOVEN et a1. (34) with minor modifications. Briefly, lymphocytes were resuspended in PBS containing 4 % bovine serum albumin at a density of 1--6 X 106/m1. One ul of the cell suspension was placed in each well of a Terasaki plate (Histo-plate, Dynatech Laboratories, Inc., Alexandria, VA), frozen at - 20°C, and lyophilized overnight. All determinations were carried OUt in quintuplicate. The reaction was started by adding 3 ul of reaction mixture consisting of 10 ul of 400 uM 14C-hypoxantin, 10 ul of 5 mM of ribose I-phosphate, 10 ~l of 0.5 M Tris buffer containing to mM EDTA, and I ~l of 10 % Triton X-tOO. The reaction was allowed to proceed for 30 min at 3rC; the substrate then was separated from the product by means of thin-layer chromatography on PEl-cellulose F (Merck-Darmstadt, Germany) using 0.2 M NaCI as solvent. The unlabeled metabolites were co-chromatographed to allow the recognition of spots under UV light. After chromatography, the spots were cut out and radioactivity was counted in a liquid scintillation counter. 'The enzyme activity was calculated from the fonnula: cpm product Enzyme activity = cpm product + cpm substrate The activity was expressed as n mole of product formed 110& cellslhr. The significance of the differences was determined using the Wilcoxon's T test.

-==-==icc:::-cc-:-=--:::==-=-­

Results Effect of ARFC in antibody synthesis

As previously reported (32), human peripheral lymphocytes can be stimulated by PWM and tetanus toxoid coupled to Sepharose 4B for production of antibody to tetanus toxoid. Using this system, we studied the effect of removal or addition of ARFC to the system on the synthesis of total and antigen specific IgG. Table 1 summarizes the results from these experiments. Depletion of ARFC from PBL increased the production of specific antibody more than 30 % and total IgG greater than 44 %. Readdi­ tion of ARFC, however, resulted in a response comparable to that of unfractionated PBL. Furthermore, removal of non-ARFC abolished pro­ duction of either total or specific IgG more than 70 %, suggesting that helper activity originated from non-ARFC, while ARFC exerted suppres­ sor activity. Determination of surface markers

Immunofluorescence staining of different fractions of PBL by a panel of monoclonal antibodies directed to T cell surface markers showed that the majority of ARFCs stained with anti-Leu2a (suppressor/cytotoxic) and Leu8, while the majority of non-ARFCs reacted with anti-Leu3a (helper/

Role of Autorosette Forming Cells in Humoral Immune Response . 5 Table 1. Effect of ARFC in in vitro IgG synthesis ng IgG/ m1l106

Cell population' and treatment

Total

Anti IT

1,965 ± 310

49 ± 7b

3,454 ± 711

69 ± 12

T-edl, + PWM + IT

3,012 ± 513

61 ± 10

4. B-cells + non-ARFC enriched + ARFC enriched + PWM + TT

1,413 ± 230

38 ± 10

1. Unfractionated PBL

+ PWM + IT

2. ARFC depleted PBL + PWM + IT 3. B-cells

+

ARFC depleted

+ ARFC enriched + PWM + IT B-eell, + PWM + IT

5. B-cells

650 ± 117

18 ± 5

6.

311 ± 52

8± 5

• 0.5 ml of a cell suspension containing 2XI0(, cells per ml of autologous combined cells cultured in 24 well microuter plate for 8 days. Ratio of T cells to B cells was 3:1. b Mean value of 4 experiments ± S.E. Statistical studies: 2 vs 1: p < 0.001 for 3 vs 1: p < 0.001 for 4 vs 1: p < 0.001 for S vs 1: p < 0.001 for S vs 2: p < 0.001 for S vs 3: p < 0.001 for

total total total total total total

IgG production. IgG production. IgG and p < 0.08 for anti-IT production. and anti- TI production. and anti- TI production. and anti-IT production.

inducer) and only 10 % of them with Leu8 (Table 2). Concanavalin A stimulation, however, increased the population reactive with Leu2a but did not increase the populations of T cells reactive with Leu3a indicating that perhaps the ARFC population contains precursors of suppressor cells generated by Con A. Table 2. Determination of ARFC surface markers by monoclonal antibodies Percent reactivity with a

Cell populations Anti Leu 2a

Anti Leu 3a

Anti Leu 4

Anti Leu 8

PBL

15 ± 3

46± 5

65 ± 5

60± gb

PBL stimulated with ConA

38 ± 4

41 ± 5

73 ± 6

68 ± 5

E-rosette forming cells

26 ±3

54 ±6

95 ± 3

70± 9

2± I

2±1

ARFC enriched

55 ±3

35 ± 6

93 ± 5

60 ± 7

ARFC depleted

30 ±4

69 ± 5

90 ± 7

10± 3

Non-E rosette forming cells

3±I2

± I

• Reaction was determined by indirect immunofluorescence staining as described in Materials and Methods. b Mean value of 4 experiments ± S.E.

6

N.

KHANSARI

et al.

Table 3. Autologous and allogeneic mixed lymphocyte reaction of T- and B-cells Effector cella

CPM

I. Unfractionatcd T-cell

With autologous stimulator cell

With allogeneic stimulator cell

26,319 ± 1,150

21,412

2. ARJ.·C depleted T -cells

16,235 ±

3. ARPC enriched T-cells

49,671 ± 1,530

961

± 961 h

15,119 ±

831

50,376 ± 1,932

• 1 X 106 responder (T-cells) were incubated with same number of irradiated (2000 R) stimulat­ ing cells (non-E-roseue fanning, monocyte depleted cell) in a final volume of 0.2 ml in a flat bottom 96 well microtiter, plated for 5 days, and then pulsed with 3H-thymidine for 18 hours. b Mean value of 5 experiments ± S.E. Statistical studies: 2 vs 1: p < 0.001 for .proliferation of responder cells with autologous and allogeneic stimulator. 3 vs 1: p < 0.001 for proliferation of responder cells with autologous and allogeneic stimulator.

Proliferative response of ARFC

Human T cells can be stimulated by autologous irradiated non-T cells resulting in proliferation of T cells with a peak of thymidine incorporation on day 6 (35-38). Using this system, we studied proliferative response of

PBL

E~FC

ARFC NON-ARFC

Fig. 1. Activity of purine nucleoside phosphorylase enzyme content of different T lymphocyte subpopulations. Results are' expressed as mean value of 16 experiments ± S. E.

Role of Autorosette Forming Cells in Humoral Immune Response· 7

different fractions of the T cells with autologous and allogeneic irradiated B cells. As shown in Table 3, removal of ARFC from unfractionated T cells resulted in a 40 % lower response than that of unfractionated. In contrast, enrichment of ARFC resulted in a significant increase of proliferation (> 4S %). The magnitude of the T cell response to autologous B cells in each experiment, however, was comparable with that of corresponding experiments of T cell responses to allogeneic B cells (Table 3).

Purine nucleoside phosphorylase activity of different T cell subpopulations As previously reported (39, 40), the PNP enzyme content of lympho­ cytes is associated with the ontogeny of these cells; therefore, we took advantage of this marker to study any differences in enzyme content between ARFC and non-ARFC. As indicated in Figure 1, there was a significant (P < 0.01) increase of PNP activity in ARFC compared to either unfractionated PBL, un fractionated T cells, or non-ARFC. Discussion

The biological significance and characteristics of autologous rosette forming cells have been controversial since MICHLEM's original report more than a decade ~go (43). However, their proliferative responses in reaction with autologous or allogeneic non-T cells (12,13, IS, 41, 42) and their association with certain immunological defects (24-2S) demonstrate a potential for immunorcgulation. Indeed, 'the investigation reported herein demonstrates the suppressor function of ARFC on humoral immune responses. As shown in Table 1, co-culture of autologous B cells with ARFC depleted T cells produced significantly higher immunoglobulin (total or specific), while addition of ARFC lowered the production of total and specific IgG comparable to that of unfractionated PBL. Using a panel of monoclonal antibodies to known T cell surface markers, we further demonstrated that ARPCs have characteristics of suppressor cells. As shown in Table 2, the majority of ARFC enriched populations were reactive with anti-Leu2 (suppressor/cytotoxic marker). In contrast, the majority of ARFC depleted populations which were able to help autolo­ gous B cells for antibody production were reactive with anti"Leu3 (helper! inducer marker). Furthermore, greater than 60 % of ARFC enriched popu­ lations were reactive with anti-LeuS while only 10 % of ARFC depleted populations were reactive to this monoclonal antibody. This confirms indirectly our results since it has been reported that Leu3+,S- T cells have helper activity in antibody synthesis while Leu3 +,S+ T cells fail to induce antibody production when co-cultured with autologous B cells (44). Human peripheral blood T lymphocytes can be activated by ConA to become suppressor cells, but not to exert detectable helper effects (45). ConA activation also enhances the auto rosette population markedly (46).

8 . N. KHANSARI et a1.

We have also found that ConA stimulation of PBL increases the population

of Leu8+ cells; therefore, there is a direct association of suppressor activity

and the LeuS marker which is expressed on ARFC but not on non-ARFC. It has been reported that ARFCs were post-thymic precursors that could be differentiated to helper or suppressor cells following stimulation by ConA or serum thymic factor (47, 4S). Measuring purine nucleoside content of ARFCs, we found this unlikely since maturation of T cells is shown to be related to increase of PNP activity (39, 40). Furthermore, increased percent­ ages of ARFC in peripheral blood of thymectomized patients (11) and aged individuals (7) would exclude the possibility of ARFCs being immature post-thymic cells. In fact, low levels of sialic acid in the membrane of these cells (11) implies their being aged cells, as we reported earlier, that aging red blood cells (49) and platelets (unpublished data) have significantly lower sialic acid in comparison to young cells. However, as is shown in Figure 1,

there is a significant difference between PNP content of ARFC and non­ ARFC and E-RFC. This suggests a possible correlation between suppressor activity and PNP content of the lymphocytes. Although autologous rosette forming cells had suppressor activity for antibody synthesis, they had a significantly higher proliferative response (> 3 fold) in reaction with either autologous or allogeneic irradiated non-E­ rosette forming cells (B-cell enriched) than that of non-ARFC (Table 3). This indicates that the responder cell in MLR is ARFC and confirms what other investigators have found; also, there are at least two populations in

ARFCs, one of which has suppressor activity and the other is a responder cell in MLR. In fact, we have found that about 40 % of ARFC are LeuS-. Since Leu8+ cells are suppressor celis, it is likely that LeuS- which form rosette with autologous RBC are responder cells in MLR. Acknowledgment Authors thank Dr. HAL WHfITEN for editorial assistance.

References 1. LAY, W. H., N. F. MENDES, C. BIANCO, and V. NUSSENZWEIG. 1971. Binding of SRBC to a large population of human lymphocytes. Nature 230: 531. 2. JOHANSEN, K. 5., T. S. JOHANSEN, and D. N. TALMAGE. 1974. T cell rosette fonnation in primates, pigs, and guinea pigs. J. Allergy Clin. Immunol. 54: 86. 3. BAXLEY, G., G. B. BISHOP, A. C. COOPER, and H. H. WORTIS. 1973. Rosetting of human red blood cells to thymocytes and thymus-derived cells. Clin. Exp. Immunol. 15: 385. 4. SANDILANDS, G. P., K. G. GRAY, A. E. COONEY,j. D. BROWNING, andJ. R. ANDERSON. 1974. Autorosette formation by human thymocytes and lymphocytes. Lancet i: 27. 5. KAPLAN, J. 1975. Human T lymphocytes form rosettes with autologous and allogeneic human red blood cells. Clin. Immuno!. Immunopatho!. 3: 471. 6. GLUCHMAN, J. C, and P. M. MONT.AM:BAULT. 1975. Spontaneous autorosette forming cells in man. A marker of subset population of T lymphocytes? Clin. Exp. Immuno!. 22: 302.

Role of Autorosette Forming Cells in Humoral Immune Response' 9 7. CHARREIRE, J., and J. F. BACH. 1982. Studies on mouse auto-reactive cells. II. Cytotoxic­ ity exerted by mouse lymphocytes against syngeneic erythrocytes. Scand. J. Immunol. 16: I.

8. KUCHRoo, V. K., and P. A. JENNINGS. 1982. Rosetting of bovine T lymphocytes with autologous and allogeneic rcd blood cells. Vet. Immuno!. Immunopathol. 3: 461. 9. SIA, D. Y., and C. R. PARISH. 1980. Anti-self receptors. II. Demonstration of H-2L region restricted receptors on subpopulations of peripheral T and B lymphocytes. J. Immuno!. 124: 2366. 10. NABARRA, B., and J. CHAR.REIRE. 1979. Morphological study of autologous rosettes (A­ RFC). Scand. J. Immuno!. 9: 15. 11. SANDILANDS, G. P., K. GRAY, A. COONEY, J. D. BROWNING, and]. R. ANDERSON. 1975. Formation of auto-rosette by peripheral blood lymphocytes. Clin. Exp. Immuno!. 22: 493. 12. RUCHETON, M., J. CARAUX, A. REY, C. THIERRY, C. ESTEVE, H. VALLES, J. DUFER, A. DESPLACES, D. ZAGURY, and B. SERROU. 1981. Human autologous rosette. II. Further characterization: Markers and functions. Cell. Immuno!. 64: 312. 13. SMOLEN, ]. S., S. O. SHARROW, and A. D. STEINBERG. 1981. Characterization of autologous rosette-forming cells: A nonrestricted phenomenon. J. Immuno!. 127: 737. 14. KUMAGAI, S., T. SUGINOSHITA, and I. GREEN. 1982. Autologous rosette-fanning T cells (ARFf) regulate response of T cells. Hum. Immuno!. 5: 35. 15. SUGAWARA, I., and R. PALACIOS. 1982. Interleukin-2 and serum thymic factor enable autologous rosette-fonning T lymphocytes to generate helper and cytotoxic functions. Scand. J. Immunol. 15: 233. 16. WEKSLER, M. E., C. E. MOODY, R. F. OSTRY, and B. A. CASAZZA. 1980. Lymphocyte transfonnation induced by autologous cells. X. Soluble factors that generate cytotoxic T lymphocytes. j. Exp. M,d. 152: 2845. 17. LAMBERMONT, M., J. WYBRAN, and A. GOVAR£TS. 1977. Human autologous and allogeneic rosettes. J. Immunol. Meth. 15: 157. 18. SCHEFFEL, J. W., and S. J. SWARTZ. 1982. AMLR-reactive T cells isolated by autologous rosette formation. Clin. Immunol. lmmunopathol. 24: 93. 19. DAMLE, N. K., J. A. HANSEN, R. A. GOOD, and S. GUPTA. 1981. Monoclonal antibody analysis of human T lymphocyte subpopulations exhibiting autologous mixed lympho­ cyte reaction. Proc. Nat. Acad. Sci. U.S.A. 78: 5096. 20. KOZAK, R. W., C. E. MOODY, L. STAIANO-CAICO, and M. E. WEKSLER. 1982. Lympho­ cyte transformation induced by autologous cells. XII. Quantitative and qualitative differences between human autologous and allogeneic reactive T lymphocytes. J. Immuno!. 128: 1723. 21. VANWAUKE,J., and J. GOOSENS. 1981. Monoclonal anti-human T lymphocyte antibodies. Enumeration and characterization of T cell subsets. Immunology 42: 157. 22. WEKSLER, M. E., and R. W. KOZAK. 1977. Lymphocyte transformation induced by autologous cells. V. Generation of immunologic memory and specificity during autolo­ gous mixed lymphocyte reaction. J. Exp. Med. 146: 1833. 23. UCHIDA, A., and M. MICKSCHE. 1982. Autologous mixed lymphocyte reaction in the peripheral blood and pleural effusions of cancer patients. J. Clin. Invest. 70: 98. 24. HAN, T., M. L. BLOOM, B. DADEY, G. BENNETf,J. MINOWADA, A. A. SANDBERG, and H. OZER. 1982. Lack of autologous mixed lymphocyte reaction in patients with chronic lymphocytic leukemia: Evidence for autoreactive T-cell dysfunction not correlated with phenotype, karyotype, or clinical status. Blood 60: 1075. 25. GUP'fA, S. 1982. Autologous mixed lymphocyte reaction in man. VIII. Deficiency of autologous mixed lymphocyte reaction in patient with selective 19A deficiency. Immunol. Lett. 5: 101. 26. GUNCH£R, L. H., A. D. STEINBERG, S. B. HOUSE, and I. GREEN. 1980. The autologous mixed lymphocyte reaction in strains of mice with autoimmune disease. J. Immuno!. 125: 1832.

10

N. KHANSARI et al.

27. HOM., j. T., and N. TALAL. 1982. Decreased syngeneic mixed lymphocyte response in autoimmune susceptible mice. Scand. j. Immunol. 15: 195. 28. SMITH, j. B., and R. D. PASTERNAK. 1978. Syngeneic mixed lymphocyte reaction mice: Strain distribution, kinetics, participating cells and absence in NZB mice. J. Immuno!. 121: 1889. 29. GELFAND, E. W., J. J. Lt.:t.:, and H. M. DOSCH. 1979. Selective toxicity of purine deoxynucleoside for human lymphocyte growth and function. Proc. Nat. Acad. Sci. U.S.A. 76: 1998. 30. COWAN, M. J., M. FRAGA, J. ANDREW, N. LAMERIS-MART1N, and A. J. AMMANN. 1982. Purine salvage pathway enzyme activities in human T-, B- and null lymphocyte popula­ tions. Cell. Immuno!. 67: 121. 31. GRAYSON, J., N. J. DOOLEY, 1. R. KOSKI, and R. M. BLAESE. 1981. Immunoglobulin production induced in vitro by glucocorticoid hormones.]. Clin. Invest. 68: 1539. 32. KHANSARI, N., H. H. FUDENBERG, and E. MERLEK. 1983. In vitro humoral immune response of human peripheral blood lymphocytes to tetanus toxoid Sepharose 4B. Immunobiology 164: 42. 33. FOURNIER, C, andJ CIIARREIRE. 1981. Autologous mixed lymphocyte reaction in man. 1. Relationship with age and sex. CeiL Immunol. 60: 212. 34. VAN LAAR HOVEN,]. P. R. M., G. T. SPIERENBURG, and C. H. M. M. DE BRUYN. 1980. Enzymes of purine nucleoside metabolism in human lymphocytes. J. Immunol. Meth. 39: 47. 35. OPELZ, G., M. KruCHI, M. TAK~SUGI, and P. I. TF,RASAKI. 1975. Autologous stimulation of human lymphocyte subpopulat'ions. J. Exp. Med. 142: 1327. 36. KUNTZ, M. M., J. B. INNES', and M. E. WEKSLER. 1976. Lymphocyte transformation induced by autologous cells. j.: Exp. Med. 143: 1042. 37. ILfELD, D. N., R. S. KRAKAUER, and R. M. BLAESE. 1977. Suppression of the human autologous mixed lymphocyte reaction by physiologic concentration of hydrocortisone. J. Immunol. 119: 428. 38. HEU.MAN, D., and J. STOBO. 1982. The autologous mixed lymphocyte reaction. Arth. Rheum. 25: 121. 39. BARTON, R., F. MARTINIUK, R. HIRSCHHORN, and T. GOLDSCHNEIDER. 1979. Distribu­ tion of adenosine deaminase among lymphocyte populations in the rat. J. Immunol. 122: 216. 40. MA, D. D. P., T. A. SYLWE~IROWICZ, S. GRANGER, M. MASSA lA, R. FRANKS, C.]ANOSSY, and V. HOFFBRAND. 1982. Distribution of terminal deoxynucleotidyl transferase and purine degradative and synthetic enzymes in subpopulations of human thymocytes. J. Immunol. 129: 1430. 41. PALACIOS, R., L. LLORENTE, D. ALARCON-SEGOVIA, A. RUIZ-ARGUELLE<;, and E. DIAZ­ jOUANEN. 1980. Autologous rosette forming T cells as responding cells in autologous mixed lymphocyte reaction. J. Clin. Invest. 65: 1527. 42. UCHIDA, A. and M. MICKSCEIE. 1981. Reactivity to autologous non-T cells and suppres~ sor function of human rosettes. Immuno!. Lett. 3: 145. 43. MICKLEM, H., C. ASH, N. STAINES and N. ANDERSON. 1970. Quantitative study of cells reacting to skin allografts. Nature 227: 947. 44. GATENBY, P. A., G. S. KANSAS, C. Y. XIAN, R. L. EVANS and E. G. ENGLEMAN. 1982. Dissection of immunoregulatory subpopulations of T lymphocytes within the helper and suppressor siblineage in man. J. Immuno!. 129: 1997. 45. SAKANE, T. and I. GREEN. 1977. Human suppressor T cells induced by'~oncanavalin A: Suppressor T cells belong to distinctive T cell subclasses. J. Immunol. 119: 1169. 46. SA KANE, T., M. HONDA, Y. TANIGUCHI and H. KOTANI. 1981. Separation of concanava­ lin A-induced human suppressor and helper T cells by the autologous erythrocytes rosette technique. J. Clin. Invest. 68: 447. 47. PALACIOS, R. 1981. Role of autologous rosette-fonning T cells in the concanavalin A­ induced suppressor cell function. Cell. Immunol. 61: 273.

Role of Autorosette Forming Cells in Humoral Immune Response

1t

48. PALACIOS, R., D. ALARCON-SEGOVIA, L. LLORENTE, A. RUIZ-ARGUF.U.ES and E. DIAZ­ jOUANEN. t981. Human post-thymic precursor cells in health and disease. I. Characteri7.ation of the autologous rosette forming T cells as post-thymic precursors. Immunology 42: 127. 49. KHANSAIU, N., G. SPRINGtR, E. MERLtR and H. H. FUOENBERG. 1983. Mechanisms for the removal of senescent human erythrocytes from circulation. Mech. Age. Develop., 21: 49. Dr. N. KHANSARI, Department of Basic and Clinical Immunology and Microbiology, Medical University of South Carolina, 171 Ashley Ave., Charleston, South Carolina 29425,

U.s.A .