Kinetics of cytokine production by thymocytes during murine AIDS caused by LP-BM5 retrovirus infection

Immunology Letters, 41 ( 1995) 187- 192 0165-2478/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved IMLET 2432

Kinetics of cytokine production by thymocytes during murine AIDS caused by LP-BM5 retrovirus infection James Wang aqb,c,‘,Dennis S. Huang

a,c*d*2,Steve Wood

Ronald

R. Watson

a*b,c,3,Paul T. Giger

‘XX and

a,c,*

” Department of Family and Community Medicine, b Nutritional Sciences Program. ‘Alcohol Research Center, ’ Department of Microbiology and Immunology, ’ Department of Molecular and Cellular Biology, University of Arizona, Tucson, AZ 85724, USA (Received 29 April 1993; revised 21 June 1993; accepted Key words: Murine AIDS; Thymus; Cytokine;

1. Summary C57BL \ 6 mice inoculated with the murine leukemia retrovirus mixture, LP-BMS, rapidly produce murine AIDS with many functional similarities to human AIDS. Human HIV infection has recently been shown to inhibit thymocyte maturation. Therefore, the kinetics of the proliferation of thymocytes induced by Concanavalin A (ConA) and levels of cytokines produced by in vitro ConA-stimulated thymocytes were examined during the progression of murine AIDS. The proliferation of thymocytes induced by ConA was significantly enhanced by retrovirus infection at 4 weeks post-infection compared to control, but significantly inhibited during 8-12 weeks post-infection. Release of IL-2 by ConA-stimulated thymocytes was significantly increased by retrovirus infection during 2-5 weeks postinfection and 1 l- 18 weeks post-infection compared to control, but significantly decreased during 7-9 weeks post-infection. Secretion of IL-4 by ConA-stimulated thymocytes was significantly enhanced by retrovirus

’ Current address: Hemrepin MN, USA.

County

Medical

2 Current address: Department of Pathology, University, Cleveland OH, USA. ’ Current address: Ross Columbus, OH, USA.

Products

Division,

Center,

Minneapolis,

Case Western Reserve

Abbott

Laboratories,

* Corresponding author: Ronald R. Watson, Ph.D., Department of Family and Community Medicine, University of Arizona, Tucson, AZ 85724, USA. Tel.: (602) 626-6001; Fax: (602) 626-2030. SSDI 0165-2478(95)00095-X

9 June 1993)

T-cell differentiation

infection from 5 to 18 weeks post-infection compared to control. The level of IL-6 produced by ConA-stimulated thymocytes was significantly inhibited by retrovirus infection at the beginning of retrovitus infection (2-9 weeks), but significantly elevated after 1 I weeks postinfection compared to control. Release of IFN, by ConA-stimulated thymocytes, however, was significantly enhanced during the whole period of retrovirus infection compared to control, while it surged at 13 weeks post-infection. We conclude that retrovirus infection affects the thymus, producing altered T-cell differentiation via the dysregulation of thymocyte cytokine secretion.

2. Introduction LP-BM5 murine leukemia retrovirus (MuLV) infection induces murine acquired immune deficiency syndrome (AIDS) characterized by lymphadenopathy, splenomegaly, hypergammaglobulinemia, deficient Bcell responses to T-independent antigens or depressed allogenic cytotoxic T-cell response [ 1I. It produces immune dysfunctions via dysregulation of cytokine production [2,3] and possesses many functional similarities to human AIDS, including changes in T-cell number and functions [4]. LP-BM5 MuLV infection in mice is a potentially useful model to study the consequence of human immunodeficiency virus (HIV) infection in humans [5-71. Although most studies have been focused on immune changes in spleen and lymph nodes, little is known about the role of thymus in murine AIDS. The generation of functional competent T cells from precursor populations within the thymus involves sev187

era1 stages of cellular proliferation and differentiation [8,9]. It has been proposed that the control of these processes is mediated by cytokines, hormones, or by interactions with specific elements of the thymic stroma [lo]. Consequently, the alternation in T-cell subsets or T-cell functions observed after retrovirus infection in human peripheral blood cells or mouse spleen cells could be the consequence of changes at the thymus level, including thymocyte cytokine production. Histopathological studies of thymic glands from HIV-l-infected children revealed involuted architecture, reduced size and weight, and fewer Hassal’s corpuscles [11,12]. HIV-l causes changes in thymic hormone and cytokine levels such as thymosin 1 (Y and interleukin-6 (IL), which act on T-cell differentiation [13,14]. Human fetal thymocyte cell lines were infected with HIV-l in vitro and altered expression of T-cell differentiation markers [15]. However, no detailed information about cytokine production by thymocytes in the progression of AIDS was reported. We therefore examined the changes of cytokine production (IL-2, IL-4, IL-6 and Interferon-gamma (IFN)) by in vitro ConA-stimulated thymocytes to understand how retrovirus infection alters thymic cytokine secretion and thereby adversely affects thymocyte development and function.

3. Materials and methods 3.1. Animal and LP-BMS MulV infection Female C57BL\6 mice 3 to 5 weeks of age were obtained from Charles River Laboratories (Wilmington, DE). The mice were housed in transparent plastic cages with stainless steel wire lids. After 3 weeks of housing in the animal facility in Arizona Health Sciences Center, mice were then randomly assigned to the following two groups: uninfected controls and LP-BMS retrovirus infected. The retrovirus was administered intraperitoneally to mice with an ecotropic (XC> of 4.5 log,, PFU/ml, which induces disease with a time course comparable to that previously published [1,4]. Infection of adult female C57BL/6 mice with LP-BM5 MuLV leads to the rapid induction of clinical symptoms with virtually no latent phase [ 1,4]. 3.2. Cytokine standards and antisera Rat anti-murine IL-2, IFN, (interferon-gamma) and IL-6 (interleukin) monoclonal antibodies and standard recombinant murine IL-2, IFN, and IL-6 were obtained from Genzyme (Boston, MA). Rabbit anti-murine IL-2 polyclonal antibody was obtained from Collaborative (Bedford, MA). Rat anti-murine IL-4 and biotin-rat 188

anti-murine IL-4 monoclonal antibodies, and recombinant murine IL-4 standard were obtained from PharMingen (San Diego, CA). Goat anti-murine IL-6 polyclonal antibody was obtained from R&D System (Minneapolis, MN). Rabbit anti-murine IFN, serum was prepared by our laboratory. 3.3. Preparation of cytokine production 0.1 ml/well of thymocytes (1 X lO’/ml) from a pool of 4 mice of the same experiment group were cultured in 6 culture wells on 96-well tissue culture plates (Falcon, Lincoln Park, NJ) with CM. The thymocytes were stimulated with ConA (10 pg/ml, O.lml/well, Sigma, St. Louis, MO, diluted in CM) for induction of IL-2, IL-4 and IL-6 with a 24 h incubation, and IFN, by a 72 h incubation at 37”C, 5% CO, incubator. After the incubation time indicated, the plates were centrifuged for 10 min at 800 g. Supematant fluids were collected and stored at - 70°C until assayed by ELISA assay. No loss of activity has been noted during such storage. 3.4. ELISA assay The methods were described previously [ 161. In brief, the wells of 96-well microtiter plates (Immulon II, Dynatech, Chantilly, VA) were coated overnight at 4°C with 50 ~1 of anti-cytokine capture rat monoclonal antibodies specific for the measured cytokines, diluted to 1-4 pg/ml in 0.05 M bicarbonate buffer (pH 9.6). Then 50 ~1 of standard or sample cytokines (diluted in culture medium) were added for 2 h at 37°C. Then 50 ~1 of polyclonal and biotinylated monoclonal detecting antibodies diluted in PBS (1-4 pg/ml) were added into each respective plate for another hour incubation at 37°C. Then 50 ~1 of diluted strepavidin-HRP (1 : 5000, Jackson, West Grove, PA), goat anti-rabbit IgG-HRP (1 : 5000, American, Qualex, La Mirada, CA) conjugates, and donkey anti-goat IgG-HRP (1 : 5000, Jackson, West Grove, PA) were added to each respective well. The plates were then incubated as above for another hour and washed 4 times with PBS-Tween and once with PBS. Finally, 100 ~1 of substrate buffer of ABTS (Sigma, citrate buffer, 0.1 M, pH 4.2 containing 0.03% H,O,) was added to each well, and the plates were allowed to develop for 20-30 min at room temperature. Optical density was determined at 405 nm by Titertek Multiscan (Flow Lab, MacClean, VA). Sensitivity of the ELISAs for IL-2, IL-4, IL-6 and IFN, were determined to be 156, 156, 320, and 78 pg/ml, respectively. No cross-reactions between these measured cytokines in the ELISA kits were observed (data not shown).

3.5. Mitogenesis

-*- Control

of thymocytes

Thymocytes (1 X 107/ml) in 0.1 ml were cultured in 96-well flat-bottom cultured plates (Falcon) with ConA (5 pg/ml) in CM at 37”C, 5% CO, for 20 h for ConA-induced T-cell proliferation. Then they were pulsed with [3H]-thymidine (0.5 pCi/well, New England Nuclear, Boston, MA). After 4 h they were harvested by a cell sample harvester (Cambridge Technology, Cambridge, MA). Radioactivity was determined by a liquid scintillation counter (Tri-Carb, 22OOCA, Packand, Laguna Hills, CA). Data are presented as counts per minute (cpm>.

1.00

1

3.6. Statistics

0.50

Two-tailed Student’s t test was applied. P < 0.05 was considered significant difference between two groups.

0.40

1

induced by ConA

Retrovirus

_ +

infection

0.20

-

0.10

’

.

*

*

*

*

.

.

*

0

2

5

7

9

11

13

15

18

(Weeks)

4.3. Release of IL-4 As shown in Fig. lB, secretion of IL-4 by thymocytes from retrovirus-infected mice was significantly higher than that by thymocytes from uninfected mice

MURINE AIDS a

(weeks)

4

8

12

7.66 f 1.06 9.62 f 0.96 ’

10.63 f 2.05 7.95 f 0.32

11.63k2.59 5.04i- 1.48

a The values are meanf SD for 6 individual culture thymocytes/weIl with ConA (5 pg/ml) stimulation. * P < 0.05.

20

retrovirus infection after 11 weeks until 18 weeks postinfection (P < O.OS>, while that of thymocytes from uninfected mice was gradually decreased as age increased.

BY ConA DURING

CPM (IO31 at post-infection

-

Fig. 1. Kinetics of release of II-2 (A) and IL-4 (B) by in vitro ConA-stimulated thymocytes during murine AIDS. The values are meanf SD for 6 individual culture wells from a pool of 4 mouse thymuses. Secretion of IL-2 and IL-4 was evaluated by 1 X lo6 thymocytes/well in 0.2 ml volume with ConA (5 @g/ml) stimulation.

As shown in Fig. lA, the release of IL-2 produced by in vitro ConA-stimulated thymocytes was significantly higher during the initial weeks of retrovirus infection compared with uninfected control (P < 0.05). Thereafter in vitro production of IL-2 was significantly reduced by retrovirus infection (P < 0.05). Furthermore, level of IL-2 was significantly elevated again by

INDUCED

B: IL-4

Post-infection

4.2. Release of IL-2

PROLIFERATION

0.30

0.00

As shown in Table 1, in vitro thymocyte proliferation was significantly increased at 4 weeks post-retrovirus infection compared with uninfected control (P < 0.05). Thereafter the proliferation was significantly inhibited by retrovirus infection at 8 and 12 weeks postinfection (P < 0.05). However, the proliferation exhibited no significant change by retrovirus infection at 16 weeks post-infection (P < 0.05).

TABLE 1 KINETICS OF THYMOCYTE

A: IL-2

t

4. Results 4.1. In vitro thymocyte proliferation

-O- LP-BM5

wells

from

l

a pool

of 4 mouse

thymi.

16

l

The proliferation

9.44 f 2.83 11.23+ 1.48 was evaluated

by 1 X lo6

189

-*-

Control

-O- LP-BMS

compared with uninfected controls through out the experiment beginning 2 weeks post infection (P < 0.051, and then secretion of IFN,, surged at 13 weeks post-infection.

2

A: IL-6

f‘-

1

t

s

51

5. Discussion

(D i

0 0

2

5

7

9

11

13

2

5

7

9

11

13

15

10

20

OL

0

Post-Infection

15

10

20

(Weeks)

Fig. 2. Kinetics of release of IL-6 (A) and IFN-T (B) by in vitro ConA-stimulated thymocytes during murine AIDS. The values are meanf SD for 6 individual culture wells from a pool of 4 mouse thymuses. Secretion of IL-4 and IFN-, was evaluated by 1 X lo6 thymocytes/well in 0.2 ml volume with ConA (5 pg/ml) stimulation.

from 5 to 18 weeks post-infection (P < 0.05). There was a significant decline in IL-4 production by thymocytes from uninfected mice as age increased. 4.4. Release of IL-6 As shown in Fig. 2A, IL-6 secreted by thymocytes was significantly inhibited by retrovirus infection compared with uninfected controls up to 11 weeks post-infection (P < 0.051, while IL-6 production by thymocytes from uninfected mice had significantly declined as age increased (P < 0.05). However, the release of IL-6 was significantly enhanced by retrovirus infection after 11 weeks post-infection compared to continuing reduced production of IL-6 by controls (P < 0.05). 4.5. Release of IFN, As shown in Fig. 2B, IFNY production by thymocytes was significantly elevated by retrovirus infection 190

We demonstrated that murine retrovirus infection modified production of cytokines by in vitro ConAstimulated thymocytes. This expands understanding of retrovirus-induced changes in cytokines which can be partially restored by supplemental vitamin E [37,39], overcoming a vitamin E deficiency induced by the retrovirus infection [38]. The mammalian thymocytes can be subdivided into 4 major subpopulations, based on expression of the differentiation-related antigens CD4 and CD8. ‘Double-negative’ (DN) CD4- CD8- thymocytes (2-4% of total in the normal adult mouse) represent immature cells with the capacity to give rise to other subsets in irradiated host [ 191. CD4+ 8’ cells in the thymic cortex account for the majority (80-85%) of thymocytes and most of such ‘double-positive’ (DP) cells are destined to die in situ [20]. The other two ‘single-positive’ (SP) CD4-8+ and CD4+8subsets (5- 10%) in the thymic medulla represent mature T cells seen in the peripheral tissue [20]. SP subsets could be further divided into CD3+ and CD3- subsets [24]. CD3+ cells resemble mature peripheral T cells in that they are functionally immunocompetent. They include responding to a variety of mitogen stimuli and secreting cytokines, and originates from CD3-4- 8+ and CD3- 4+8- immature T cells in the thymus [ 19,201. Our previous report found that murine retrovirus infection alters T-cell differentiation by reducing CD4+CD8+ cells. This population (80-85% in normal mice) dramatically dropped to only 30% and other populations, CD4+ 8- and CD4- 8+ (SP subsets), were elevated respectively from controls 1.99% and 7.67% to 33.98% and 15.5% of thymocytes in retrovirus-infected mice at 16 weeks post-infection [17]. The continuing analysis of CD3 marker of two SP subsets by fluorescence cytometry in our laboratory indicated that the percentage of CD3+CD4+CD8subsets in the thymus from retrovirus-infected mice for 16 weeks dramatically dropped from control 74% to 7.9% of CD4+CD8subset, while the percentage of CD3+CD4_CDS’ thymocytes of CD4_CD8+ subsets did not change during murine AIDS (2.3% in murine AIDS vs. 2.2% in control mice [37,39]. Therefore SP thymocytes increased by the retrovirus infection were immature CD3- SP T cells, not mature CD3+ SP T cells, which suggests disturbance of T-cell maturation in the thymus in pro-

gression of AIDS. Since we did not study the kinetics of thymocyte subsets during the progression of murine AIDS, we speculated that the increased or decreased thymocyte proliferation and thymic cytokine release induced by retrovirus infection might partly reflect the fluctuation of populations of CD3+ mature T cells in the medulla and CD3 - immature T cells in the cortex of thymus [37,39]. Also, the functional change of mature thymocytes may be reponsible for the change of thymic cytokine secretion we observed. Taken together, retrovirus infection affects the thymus, producing altered T-cell differentiation via the dysregulation of thymic cytokine secretion [37,39]. Such changes in thymocyte differentiation may explain the dysfunction of T cells seen in AIDS. How could thymocyte differentiation be altered during the progression of murine AIDS? One possibility is that activated peripheral T cells by retrovirus infection re-enter the adult thymus, where they secret some cytokines and adversely affect regulation or differentiation of T maturation in the thymus. Agus et al. found that homing of mature T cells to the thymus of adult mice was substantial and exclusively restricted to ConAactivated T cells. However, mature resting T cells have virtually no capacity to migrate to the thymus [21]. The physiological significance of ‘back migration’ of activated T cells into the thymus may be that the presentation of retrovirus antigens in the thymus triggers selftolerance induction, causing the loss of immune response to retrovirus antigen. Alternatively, the release of cytokines by retrovirus-activated mature T cells may impair T-cell differentiation, causing secondary acquired T-cell deficiency. This notion that circulating T cells activated by retrovirus migrate into the adult thymus is in accord with the some reports [22-251. This speculation is also supported by the recent observation in our laboratory that mature mesenteric lymph node lymphocytes from mice infected with retrovirus LPBM5 migrate back to the thymus in normal mice by adoptive transfer assay (Lopez et al., unpublished data). Clearly mesenteric lymph node cells are distinct functionally and phenotypically from splenic or thymic cells during murine rctrovirus infection [41,42]. Production and action of cytokines in the human and mouse thymus have been associated with an essential role in T-cell development [lo]. In mice, expressing a human IL-2Ra chain transgene, thymocytes express a non-functional murine IL-2RP human IL-2R CY heterodimer resulting in the accumulation of T-cell precursors in the thymus and periphery [26]. Also, anti-IL-2R LY chain antibodies abrogate T-cell development which can be reversed by addition of IL-2 [27]. Addition of IL-2 to intact lobes immersed in culture medium promotes the selective outgrowth of T-cell expression T6 TCR [28].

Consequently, the dysregulation of IL-2 released by retrovirus infection we observed may be responsible for the changes in T-cell subpopulations and differentiation in the thymus during the progression to murine AIDS. The addition of IL-4 blocks T-cell development by reducing the number of DP thymocytes [29]. The constitutive production of IL-4 in IL-4 transgenic mice also results in the inhibition of DP thymocytes and mature peripheral T-cell development [30.3 I]. Consequently, we conclude that the consistently increased levels of IL-4 induced by retrovirus infection in the thymus may contribute to the impairment of T-cell development in murine or human AIDS, resulting in secondary acquired T-cell deficiency. Treatment with anti-IL-4 antibodies or IFN, prevented immune dysfunction including the normal increase in production of cytokines by Th2 spleen cells, and decreased production by Thl cells during murine AIDS [40]. IL-6 has been shown to promote the differentiation of Thy- 1 + IL-2R ’ donor thymocytes after intrathymic transfer into irradiated hosts [32]. Thus, the dysregulation of IL-6 secretion induced by retrovirus infection at the beginning of infection may also contribute to the abrogation of T-cell development in the thymus. The physiological significance of increased levels of IFNy by thymocytes, induced by retrovirus infection, may be involved in the up-regulation of MHC class I and II expression on the surface of thymic stromal cells, which play a crucial role in positive and negative selection during T-cell education (i.e., recognization of ‘self’ and ‘non-self’) in the thymus. Thus, the dysregulation of IFN, release by thymocytes may reflect the failure of T-cell education, leading to loss of tolerance to self-antigens or reaction to non-self antigens. The source of IFN, in vivo has been suggested to be mature CD4+ T cells and natural killer (NK) cells in the peripheral tissues [33]. The decreased number of CD3+CD4+CD8T subset or no changed CD3+CD4_CD8+ T subset in the thymus during the progression of murine AIDS cannot account

TABLE 2 THYMIC NK CELL ACTIVITY WEEK POST-INFECTION ’

DURING

Renovirus

(o/o) at ratio of E/T

_ +

infected

Cytotoxicity

MURINE

I

100: 1

50:

19.Ok2.2 40.5 *5.2

14.6f 1.36 28.4 f 4.8 ’

*

AIDS AS 12

25:

I

10.9*1.7 19.1 *4.9

l

a The measurement of NK cell activity was finished by fluorescence release assay [34]. The values are meanf SD for triplicate samples from a pool of 4 mouse thymi. The thymocytes were cultured in U-bottom microtiter plates with 0.2 ml volume at different ratio of effecter/target (YAC-1) cell concentration. YAC-I cell concentration was adjusted at 1 X 105/well. P < 0.05. l

191

for consistently increased secretion of IFN, by thymocytes. Thus, we measured thymic NK cell activity by fluorescence release assay [30]. The results showed that NK cell activity was increased after 12 weeks retrovirus infection (Table 2). Since NK cells are present at a very low frequency ( < 0.1% of thymocytes) in both human and mouse thymus [35,36], the increased activity of NK cells may suggest that the impairment of T-cell differentiation and changes in T-cell subpopulations in the thymus during the progression of murine AIDS could be caused by back migration of other peripheral lymphoid cells (e.g., NK cells) as well as mature peripheral T cells. The continuing elucidation of the physiological roles of cytokines in the T-cell maturation and the mechanism of these changes induced by retrovirus infection may facilitate understanding of the immunopathogenic mechanism of HIV infection on the thymus in human AIDS.

Acknowiedgements

Supported by Grant NIH AA 08037.

References Ill Chattopadhyay,

S.K., Makino, M., Hartley, J.W. and Morcse, III H.C. (1988) in Immunodeficient Animals in Experimental Medicine B.-A. Wu and J. Zheng, eds.), Karger, Basel, Switzerland, pp. 12-18. 121Gaxzinelli, R.T., Makino, M., Chattopadhway, S.K., Snapper, CM., Sher, A., Hugin, A.W. and Morese, H.C., III. (1992) J. Immunol. 148, 182. [31 Wang, Y., Huang, D.S., Giger, P.T. and Watson, R.R. (1993) Adv. Bio. Sci. 86, 335. [41 Watson, R.R. (1989) Life Sci. 44, 1. El Mosier, D.E., Yetter, R.A. and Morse, H.C. III, (1987) J. Exp. Med. 168, 1737. 161Yetter, R.A., Buller, R.M.L., Lee, J.S., Elkins, K.L., Mosier, D.E., Fredrickson, T.N. and Morse, H.C. III (1988) J. Exp. Med. 168, 623. [71 Cemy, A., Hugin, A.W., Hardy, R.R., Hayakawa, K., Zinkernagel, R.M., Makino, M. and Morse, H.C. III (1990) J. Exp. Med. 171, 315. [81 Scollay, R., Wilson, A., D’Amico, A., Kelly, K., Egreton, M., Pearse, M., Wu, L. and Shortamn, K., (1988) Immunol. Rev. 104. 81. [91 Lesley, J., Trotter, J., Schutle, and Hyman, R., (1990) Immunology 128, 63. [lOI Carding, S.R., Hayday, A.C. and Bottomly, K. (1991) Immunol. Today 12, 239. 1111 Joshi, V.V. and Oleske, J.M. (1985) Arch. Pathol. Lab. Med. 109, 142. [I21 Schuurman, H.J., Krone, W.J.A., Broekhuizen, R., Van Baarien, J., Van Veen, P., Goldstein, A.L., Huber, J. and Goudsmit J. (1991) Am. J. Pathol. 134, 1329.

192

[13] Schnittman, SM., Singer, K.H., Greenhouse, J.J., Stanley, S.K., Whichard, L.P., Haynes, B.F. and Fauci, A.S. (1991) J. Immunol. 147, 2553. [14] Schnittman, S.M., Singer, K.H., Greenhouse, J.J., Stanley, SK. Whichard, L.P., Haynes, B.F. and Fauci, A.S. (1991) VII lnternational Conference on AIDS. Abstract Book, Vol. 2, Florence, Italy, p. 153. [1.5] Tanaka, K.E., Hatch, W.C., Kress, Y., Soeiro, R., Calvelli, T., Rashbaum. W.K., Rubinstein, A. and Lyman, W.D. (1992) J. AIDS 594. [16] Wang, Y., Huang, D.S., Giger, P.T. and Watson, R.R. (1993) Alcoholism: Clin. Exp. Res., 17, 1035. [17] Lopez, M.C., Colombo, L.L., Huang, D.S., Wang, Y. and Watson, R.R. (1992) Clin. Immunol. Immonopatbol. 65, 45. [18] Fowlkes, B.J., Edison, L., Mathieson, B.J. and Chused, T.M. (1985) J. Exp. Med. 162, 802. [19] MacDonald, H.R., Budd, R.C. and Howe, R.C., (1988) Eur. J. Immunol. 18, 519. [20] Ceredig, R., Glasebrook, A.L. and MacDonald, H.R. (1982) J. Exp. Med. 155, 358. [21] Agus, D.B., Surh, C.D. and Sprent, J. (1991) J. Exp. Med. 173, 1039. [22] Lo, D., Burkly, L.C., Widera, G., Cowing, C., Flavell, R.A., Palmiter, R.D. and Brinster, R.L. (1988) Cell 53, 159. [23] Naparstek. Y., Ben-Nun, A., Holoshitz, J., Reshef, T., Frenkel, A., Rosenberg, M. and Cohen, I.R. (1982) Nature (Land.) 300, 262. [24] Naparstek, Y., Ben-Nun, A., Holoshitz, J., Reshef, T., Frenkel, A., Rosenberg, M. and Cohen, I.R. (1983) Eur. J. Immunol. 13, 418. [25] Fink, P.J., Bevan, M.J. and Weissman, I.L. (1984) J. Exp. Med. 159, 436. [26] Gutierrez-Ramos, J.C., Martinez-A.C., Kohler, G. et al. (1990) Immunol. Res. 140, 661. [U] Zuniga-Pflucker, J.C. and Kruisbeek, A.M. (1989) J. Immunol. 144, 3736. [28] Ceredig, R., Medveczky, J. and Skulimowski, A. (1989) J. Immunol. [29] Plum, J., De Smedt, M., Leclercq, G. et al. (1990) J. Immunol. 145, 1066. [30] Tepper, R.I., Levinson, D.A., Stanger, B.Z. et al. (1990) Cell 62, 457. 1311 Lewis, D.D., Yu, C.C., Forbush, K.A. et al. (1991) J. Exp. Med. 173, 89. [32] Nakano, N., Kikutani, H. and Kishimoto, T. (1990) Dev. Immunol. I, 77. [33] Scott, P. and Kaufman, S.H.E. (1991) Immunol. Today 142, 3353. [34] Poet, T.S., Pillai, R., Wood, S. and Watson, R.R. (1991) Toxicol. Len. 59, 1. [35] Mingari, M.C., Poggi, A. et al. (1991) J. Exp. Med. 174, 21. [36] Garni-Wagner, B.A., Wine, I.L., Tutt, M.M. et al. (1990) J. Immunol. 144, 796. [37] Wang, Y., Huang, D.S. and Watson, R.R. (1993) Immunol. Res. 12, 358. [38] Wang, Y., Liang, B. and Watson, R.R. (1994) Nun‘. Res. 14, 1031. [39] Wang, Y. and Watson, R.R. (1994) Thymus 22, 153. [40] Wang, Y., Ardestani, S.K., Liang, B., Beckham, C. and Watson, R.R. (1994) Immunology, 83, 384. [41] Lopez, M.C., Huang, D.S., Way, D.L. and Watson, R.R. (1995) Adv. Exp. Med; pp. 103991042. [42] Huang, D.S., Wang, Y., Marchalonis, J.J. and Watson, R.R. (1994) Reg. Immunol. 5, 325.