Arch~e~ of

Original contributions 9 Springer-Verlag 1992 Arch Dermatol Res (1992) 284:189-192

Epidermal and splenic antigen-presenting cell function in a retrovirally induced murine immunodeficiency syndrome (MAIDS)* A. Cerny t, S. Izui 2, J.-H. Saurat 3, F. A. Waldvogel 1, H. C. Morse III 4, and C. Hauser 3 1 Departments of Medicine, 2 Pathology, 3 Dermatology, H6pital Cantonal Universitaire, 24, rue Micheli-du-Crest, CH-1211 Geneva, Switzerland, ~A Laboratory of Immunopathology,National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA Received February 24, 1992

Summary. Since alterations of epidermal Langerhans cells (LC) have been observed in humans infected with HIV, we investigated the morphology and function of these cells in murine acquired immunodeficiency syndrome (MAIDS), a murine model closely resembling human AIDS. The number as well as the shape of dendritic MHC class II + cells from ear skin of C57BL/6 mice were similar in normal and infected animals. In mixed epidermal cell (EC) lymphocyte cultures, EC from infected mice and from normal mice stimulated allogeneic T cell proliferation to the same extent. In contrast to T cells from normal mice, however, T cells from infected mice did not respond to allogeneic spleen cells, confirming the presence of a T-cell defect in MAIDS. Subcutaneous injection of syngeneic mice with trinitrophenyl-modified MAIDS EC resulted in delayed ear swelling responses after challenge that were equivalent to those induced by hapten-modified EC from normal mice, suggesting that the contact sensitivity inducing potential of MAIDS LC was preserved. To investigate antigen presenting and processing function, EC and spleen cells were tested with the ovalbumin-specific IAb-restricted T cell hybridoma BO.17.10 and either ovalbumin 323-339 peptide or intact ovalbumin protein. MAIDS spleen cells had a reduced antigen presenting capacity compared with normal spleen cells, whereas EC from these mice showed the same processing and presenting capacity as normal controls. In summary, our results demonstrate that the frequency, morphology, level of MHC class II antigen expression and ability to process and present antigen is normal for LC from mice with MAIDS whereas the function of splenic T cells and APC from infected mice is significantly impaired.

Correspondence to. C. Hauser * Presented in part at the annual meetingof the European Society for Dermatological Research, Copenhagen, Denmark, 1-4 May 1991 and publishedas an abstract, J InvestDermatol (1991)96:1014

Key words: Langerhans cells - Murine AIDS - Antigen presenting cell

Susceptible mouse strains develop an immunodeficiency disease after infection with the murine leukaemia viruses designated LP-BM5 MuLV [1, 6, 7, 15, 16]. The disease has a natural history in many ways comparable to HIV-l-induced AIDS in humans and has therefore been termed murine acquired immunodeficiency syndrome (MAIDS). This animal model has proved to be useful particularly in studies on the pathogenesis as well as on the efficacy of therapeutic interventions on retrovirusinduced immunodeficiency [2, 5, 9]. Alterations of Langerhans cells (LC) have been reported in humans with AIDS. Belsito et al. described decreased LC density as tested by ATPase and MHC class II reactivity of epidermal sheet preparations [3]. In addition, the presence of p17 and p24 HIV antigens in LC, some with blunted dendrites, has been reported by Tschachler et al. using immunofluorescence techniques [21]. Furthermore, budding of viral particles from human LC has been described as well as the recovery of HIV from monocyte/macrophage cultures incubated with skin biopsy specimen from an AIDS patient [17]. Several groups have investigated the presence of HIV RNA and/or DNA within the human epidermis by the polymerase chain reaction [11, 12, 22]. The percentage of positive signals derived from epidermis or from preparations enriched in LC varied widely. Kalter et al., however, found no evidence of LC harbouring HIV or signs of LC alterations using several techniques [11]. Although several groups have investigated virological and morphological aspects of LC in AIDS, to our knowlede, there are no studies on LC function in HIV infection. Since human AIDS and MAIDS might share retrovirus-induced alterations of epidermal antigen presenting cells (APC), we examined the morphology and function of epidermal LC in MAIDS. Here, we report on the expression of major histocopatibility complex (MHC)

190 class II molecules on epidermal LC. I n addition, we describe A P C function in b o t h the epidermal a n d splenic compartments.

St. Louis, Mo., USA). After 1 day, supernatants were removed and their IL-2 activity determined with CTLL cells as previously described [8].

Results Materials and methods

Immunofluorescence studies Mice and viruses Inbred C57BL/6 (B6; H-2b) and BALB/c (H-2 ~) mice were obtained from IFFA Credo, I'Arbresle, France, and kept under conventional conditions. Mice were injected at the age of 4 weeks i.p. with 0.1 ml LP BM5 MuLV stock containing a recently described replicationdefective component [6] and approximately 101 ffu/ml MCF virus and 103.8- 10s pfu/ml ecotropic MuLV. Studies were done 12 weeks after virus infection, corresponding to an advanced stage of disease with spleen weights in the range of 350 to 600 nag.

Immunofluorescence studies Epidermal sheet preparations [10]were incubated overnight at 37 ~ with undiluted culture supernatant from the M5/114.15.2 hybridoma (anti-mouse Ia, rat IgG2b; ATCC, Rockville, Md., USA) or from the GK 1.5 hybridoma (anti-mouse CD4, rat IgG2b; ATCC) as an isotype control. Washed sheets were then incubated for 3 h at 37 ~ with affinity-isolated goat F(ab')2 anti-mouse immunoglobulins conjugated with FITC (1:20; Tago, Burlingame, Calif., USA). Washed and mounted sheets were scored under a microscope with epiimmunofluorescence at a magnification of 250. The number of dendritically shaped cells in five high-power fields per sample was counted. For flow cytometric analysis, freshly prepared EC were labelled with the same antibody combinations as above except that the second antibody was diluted 1:100. Analysis was done with a FACSCAN (Becton and Dickinson, Mountain View, Calif., USA).

Sensitization by injection of hapten-modified epidermal cells (EC) Ear EC were obtained by trypsinization as previously described [8]. The EC suspension, containing 3% MHC class II + cells, was modified with 1 mM trinitrobenzene sulphonic acid [8] (Eastman, Rochester, N.Y.) and injected subcutaneously into naive syngeneic recipients (7 • 105 EC per animal) as previously described [20]. After 7 days, the animals were challenged with 1% trinitrochlorobenzene (Polysciences, Warrington, Pa., USA) in olive oil. Ear thickness was measured 24 h after challenge employing a dial gauge caliper (Interapid, Rolle, Switzerland) using the contralateral olive-oilpainted ear as control. Results are expressed as delta ear swelling response.

CeH cultures Seven-day mixed EC lymphocyte cultures were set up as previously described [13], with the exception that nylon-wool-passed spleen cells were used as responders. EC or spleen cells from control and MAIDS mice were incubated with chicken ovalbumin (lot 19F 8105; Sigma) or ovalbumin peptide 323-339 (kindly provided by Dr. A. Sette, Cytel, La Jolla, Calif.) and the IAb-restricted ovalburnin specific T-cell hybridoma BO.17.10 (5 • 104 well, kindly provided by Dr. L. Adorni, Sandoz, Basel, Switzerland [19]. Cultures were set up in triplicate in 96-well flat-bottommed tissue culture clusters (Costar, Cambridge, Mass., USA) with a volume of 200 ixl/well. Culture medium was RPMI1640 containing 10% FCS, 100 U/ml penicillin, 100 Ixg/ml streptomycin, 2 IxM glutamine, 5 x 10- s M 2-mercaptoethanol (all from Gibco, Basel, Switzerland), and 1 ~tg/ml indomethacin (Sigma.

Neither the shape n o r the n u m b e r o f M H C class II + E C in the sheet p r e p a r a t i o n s differed between control B6 (43.2 _+ 4.0 positive cells/high-power field, n = 6 mice) a n d M A I D S mice (38.7 +_ 4.7 positive cells/high-power field, n = 6 mice, p > 0.05). F l o w cytometric analysis confirmed that the n u m b e r of M H C class I I + E C did not differ between the two g r o u p s of mice (2.02 _ 0.36% positive cells in uninfected mice and 2.19 _ 0.46% in infected mice, n = 4 mice). I n addition, the m e a n fluorescence intensity of M H C class II + E C from control and M A I D S mice was c o m p a r a b l e (1059 ___ 49 in uninfected mice and 1017 + 10 in infected mice, n = 4 mice).

Contact sensitivity It has previously been s h o w n t h a t T-cell function in M A I D S is deficient [14-16]. W h e n c o n t r o l and M A I D S mice were painted with T N C B a n d challenged 5 days later on the ear, only the c o n t r o l mice responded with a significant ear swelling response (controls, delta = 23.3 + 4.2 m m • 10 -z, n = 6; M A I D S , delta = 5.0 ___ 4.6 m m • 10 -2, n = 4; challenge only, delta = 7.3 + 1.8 m m • 10 -2, n = 6). Thus, the failure to develop c o n t a c t sensitivity in M A I D S mice, as seen in preliminary experiments, might have been due to a T-cell defect. T o test L C function of M A I D S mice, we p r e p a r e d trinitrophenyl-modified E C f r o m c o n t r o l and infected animals. The cells were injected into naive syngeneic recipients. This p r o c e d u r e lead to successful induction of c o n t a c t sensitivity in b o t h g r o u p s of mice (mice injected with control trinitrophenyl-EC, delta = 18.0 + 3.5 m m • 10 -2, n = 4; mice injected with M A I D S trinitrophenyl-EC, delta = 19.0 _+ 1.7 m m • 10 -2, n = 3; challenge o n l y , delta = 7.3 + 1.8 m m x 10 -2, n = 6) indicating an intact c o n t a c t sensitivity-inducing capacity of M A I D S LC.

APC function T o investigate the A P C function o f EC, allogeneic mixed E C - l y m p h o c y t e cultures were prepared. W h e n n o r m a l B A L B / c E C were used as stimulators, a strong response o f the control B6 T cells was observed (Fig. 1A). In contrast, T cells f r o m infected mice s h o w e d no response (Fig. 1 B) indicating a deficient T-cell responsiveness in M A I D S mice as previously reported [14-16]. I n contrast, E C p r e p a r e d f r o m the same infected mice s h o w n to have defective T-cell function elicited proliferative responses f r o m B A L B / c T cells equivalent to the responses induced by E C f r o m uninfected mice (Fig. 1 A). This indicated preserved allostimulatory capacity o f M A I D S EC. To reveal any A P C dysfunction n o t detectable with allostimulation a n d in order to distinguish between

191

12

(~

0

30

ECNL A

x

x

,t: 4

2 ~,

o

[]

NL T cells B

o

J I

150

I

I

I

150

75 37 18 epidermal cell number x 10-3

75

abnormalities related to antigen processing or presentation, we used the T-cell hybridoma BO17.10. This hybridoma responds to the ovalbumin peptide 323-339 a n d I A b [19]. It produces IL-2 independently of second signals. The response to the intact protein is dependent on both processing and presenting function whereas the response to the peptide depends on presenting function alone. When MAIDS spleen cells were used as APC in combination with the peptide, the hybridoma response was clearly diminished when compared with the response to normal spleen cells (Fig. 2B). This indicates a severe presenting defect in MAIDS spleen cells. In contrast, the response to MAIDS EC plus peptide was preserved when compared with control EC (Fig. 2A), demonstrating the absence in MAIDS EC of the presenting defect seen in MAIDS spleen cells. When intact ovalbumin was used as antigen, again, compared with normal spleen cells, a

(~

O ECNL A

'~

os

37

1"8

epidermal cell number x 10-3

I

[~

Fig. l A, B. Mixed EC lymphocyte reaction with cells from MAIDS mice and normal control mice. Various numbers of 1500-rad-irradiated EC from normal control B6 mice (A) (NL), or from B6 mice with MAIDS (A), or from Balb/c mice (B) were cultured for 7 days with l0 s nylon-wool-passed spleen cells from Balb/c mice (A), from normal control B6 (B) (NL) or from B6 mice with MAIDS (B). Proliferation was assessed by 3H-thymidine incorporation during the final 12 h of culture. Points represent the mean of triplicate cultures. Standard deviations, usually < 15%, are not shown

deficient response with MAIDS spleen cells was observed (Fig. 2D). EC from control and MAIDS mice induced a similar response to intact ovalbumin (Fig. 2C), demonstrating the preserved processing function of the MAIDS EC for intact ovalbumin protein.

Discussion

Direct assessment of the presence of all components of the LP-BM5 retrovirus mixture in skin of infected mice is complicated by the fact that the normal mouse genome contains numerous endogenous species that cross-react with probes for each of the LP-BM5 viruses. However, recent studies have shown that MAIDS can be transferred by skin grafts [18]. Moreover, ecotropic virus can be recovered from cell-free extracts prepared from the skin (J. W. Hartley, A. S. Rosenberg, H. C. Morse III, un-

I-ISCNL

~ ,~o

B ]

s

lOO1

' 8

!t

8 sot

'~

~o

3ol t

zs

o.6

o.,,

o9 ~c.~ c ECMAIDS

%

;o

Peptbdeconcentration (~g/mi)

~'o

Ls

o'.6

o.'~,

75

x

z

~

o

i,: flol-

so

'g 2s 2 180

60 20 cell numberx 10-3

7

180

gO ~;0 cell ~umberx 10.3

7

Fig. 2 A - D . Spleen cells, but not EC, from mice with MAIDS show an APC defect. BO.17.10 cells (5 X 104) were cultured for t day with either ovalbumin peptide 323-339 at the indicated concentrations (A, B) or intact ovalbumin protein (10 mg/ml) (C, D) and EC (A, C) or spleen cells (SC) (B, D) from normal B6 control mice (NL) or from M A I D S mice. The titration of the peptide was performed at 1.8 x 105 EC or spleen cells per well. CTLL growth-promoting activity of the supernatants diluted at 1 : 4 was determined with CTLL cells as previously described [8]. The CTLL response was determined by 3H-thymidine incorporation. Points represent the mean of triplicate cultures. Standard deviations, usually < 15%, are not shown

192 published observations). Although the cellular localization of these viruses is not known, they m a y reside in bone marrow-derived cells, including LC, as a high proportion of bone m a r r o w cells from mice with M A I D S are productively infected (R. A. Yetter, H. C. Morse III, unpublished observations). This view is supported by the observation that skin from M A I D S mice is rejected by uninfected animals, a process most likely requiring activation of T cells by LC from the graft [18]. O u r studies confirm and extend previous results obtained by Mizuochi et al. demonstrating an antigenpresenting defect in adherent spleen cells of mice with M A I D S [14]. This defect was ascribed, at least in part, to impaired costimulatory activity of A P C for certain helper T-cell clones. The present analyses of splenic A P C function were performed with a T-cell h y b r i d o m a that does not require costimulation for activation by antigen. The reduced ability of spleen cells from infected mice to present the ovalbumin peptide to this cell line suggests that alterations other than those affecting costimulation contribute to this abnormality. As processing precedes presentation, the presentation defect identified here did not allow the detection of any processing abnormality by the use of intact ovalbumin protein. M o r e recent experiments by Hiigin et al. suggest that a virus-encoded superantigen presented on infected B cells m a y be the driving force of a rapid and extensive T-cell activation and of the subsequent breakdown of both humoral and cellar immunity [9]. The presence of virus-encoded superantigens on splenic A P C in M A I D S m a y have interfered with the presentation of the ovalburain peptide. The role of A P C other than B cells in such a scenario remains to be elucidated, but the failure of mice depleted of B cells to develop T-cell abnormalities after infection suggest that other APC, including macrophages, dendritic cells and LC, are not crucial for development of M A I D S [4]. O u r findings of conserved epidermal LC m o r p h o l o g y and function in the presence of a splenic A P C defect are in accord with the above model and studies done in individuals infected with H I V [11].

Acknowledgements. This study was supported in part by the Geigy-Jubilfiums-Stiftung (to A.C.), the Foundation for the Support of AIDS Research in Switzerland (grant No 7009 to A.C.), the Swiss National Foundation for Scientific Research (grant No 3.115-0.88 and 32-27/59.89 to C.H. and 31.28782.90 to SI) and contract NO 1-AI-72622 to Microbiologic Associates, Inc., Bethesda (to H.C.M.).

References 1. AzizDC, HannaZ, JolicoeurP (1989) Severeimmunodeficiency induced by a defective murine leukemia virus. Nature 338: 505-508 2. Basham T, Holdener T, Merigan T (1991) Intermittent, alternating, and concurrent regimens of zidovudine and 2'-3'dideoxycytidine in the LP-BM5 murine induced immunodeficiency model. J Infect Dis 163:869-872 3. Belsito DV, Sanchez MR, Baer RL, Valentine F, Thorbecke GJ (1984) Reduced Langerhans' cell IA antigen and ATPase activity in patients with the acquired immunodeficiency syndrome. N Engl J Med 310:1279-1279 4. Cerny A, Hfigin AW, Hardy RR, Hayakawa K, Zinkernagel

RM, Makino M, Morse III HC (1990) B cells are required for induction of T cell abnormalities in a routine retrovirus-induced immunodeficiency syndrome. J Exp Med 171:315-320 5. Cerny A, Merino R, Makino M, Waldvogel FA, Morse III HC, Izui S (1991) Protective effect of cyclosporin A on immune abnormalities observed in the murine acquired immunodeficiency syndrome. Eur J Immunol 21:1747-1750 6. Chattopadhyay SK, Morse III HC, Makino M, Ruscetti SK, Hartley JW (1989) A defective virus is associated with induction of a murine retrovirus-induced immunodeficiency syndrome, MAIDS. Proc Natt Acad Sci USA 86:3862-3866 7. Hartley JW, Fredrickson TN, Yetter RA, Makino M, Morse III HC (1989) Retrovirus-induced murine acquired immunodeficiency syndrome: natural history of infection and differing susceptibility of inbred mouse strains. J Virol 64:1223-1231 8. Hauser C, Snapper CM, Ohara J, Paul WE, Katz SI (1989) T helper cells grown with hapten-modified Langerhans cells produce interleukin 4 and stimulate IgE production. Eur J Immunol 19:245-251 9. Hiigin AW, Vacchio MS, Morse III HC (1991) A virus-encoded 'superantigen' in a retrovirus-induced immunodeficiency syndrome of mice. Science 252:424-427 10. Juhlin L, Shelly WB (1977) New staining techniques for the Langerhans cell. Acta Derm Vencreol (Stockh) 57:289-296 I1. KaIter DC, Greenhouse J J, Orenstein ]'M, Schnittman SM, Gendelman HE, Melzer MS (1991) Epidermal Langerhans cells are not principal reservoirs of virus in HIV disease. J Immunol 146:3396-3404 12. Kanitakis J, Escaich S, Trepo C, Thivolet J (1991) Detection of human immunodeficiency virus-DNA and RNA in the skin of HIV-infected patients using the polymerase chain reaction. J Invest Dermatol 97:91-96 13. Main RK, Cochrum KC, Jones MJ, Kountz SL (1971) Immunological potential of the in vitro mixed skin cell-leukocyte reaction. Nature 229:89 9t 14. Mizuochi T, Mizuguchi J, Uchida T, Ohnishi K, Nakanishi M, Asano Y, Kakiuchi T, Fukushima Y, Okuyama K, Morse III HC, Komuro K (1990) A selective signaling defect in helper T cells induced by antigen-presenting cells from mice with murine acquired immunodeficiency syndrome. J Immunol 144: 313-316 15. Morse III HC, Yetter RA, Via CS, Hardy RR, Cerny A, Hayakawa K, Hiigin AW, Miller MW, Holmes KL, Shearer GM (1989) Functional and phenotypic alterations in T cell subsets during the course of MAIDS, a murine retrovirusinduced immunodeficiency syndrome. J Immunol 143:844-850 16. Mosier DE, Yetter RA, Morse III HC (1985) Retroviral induction of acute lymphoproliferative disease and profound immunosuppression in adult C57BL/6 mice. J Exp Med 161: 766-784 17. Rappersberger K, Gartner S, Schenk P, Stingl G, Groh V, Tschachler E, Mann DL, Wolff K, Konrad K, Popovic M (1988) Langerhans cells are an actual site of HIV-1 replication. Intervirology 29:185-194 18. Rosenberg AS, Maniero TG, Morse III HC (1991) In vivo immunologic deficits in mice with the acquired immunodeficiency syndrome and the effect of LP-BM5 infection on rejection of skin from infected mice. Transplant. Proc 23: 167-169 19. Shimonkevitz R, Kappler J, Marrack P, Grey HM (1983) Antigen recognition by H-2-restricted T cells. I. Cell-free antigen processing. J Exp Med 158:303-316 20. Tamaki K, Fujiwara H, Katz SI (1981) The role of epidermal cells in the induction and suppression of contact sensitivity. J Invest Dermatol 76:275-278 21. Tschachler E, Groh V, Popovic M, Mann DL, Wolff K, Stingl G (1987) Epidermal Langerhans cells - a target for HTLVIII/LAV infection. J Invest DermatoI 88:233-237 22. Zambruno G, Mori L, Marconi A, Mongiardo N, De Rienzo B, Bertazzoni U, Gianetti A (1991) Detection of HIV-1 in epidermal Langerhans cells of HIV-infected patients using the polymerase chain reaction. J Invest Dermatol 96:979-982