Eur. J. Immunol. 1990.20: 2451-2456
Joachim Ennen, Inka Seipp, Stephen G. Norley and Reinhard Kurth
Decreased accessory cell activity of HIV 1-infccted MQ,
2451
Decreased accessory cell function of macrophages after infection with human immunodeficiency virus type 1 in vitro*
Paul-Ehrlich-Institute, Langen Peripheral blood monocytes from human immunodeficiency virus (H1V)infected individuals or AIDS-related complex/AIDS patients ex vivo exhibit distinct alterations in some but not all immune functions. In studies presented here, monocytes from healthy donors were infected with HIV 1 in v i m and co-cultures with autologous uninfected T lymphocytes were set up. The monocyte/macrophage (M@)-dependent T cell function was determined by measurement of proliferative and secretory [interleukin (IL) 2, interferon-y] responses to lectin (phytohemagglutinin), mitogen (anti-CD3 monoclonal antibody), or recall antigen (tetanus toxoid, tuberculin). Accessory function of MQ, was normal after HIV infection when optimal amounts (lOY0-20%) were added to the T lymphocytes. However, HIV infection of MQ, significantly decreased T cell proliferative responses and secretion of IL 2 when supplemented at limited dilution (0.5%-5%), although interferon-y production was not affected.Whereas the lipopolysaccharide-triggered MQ, production of IL 1 was not impaired by HIV 1infection, there was a significant decrease in this response when anti-CD3 monoclonal antibody or tetanus toxoid were used t o trigger the peripheral blood mononuclear cells. The impairment of proliferation of T lymphocytes in the presence of HIV 1-infected MQ, could be overcome by addition of exogenous IL 1.Taken together, these data clearly show that the mononuclear phagocytedependent enhancement of stimulated T cell proliferation and lymphokine secretion is decreased when the restricted numbers of monocytes/M@ are HIV 1 infected. There are, therefore, two possible roles of MQ, in HIV infection and progression to disease. First, as a reservoir and vehicle for dissemination of the virus, and second, as an immune cell whose essential functions are impaired by infection.
1 Introduction
from HIV-infected donors produced markedly elevated amounts of IL6 [17].
One of the striking features of AIDS is the reduction and final depletion of CD4+ T lymphocytes, which parallels the development of immunodeficiency [ 1, 21. Nevertheless, the loss of CD4+ T cells does not adequately explain all the immunopathological effects observed during HIV infection [3-51. Even during the early stages of HIV 1infection there is a failure of lymphocytes to recognize and respond to soluble antigens, although levels of CD4+ T lymphocytes remain normal [6, 71. A number of reports demonstrate that cells of the monocyte (Mo)/MQ,lineage are targets of the HIV [%lo], suggesting further potential mechanisms for the impaired immune responsiveness [ l l , 121. Mo from AIDS patients exhibit a significantly suppressed response to specific chemotactic ligands [ 13, 141. Furthermore, high levels of TNF-a [ 151 and IL 1 [161 are produced by purified blood Mo from HIV-infected patients and cultured PBMC [I 85171
*
This work was supported in part by the Federal Ministries of Health and of Research and Technology.
Correspondence: Joachim Ennen, Paul-Ehrlich-Institute, PaulEhrlich-Str. 51-59, D-6070 Langen 1, FRG Abbreviations: GM-CSF Granulocyte-M~D-CSF toxoid Mo: Monocyte
Tetanus
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
Besides the Mo functional activities of phagocytosis, bacterial and viral killing [MI, and tumoral cell cytotoxicity [19] the cells of the mononuclear phagocyte lineage also play an important role in the initiation and regulation of immune responses through antigen processing/presentation and monokine release [20, 211. A number of studies have reported defects in accessory activity or APC function in patients with AIDS [22,23]. The issue of whether HIV-infected hosts develop underlying defects in Mo/MQ, function prior to, or in addition to, a decrease in CD4+ cell numbers and/or disturbance of Th cell function is not yet resolved. Such a functional immunological perturbation in asymptomatic carriers, manifested as impaired lymphocyte proliferative responses to antigenic stimuli, could represent the initial cause of immunopathogenesis in AIDS. This study was designed t o examine the accessory activity of human MQ, in both antigenic and mitogenic stimulation of T lymphocyte function after infection of the MQ, with HIV 1. we present evidence that when present in excess, HIV-infected MQ, show normal accessory cell activity for T 'ymphocyte proliferation Or 'ymphokine secretion*However, PBMC populations composed of T lymphocytes supplemented with suboptimal amounts of M@ exhibit significantly decreased MQ-dependent Tcell activation if M a are infected with HIV 1.
+
0014-2980/90/1111-2451$3.50 .25/0
2452
Eur. J. Immunol. 1990.20: 2451-2456
J. Ennen, I. Seipp, S. G. Norley and R. Kurth
2 Material and methods 2.1 Isolation and purification of human PBMC Human PBMC were isolated by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) gradient centrifugation from buffy coats of healthy blood donors. T lymphocytes were purified by rosetting with neuraminidase-treated SRBC followed by an additional density gradient centrifugation step [24]. Mo were purified by adherence to plastic tissue culture flasks precoated with FCS at 4°C for 30 min [25]. Mo were cultured in the presence of 500 U/ml recombinant human granulocyte-M@ (GM)-CSF (Boehringer Mannheim, Mannheim, FRG) in RPMI 1640 containing 10% FCS, 2 mM L-glutamine, and antibiotics [26]. T h eT lymphocytes were frozen and stored in liquid nitrogen until use. 2.2 Virus stocks and infection of MOM@ The HTLV-IIIB strain of HIV was serially passaged in human peripheral Mo/M@ target cells. Over time, virus in the SN of these cultures showed an increased ability to infect fresh monocyte cultures, as determined by measurement of infected cell numbers and of SN HIV 1-p24 levels. This Mo-adapted HIV strain was termed HTLV-IIIB-MO. GM-CSF-treated MOM@growing as monolayers in culture flasks were inoculated on day 2 with lo5 TCIDSOcell-free HIV 1 (titration on MT4 cells) and allowed t o adsorb overnight. Subsequently, cultures were washed and supplemented with fresh medium. HIV 1-infected cells were tested for expression of HIV 1 antigens by immunofluorescence assay. HIV 1-p24 protein in SN was detected using an ELISA antigen-trapping assay [27].
The amount of IL2 secreted was evaluated by measuring the proliferative response of the IL 2-dependent murine cytotoxic cell line CTL-6 in the presence of culture SN. For the quantitative evaluation, a reference standard was included in each assay. Culture SN were screened for IL l p production using an antigen-capture ELISA testkit (Biochrom, Berlin, FRG). Where appropriate, cultures were supplemented with recombinant human IL 1p (Boehringer Mannheim).
3 Results 3.1 HIV infection of MOM@ HIV infection of Mo/M@ was monitored by indirect immunofluorescence assay of infected cells and by measurement of HIV 1-p24 in culture SN. Cells that were HIV 1-antigen positive were first detected between 6 and 9 days after infection (Fig. l ) , after which the number of infected cells increased continuously to a mean of 49% on day 24. HIV 1-p24 antigen was first detected on day 3 to 6, and the kinetics of p24 release paralleled the appearance of infected cells. Taken together, infection with the Moadapted HIV 1strain HTLV-IIIB-MO resulted in a significant infection of the MO culture, which is in contrast to infection with the parental strain HTLV-IIIB (data not shown). 3.2 Accessory function of HIV 1-M@in T cell function
On day 21 or 24 after infection with HIV 1, M@ cultures were harvested and used to supplement autologous, uninfected T lymphocyte cultures at a ratio of 1/5. Proliferation was triggered either by mitogens (PHA, anti-CD3 mAb) or 2.3 Co-cultures of Mo and T lymphocytes specific antigens (PPD, TT) and DNA synthesis was On day21 to 24 purified T lymphocytes were thawed, subsequently measured. Controls included cultures supplewashed, and aliquoted into U-well microtiter plates at lo5 mented with uninfected MQ and cultures containing T cells/well. HIV 1-infected MQ or control MQ were added in lymphocytes alone. Only background [3H]dThd incorporaamounts of 0.5% to 20% in relation t o T lymphocytes. For mitogen stimulation PHA (5 pg/ml; Wellcome, Dartford, GB) or anti-CD3 mAb (2 pg/ml, IgG2,; Dianova, Hamburg, FRG) was then added to the culture. Antigenic stimulation was performed by addition of soluble tetanus toxoid (TI') diluted 1/5000 (4285 Lf/ml; Behring Institute, Marburg, FRG) or PPD (Tuberculin-GGT, Behring) at a concentration of 25 pg/ml. Monokine secretion was triggered with LPS from S. abortus equi (10 pg/ml; Sigma, Taufkirchen, FRG), while controls for in viwo activated T cells received 100 U/ml of purified human IL2 (Biotest, Dreieich, FRG). O n day 3 (mitogenic stimulation) or day 5 (antigenic stimulation) culture SN were transferred to microtiter plates for evaluation of cytokine activities. Cell cultures were then pulsed overnight with 1 pCi/well 0 10 20 30 = 37 kBq/well of [3H]dThd and, finally, cells were hardays p i . vested for measurement of [3H]dThd incorporation into DNA. Data are presented as mean cpm & SD of triplicate Figure 1. Kinetics of appearance of HIV 1+ cells and secretion of cultures. HIV 1-p24 antigen in cultures of human M@ after infection with 2.4 Assays for cytokines
The antigen-capture test for quantification of human IFN-y in culture SN was performed as described previously [25].
HIV 1.Two days after purification of human Mo and culture in the presence of GM-CSF (500 Ulml) cells were infected with HIV (day 0). Percentage of cells staining positive for HIV 1proteins (W) was measured in the indirect immunofluorescence assay. The amount of HIV 1core antigen p24 in the SN (0)was measured by ELISA as described in Sect. 2.2.
Decreased accessory cell activity of HIV 1-infected MQ,
Eur. J. Immunol. 1990. 20: 2451-2456 tion was detected in cultures of infected and uninfected MQ, pretreated with GM-CSF (Fig. 2a). However, there was low but significant DNA synthesis detected in highly enriched T cell cultures triggered by mitogen o r by antigen.There was a pronounced cooperative effect between M@ and T lympho-
2453
cytes in the mixed cultures. This enhancement resulting from cell-to-cell interaction, was not affected by HIV 1 infection of the MQ,. In addition, culture SN were screened in parallel for biologically active IL2 and for the presence of I F N y (Fig. 2b and c). Here, a similar pattern was observed, i.e. no IL 2 or IFN-y activity in SN from pure MQ, and only low activity in culture fluids from T lymphocytes. However, there was significant activity of both lymphokines when co-cultures of M a and Tcells were set up. During co-culture, HIV 1-infected MQ, release virus particles into the culture fluid.To control the effect of infectious virus or free viral proteins on proliferative and secretory activities of T lymphocytes we added lo3 to lo5 TCID50 of HIV 1 to co-cultures. No inhibitory effect was observed in control cultures supplemented with infectious HIV l-standard SN (data not shown). The number of infected M@ in culture after inoculation with HIV 1 was rather variable (between 31% to 59% HIV+ cells on day 24 post infection) and there was consequently a considerable number of uninfected M@ present in the co-cultures of HIV 1-infected MQ, with T lymphocytes. We therefore examined T lymphocyte function in the presence of different quantities of infected or uninfected [email protected], the proliferative response of Tcells to anti-CD3 antibody decreased as the number of MQ, became limiting, and in the presence of only 2% MQ, the proliferative response was half-maximal (Fig. 3a). In the cultures consisting of T lymphocytes plus HIV 1-infected M@ the titration curve was strikingly different in the range of 0.5% to 5% M a supplementation. Here supplementation with 2% HIV l-infected MQ, resulted in only 15% of the maximal Tcell response. Similar results were obtained when the antigen 'IT was used as the T cell trigger (Fig. 3a).
._ +*
z:
0
8
5:
-8 rn
+
rn
% ++ " 8
C ~ population U
2
rn
0
+
.-8+
z
Figure2. Accessory activity of M@ after HIV 1 infection as detected in Tcell proliferation or lymphokine secretion assays. On day 21 to 24 after inoculation of Mo with HIV 1the M@ cultures were harvested and PBMC cultures of autologousT lymphocytes supplemented with 20% MQ, were established. Proliferative responses (a) were measured by [3H]dThdincorporation on day 3 after PHA (m) or anti-CD3 mAb (B) stimulation and on day 5 after antigen-induced(IT ,PPD E3) stimulation. Secreted IL 2 in culture SN (b) was measured in the IL2-dependent murine CTL-6 cell assay and activity of IFN-y was evaluated in specific ELISA (c). Controls include LPS supplementation ( ), IL2 addition (B) and medium control (0).IL2 units were calculatedby comparison with an internal laboratory standard and IFN-y levels were compared with an rIFN-y standard. Results are expressed as the average of triplicates f SD. The data are representative of five replicate experiments.
These data suggest that HIV 1-infected MQ, have a reduced ability t o provide accessory function t o T 1ymphocytes.This fact is most obvious when restricted numbers of MQ, are present in the culture. However, at optimal concentrations the uninfected fraction in the HIV 1-infected MQ, population provides sufficient accessory activity to give normal proliferative Tcell responses. There were differences in the effect that infection of MQ, had on lymphokine production in mixed cultures stimulated by mitogen o r antigen. Whereas levels of IL2 decreased in parallel with the lymphoproliferative response (Fig. 3b), there was no such inhibition of I F N y production (Fig. 3c).
3.3 IL1 activity in co-culture SN Since IL 1 is involved in mitogen- and antigen-triggered MQ-dependent enhancement of T lymphocyte proliferation and lymphokine secretion, we screened SN for activity of this monokine. Fig. 4a shows the results for IL1 production by antLCD3 mAb-stimulated PBMC cultures. No differences in the low I L 1 activity in SN from pure HIV 1-infected M a or control M@ were observed. Only in co-cultures of MQ, withTcells could we detect considerable enhancement of I L 1 activity resulting from cell-to-cell interaction. However, IL 1secretion was significantly lower
2454
Eur. J. Immunol. 1990. 20: 2451-2456
J. Ennen, I. Seipp, S. G . Norley and R. Kurth
Tr
20
0
-
Tr I
I
I
I
I
I
I
l
l
l
l
% of macrophages added to T cells
Figure 3. Titration of HIV 1-infected M@ accessory activity onT lymphocyte function. Increasing numbers of HIV 1-infected MO (m) or uninfected control MO ( 0 )were co-cultured with autologous T lymphocytes. On day 3 (anti-CD3 mAb stimulation) or day 5 (IT stimulation) proliferation (a) was measured and activities of IL2 (b) and IFN-y (c) in the SN were determined.The responses of PBMC populations of T lymphocytes plus 20% MQ supplementation were referred to as 100%.
when HIV 1-infected M@ were present in PBMC cultures triggered by anti-CD3 mAb. Similar results we obtained with TT as the T cell trigger (data not shown). To answer the question of whether I L 1 production is generally disturbed in HIV 1-infected M@,werepeated the experiments with the addition of LPS stimulation of the cultures. In fact no cooperative effect of T lymphocytes and M a (with respect to enhanced I L 1 production) was observed and, furthermore, IL 1secretion triggered by LPS was not affected by HIV 1 infection of the M@ (Fig. 4b). Finally, LPS-triggered stimulation of IL 1 release is threeto sevenfold higher than anti-CD3-mediated induction. To evaluate the role of IL1 in the impairment of MQ-T lymphocyte cooperation further we supplemented antiCD3-stimulated cultures with exogenous human rIL 1. As shown inTable 1there were high proliferative responses by T lymphocytes even in the presence of HIV 1-infected M@. We conclude from these data that interaction of mitogenor antigen-stimulated T lymphocytes with M@ is affected by HIV 1infection of the M@ and that a possible reason for this low proliferation is the significantly decreased production of IL 1which can be restored by exogenous addition of this monokine.
4 Discussion In this study we examined the functional capacity of HIV 1-infected M@ to cooperate with T lymphocytes in antigen- or mitogen-triggered T lymphocyte responses. We observed a reduced ability of M a to support the proliferation of T lymphocytes and secretion of IL2 after infection with HIV 1. The inhibition of cooperation-dependent T cell proliferation in the presence of HIV 1-infected M@ in vitro is a novel observation, although the mechanism of inhibition is so far unknown. Proliferation of PBMC in response to soluble antigens involves many steps including processing and presentation, T lymphocyte activation and lymphokine secretion [21]. In antigenic stimulation the processing/presentation of antigen and interaction of the MHC class I1 complex with the CD4 molecules of TI, lymphocytes are crucial steps for activation of specific T lymphocytes [28]. Clearly, the inhibition of M@ accessory activity after HIV 1 infection could occur at any of these stages. Since PHA- or anti-CD3-triggered proliferative T cell responses or IL 2 secretion are also decreased in the presence of infected M@, it is likely that expression of HIV proteins may interfere generally with monocytic accessory capacities (i.e. those common for antigen- and mitogen-triggered res-
Eur. J. Immunol. 1990. 20: 2451-2456
Decreased accessory cell activity of HIV 1-infected MQ,
2455
Table 1. Restoration of T lymphocyte proliferation in the presence of HIV 1-infected MQ, by supplementation with IL 1
100
80
Ccll population
IL 1f3 added (U/ml)
['HIdThd incorpora tion
60
T cells HIV-infected M a a )
500 125
+
40
T cells + control MQ,
125
20 h
F3
v 4
0 0,1
None 500
1
10
100
Tcells alone HIV-infected MQ, alone Control MQ,alone
None None
None None
42 300
+
3700 38700 f 5150 15200 f 2 m 49 200 t 4sw 16400 f 62W 48 300 _t SYOO 8700 k 1100 2100+ Y70 33w+ 700
a) Cultures of T lymphocytes plus 2% MQ, were stimulated for 3 days with anti-CD3 mAb. rIL l p was added as indicated on day 0 of culture. Data are expressed as mean f SD from one of
three representative experiments.
Interestingly, the phenomenon of reduced IL 1 release by HIV-infected M a in mitogen- or antigen-driven T lymphocyte cultures is not a common feature of these cells. In contrast, LPS induction of IL1 secretion is similar in control M a and HIV 1-infected M a (Fig. 4b),whereas the cell-to-cell interaction-dependent enhancement of IL 1 secretion stimulated by anti-CD3 mAb (Fig. 4a) or T T (data not shown) was significantly decreased. In further 0,1 1 10 100 support of a causative role for low IL1 levels in the % macrophages added to culture inhibition of Tcell function in the presence of HIV-infected M a we could restore anti-CD3-triggered proliferative Figure 4. IL 1 activity in culture fluids from cultures of anti-CD3 mAb (a) or LPS-stimulated (b) MQ,or MQ, plus T lymphocytes. responses by supplementation of cultures with exogenous HIV-infected MQ, (M) or control MQ, ( 0 )were either cultured IL1 (Table 1). Similarly, it has been shown that the alone or in co-culturewithTcells (HIV-infectedMQ,plusTcells, A; reduction of Tcell proliferation in the presence of HIVcontrol MQ, plus T cells (A). Detection of IL If3 by ELISA is infected U-937 cells could be restored by addition of IL 1 or described in Sect. 2.4. IL2 [31]. ponses) rather than events specific for antigen processing or presentation. For mitogenic stimulation with antibody specific for the TcR [29], interaction with mononuclear phagocytes via their FcyR is required [30]. This cellular interaction probably provides Ma-derived signals (e.g. IL 1) which are presumably necessary for T lymphocyte activation and subsequent proliferation. We tested expression of the FcyR on the surface membrane of HIV-infected M a and failed t o detect any significant alteration (unpublished observation). In agreement with this result Petit et al. [31] reported normal expression of membrane FcyR on the HIV-infected U-937 promonocytic cell line. However, the normal expression of FcyR on mononuclear phagocytes after HIV infection does not necessarily imply full functional activity. Mann et al. [32] demonstrated a diminished ability of HIV 1-infected peripheral monocytes t o stimulate allogenic cells in comparison with uninfected cells. However, the production or secretion of monokines which could account for defective allogenic stimulation had not been measured. Our results strongly suggest that the decrease in mitogenic or antigenic triggering of T cell responses after HIV infection of M a might be due t o low IL1 production.
We have attributed the effect of inhibition of M a dependent Tcell function to Mo/MQ,HIV infection. Recent evidence suggests that purified native [33] or recombinant [7,34] gp120 may selectively suppress the immune system, presumably through its direct interaction with CD4 determinants onT lymphocytes. However, these studies use high concentrations of viral glycoproteins, whereas the addition of lo5 TCIDSOof HIV 1 to control cultures in our experiments had no significant effect on Ma-dependent enhancement of Tcell function (data not shown). The experiments studying the effect of HIV 1 infection of M a on the response of T lymphocytes to mitogen or antigen suggest that HIV infection may interfere with immune recognition independent of CD4+ T lymphocyte depletion. However, for the following reasons these in v i m observations may not be extrapolated to the situation in vivo without caution: (a) the measurement of HIV proteins or nucleid acids in the circulating Mo of seropositive patients reveals a low frequency of productively infected cells (about 1 in 104, see [35]). M a , however, are the major infected cells in the brain [8] and in BM [9, 361. Our current view is that suppression of T cell activation would most likely occur within the confines of organs or of secondary lymphoid tissues where the putative reservoirs of virally
2456
J. Ennen, I. Seipp, S. G. Norley and R. Kurth
Eur. J. Immunol. 1990. 20: 2451-2456
5 Kurth, R., Intervirology 1990, in press. 6 Shearer, G. M., Bernstein, D. C., Tung, K. S. K.,Via, C. S., Redfield, R., Salahuddin, S. Z. and Gallo, R. C., J. Immunol. 1986. 137: 2514. 7 Gurley, R. J., Ikeuchi, K., Byrn, R. A., Anderson, K. and Groopman, J. E., Proc. Natl. Acad. Sci. USA 1989. 86: 1993. 8 Koenig, S., Gendelman, H. E., Orenstein, J. M., Dal Canto, M. C., Pezeshkpour, G. H., Yungbluth, M., Janotta, F., Aksamit, F., Martin, M. A. and Fauci, A. S., Science 1986.233: 1089. 9 Gartner, S., Markovits, I?, Markovitz, D. M., Kaplan, M. H., Gallo, R. C. and Popovic, M., Science 1986. 233: 215. 10 Ho, D. D., Rota,T. R. and Hirsch, M. S., J. Clin.Invest. 1986. 77: 1712. 11 Gendelman, H. E., Orenstein, J. M., Baca, L. M.,Weiser, B., Burger, H., Kalter, D. C. and Meltzer, M. S., AIDS 1989. 3: 475. There is clear evidence of functional discrepancies of 12 Roy, S. and Wainberg, M. A,, J. Leukocyte Biol. 1988. 43: 91. monocytes from AIDS patients ex vivo and MOM@from healthy donors infected with HIV in vitro. Whereas Mo 13 Smith, F! D., Ohura, K., Masur, H., Lane, H. C., Fauci, A. S. and Wahl, S. M., J. Clin. Invest. 1984. 74: 2121. accessory function from symptomatic HIV+ patients was 14 Poli, G., Bottazzi, B., Acero, R., Bersani, L., Rossi, V., reported to be defective [22,23], we measured low accesIntrona, M., Lazzarin, A., Galli, M. and Mantovani, A., Clin. sory function in mitogen- or antigen-driven assays in vitro Exp. Immunol. 1986. 62: 136. only in conditions of LD, and not when “physiological” 15 Wright, S. C., Jewett, A., Mitsuyasu, R. and Bonavida, B., J. ratios of Mo t o T lymphocytes (e.g. 1to 5) were employed. Immunol. 1988. 141: 99. This observation is most striking in view of the very low 16 Roux-Lombard, P., Modoux, C., Cruchard, A. and Dayer, J. number of infected peripheral Mo in vivo [35] and the M., Clin. Immunol. Immunopathol. 1989. 50: 374. number of 31% to 59% infected Mo/MO in vitro (this 17 Breen, E. C., Rezai, A. R., Nakajima, K., Beall, G. N., Mitsuyasu, R.T., Hirano,T., Kishimoto,T. and Martinez-Maza, report). In addition, high levels of I L 1 are produced by O., J. Immunol. 1990. 144: 480. purified blood Mo from HIV-infected patients [16]. How18 Nathan, C. F., Trans. R. SOC. Trop. Med. Hyg. 1986. 77: 620. ever, levels of LPS-triggered IL l release by purified M 0 D. 0. and Nathan, C. F., Immunol. Today 1983. 4: infected in vitro by HIVare normal in comparison to control 19 Adams, 166. MQ, (Fig. 4b). In addition, while comparing the functional 20 Johnston, R. B, Jr., N . Engl. J. Med. 1988. 318: 747. properties of M@ infected in vitro by HIV, with mono- 21 Unanue, E. R. and Allen, P. M., Science 1987. 236: 551. nuclear phagocytes ex vivo from HIV+ individuals, the 22 Prince, H. E., Moody, D. J., Shubin, B. I. andFahey, J. L., J. experimental treatment should be considered. In our case Clin. Immunoi. 1985. 5: 21. peripheral Mo were treated with GM-CSF before infection 23 Rich, E. A., Toossi, Z., Fujiwara, H., Hanigoski, R., Lederman, M. M. and Ellner, J. J., J. Lab. Clin. Med. 1988. 112: with HIV and cultured for 3 weeks until a high percentage 174. expressed HIV proteins. Taken together, these discrepan24 Ulmer, A. J., Scholz, W., Emst, M., Brandt, E. and Flad, cies clearly indicate the possible role of cofactors (e.g. H.-D., Immunobiology 1984. 166: 238. cytokines) being responsible for dysregulation of MOM@ 25 Ennen, J., Emst, M. and Flad, H.-D., Imrnunobiology 1989. activation in HIV infection in vivo. 179: 17. 26 Koyanagi, Y., OBrien, W. A. Zhao, J. Q., Golde, D. W., These results appear to have important implications for our Gasson, J. C. and Chen, I. S.Y., Science 1988. 241: 1673. understanding of the mechanisms by which a HIV-infected 27 Miiller, C., Moritz, R., Prigge, G., Weitzel, D., Kunze, R., Kage, A. and Kottgen, E., Fresenius Z. Anal. Chem. 1988.330: individual may progress from an asymptomatic state to a 352. state of profound immunosuppression leading finally to 28 Doyle, C. and Strominger, J. L., Nature 1987.330: 256. clinical AIDS. 29 Chang, T. W., Kung, P. C., Gingras, S. P. and Goldstein, G., Proc. Natl. Acad. Sci. USA 1981. 78: 1805. The authors would like to thank Dr. Callati (Bade)for supplying us 30 Smith, K. G. C., Austyn, J. M., Hariri, G., Beverley, P. C. L. with anti-IFN-y ELISA reagents. We also thank Dr. H . Herzbeck and Morris, F! J., Eur. J. Immunol. 1986. 16: 478. (Borstel) for providing the CTL-6 cells. 31 Petit, A. J. C., Tersmette, M.,Terpstra, E G., Ruud, E.Y., De Goede, R. E. Y., Van Lier, R. A. W. and Miedema, F., J. Received April 22, 1990; in revised form July 3, 1990. Immunol. 1988. 140: 1485. 32 Mann, D. L., Gartner, S., LeSane, F., Blattner, W. A. and Popovic, M., Clin. Immunol. Immunopathol. 1990.54: 174. 5 References 33 Mann, D. L., LeSane,F., Popovic, L. O., Arthur, L. O., Robey, W. G., Blattner,W. A. and Newman, M. J., J. Immunol. 1987. 1 Lane, H. C., Depper, J. M., Greene, W. C., Whalen, B. S., 138: 2640. Waldmann,T. A. and Fauci, A. S., N. Engl. J. Med. 1985.313: 34 Shalaby, M. R., Krowka, J. F., Gregory,T. J., Hirabayashi, S. 79. E., McCabe, S. M., Kaufman, D. S., Stites, D. P. and Ammann, 2 Nicholson, J. K. A., McDougal, J. S., Jaffe, H. W., Spira,T. J., A. J., Cell. Immunol. 1987. 110: 140. Haverkos, H. W., Kennedy, M. S., Jones, B. M., Darrow,W.W., 35 Schmittmann, S. M., Psallidopoulos, M. C., Lane, H. C., Morgan,W. M. and Hubbard, M., Ann. Intern. Med. 1985.103: Thompson, L., Baseler, M., Massari, F., Fox, C. H., Salzman, 37. N. P. and Fauci, A. S., Science 1989. 245: 305. 3 Fauci, A. S., Science 1988. 239: 617. 36 Ho, D. D., Pomerantz, R. J. and Kaplan, J. C., N. Engl. J. Med. 4 Sattentau, Q. J. and Weiss, R. A., Cell 1988. 52: 631. 1987. 317: 278.
infected M@ are presumed to be in residence. (b) Definitive studies on the antigen-presenting capacity of M o m 0 from HIV+ and AIDS-related complex/AIDS patients have not been done. LD analysis of the accessory cell activity of HIV-infected MOM@ex vivo would provide information on the M o m 0 function in antigen-specific responses. However, functional inhibition may result from defective regulation of mononuclear phagocyte activity by cytokines or from direct infection of monocytes themselves. (c) Little is known about the critical monokine concentrations in the microenvironments of tissues or organs where MWT cell interactions take place. Clearly, more studies are needed to determine the biological significance of the decreased cooperative cell-to-cell interaction of HIV-infected mononuclear phagocytes with T lymphocytes.
Joachim Ennen, Inka Seipp, Stephen G. Norley and Reinhard Kurth
Decreased accessory cell activity of HIV 1-infccted MQ,
2451
Decreased accessory cell function of macrophages after infection with human immunodeficiency virus type 1 in vitro*
Paul-Ehrlich-Institute, Langen Peripheral blood monocytes from human immunodeficiency virus (H1V)infected individuals or AIDS-related complex/AIDS patients ex vivo exhibit distinct alterations in some but not all immune functions. In studies presented here, monocytes from healthy donors were infected with HIV 1 in v i m and co-cultures with autologous uninfected T lymphocytes were set up. The monocyte/macrophage (M@)-dependent T cell function was determined by measurement of proliferative and secretory [interleukin (IL) 2, interferon-y] responses to lectin (phytohemagglutinin), mitogen (anti-CD3 monoclonal antibody), or recall antigen (tetanus toxoid, tuberculin). Accessory function of MQ, was normal after HIV infection when optimal amounts (lOY0-20%) were added to the T lymphocytes. However, HIV infection of MQ, significantly decreased T cell proliferative responses and secretion of IL 2 when supplemented at limited dilution (0.5%-5%), although interferon-y production was not affected.Whereas the lipopolysaccharide-triggered MQ, production of IL 1 was not impaired by HIV 1infection, there was a significant decrease in this response when anti-CD3 monoclonal antibody or tetanus toxoid were used t o trigger the peripheral blood mononuclear cells. The impairment of proliferation of T lymphocytes in the presence of HIV 1-infected MQ, could be overcome by addition of exogenous IL 1.Taken together, these data clearly show that the mononuclear phagocytedependent enhancement of stimulated T cell proliferation and lymphokine secretion is decreased when the restricted numbers of monocytes/M@ are HIV 1 infected. There are, therefore, two possible roles of MQ, in HIV infection and progression to disease. First, as a reservoir and vehicle for dissemination of the virus, and second, as an immune cell whose essential functions are impaired by infection.
1 Introduction
from HIV-infected donors produced markedly elevated amounts of IL6 [17].
One of the striking features of AIDS is the reduction and final depletion of CD4+ T lymphocytes, which parallels the development of immunodeficiency [ 1, 21. Nevertheless, the loss of CD4+ T cells does not adequately explain all the immunopathological effects observed during HIV infection [3-51. Even during the early stages of HIV 1infection there is a failure of lymphocytes to recognize and respond to soluble antigens, although levels of CD4+ T lymphocytes remain normal [6, 71. A number of reports demonstrate that cells of the monocyte (Mo)/MQ,lineage are targets of the HIV [%lo], suggesting further potential mechanisms for the impaired immune responsiveness [ l l , 121. Mo from AIDS patients exhibit a significantly suppressed response to specific chemotactic ligands [ 13, 141. Furthermore, high levels of TNF-a [ 151 and IL 1 [161 are produced by purified blood Mo from HIV-infected patients and cultured PBMC [I 85171
*
This work was supported in part by the Federal Ministries of Health and of Research and Technology.
Correspondence: Joachim Ennen, Paul-Ehrlich-Institute, PaulEhrlich-Str. 51-59, D-6070 Langen 1, FRG Abbreviations: GM-CSF Granulocyte-M~D-CSF toxoid Mo: Monocyte
Tetanus
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
Besides the Mo functional activities of phagocytosis, bacterial and viral killing [MI, and tumoral cell cytotoxicity [19] the cells of the mononuclear phagocyte lineage also play an important role in the initiation and regulation of immune responses through antigen processing/presentation and monokine release [20, 211. A number of studies have reported defects in accessory activity or APC function in patients with AIDS [22,23]. The issue of whether HIV-infected hosts develop underlying defects in Mo/MQ, function prior to, or in addition to, a decrease in CD4+ cell numbers and/or disturbance of Th cell function is not yet resolved. Such a functional immunological perturbation in asymptomatic carriers, manifested as impaired lymphocyte proliferative responses to antigenic stimuli, could represent the initial cause of immunopathogenesis in AIDS. This study was designed t o examine the accessory activity of human MQ, in both antigenic and mitogenic stimulation of T lymphocyte function after infection of the MQ, with HIV 1. we present evidence that when present in excess, HIV-infected MQ, show normal accessory cell activity for T 'ymphocyte proliferation Or 'ymphokine secretion*However, PBMC populations composed of T lymphocytes supplemented with suboptimal amounts of M@ exhibit significantly decreased MQ-dependent Tcell activation if M a are infected with HIV 1.
+
0014-2980/90/1111-2451$3.50 .25/0
2452
Eur. J. Immunol. 1990.20: 2451-2456
J. Ennen, I. Seipp, S. G. Norley and R. Kurth
2 Material and methods 2.1 Isolation and purification of human PBMC Human PBMC were isolated by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) gradient centrifugation from buffy coats of healthy blood donors. T lymphocytes were purified by rosetting with neuraminidase-treated SRBC followed by an additional density gradient centrifugation step [24]. Mo were purified by adherence to plastic tissue culture flasks precoated with FCS at 4°C for 30 min [25]. Mo were cultured in the presence of 500 U/ml recombinant human granulocyte-M@ (GM)-CSF (Boehringer Mannheim, Mannheim, FRG) in RPMI 1640 containing 10% FCS, 2 mM L-glutamine, and antibiotics [26]. T h eT lymphocytes were frozen and stored in liquid nitrogen until use. 2.2 Virus stocks and infection of MOM@ The HTLV-IIIB strain of HIV was serially passaged in human peripheral Mo/M@ target cells. Over time, virus in the SN of these cultures showed an increased ability to infect fresh monocyte cultures, as determined by measurement of infected cell numbers and of SN HIV 1-p24 levels. This Mo-adapted HIV strain was termed HTLV-IIIB-MO. GM-CSF-treated MOM@growing as monolayers in culture flasks were inoculated on day 2 with lo5 TCIDSOcell-free HIV 1 (titration on MT4 cells) and allowed t o adsorb overnight. Subsequently, cultures were washed and supplemented with fresh medium. HIV 1-infected cells were tested for expression of HIV 1 antigens by immunofluorescence assay. HIV 1-p24 protein in SN was detected using an ELISA antigen-trapping assay [27].
The amount of IL2 secreted was evaluated by measuring the proliferative response of the IL 2-dependent murine cytotoxic cell line CTL-6 in the presence of culture SN. For the quantitative evaluation, a reference standard was included in each assay. Culture SN were screened for IL l p production using an antigen-capture ELISA testkit (Biochrom, Berlin, FRG). Where appropriate, cultures were supplemented with recombinant human IL 1p (Boehringer Mannheim).
3 Results 3.1 HIV infection of MOM@ HIV infection of Mo/M@ was monitored by indirect immunofluorescence assay of infected cells and by measurement of HIV 1-p24 in culture SN. Cells that were HIV 1-antigen positive were first detected between 6 and 9 days after infection (Fig. l ) , after which the number of infected cells increased continuously to a mean of 49% on day 24. HIV 1-p24 antigen was first detected on day 3 to 6, and the kinetics of p24 release paralleled the appearance of infected cells. Taken together, infection with the Moadapted HIV 1strain HTLV-IIIB-MO resulted in a significant infection of the MO culture, which is in contrast to infection with the parental strain HTLV-IIIB (data not shown). 3.2 Accessory function of HIV 1-M@in T cell function
On day 21 or 24 after infection with HIV 1, M@ cultures were harvested and used to supplement autologous, uninfected T lymphocyte cultures at a ratio of 1/5. Proliferation was triggered either by mitogens (PHA, anti-CD3 mAb) or 2.3 Co-cultures of Mo and T lymphocytes specific antigens (PPD, TT) and DNA synthesis was On day21 to 24 purified T lymphocytes were thawed, subsequently measured. Controls included cultures supplewashed, and aliquoted into U-well microtiter plates at lo5 mented with uninfected MQ and cultures containing T cells/well. HIV 1-infected MQ or control MQ were added in lymphocytes alone. Only background [3H]dThd incorporaamounts of 0.5% to 20% in relation t o T lymphocytes. For mitogen stimulation PHA (5 pg/ml; Wellcome, Dartford, GB) or anti-CD3 mAb (2 pg/ml, IgG2,; Dianova, Hamburg, FRG) was then added to the culture. Antigenic stimulation was performed by addition of soluble tetanus toxoid (TI') diluted 1/5000 (4285 Lf/ml; Behring Institute, Marburg, FRG) or PPD (Tuberculin-GGT, Behring) at a concentration of 25 pg/ml. Monokine secretion was triggered with LPS from S. abortus equi (10 pg/ml; Sigma, Taufkirchen, FRG), while controls for in viwo activated T cells received 100 U/ml of purified human IL2 (Biotest, Dreieich, FRG). O n day 3 (mitogenic stimulation) or day 5 (antigenic stimulation) culture SN were transferred to microtiter plates for evaluation of cytokine activities. Cell cultures were then pulsed overnight with 1 pCi/well 0 10 20 30 = 37 kBq/well of [3H]dThd and, finally, cells were hardays p i . vested for measurement of [3H]dThd incorporation into DNA. Data are presented as mean cpm & SD of triplicate Figure 1. Kinetics of appearance of HIV 1+ cells and secretion of cultures. HIV 1-p24 antigen in cultures of human M@ after infection with 2.4 Assays for cytokines
The antigen-capture test for quantification of human IFN-y in culture SN was performed as described previously [25].
HIV 1.Two days after purification of human Mo and culture in the presence of GM-CSF (500 Ulml) cells were infected with HIV (day 0). Percentage of cells staining positive for HIV 1proteins (W) was measured in the indirect immunofluorescence assay. The amount of HIV 1core antigen p24 in the SN (0)was measured by ELISA as described in Sect. 2.2.
Decreased accessory cell activity of HIV 1-infected MQ,
Eur. J. Immunol. 1990. 20: 2451-2456 tion was detected in cultures of infected and uninfected MQ, pretreated with GM-CSF (Fig. 2a). However, there was low but significant DNA synthesis detected in highly enriched T cell cultures triggered by mitogen o r by antigen.There was a pronounced cooperative effect between M@ and T lympho-
2453
cytes in the mixed cultures. This enhancement resulting from cell-to-cell interaction, was not affected by HIV 1 infection of the MQ,. In addition, culture SN were screened in parallel for biologically active IL2 and for the presence of I F N y (Fig. 2b and c). Here, a similar pattern was observed, i.e. no IL 2 or IFN-y activity in SN from pure MQ, and only low activity in culture fluids from T lymphocytes. However, there was significant activity of both lymphokines when co-cultures of M a and Tcells were set up. During co-culture, HIV 1-infected MQ, release virus particles into the culture fluid.To control the effect of infectious virus or free viral proteins on proliferative and secretory activities of T lymphocytes we added lo3 to lo5 TCID50 of HIV 1 to co-cultures. No inhibitory effect was observed in control cultures supplemented with infectious HIV l-standard SN (data not shown). The number of infected M@ in culture after inoculation with HIV 1 was rather variable (between 31% to 59% HIV+ cells on day 24 post infection) and there was consequently a considerable number of uninfected M@ present in the co-cultures of HIV 1-infected MQ, with T lymphocytes. We therefore examined T lymphocyte function in the presence of different quantities of infected or uninfected [email protected], the proliferative response of Tcells to anti-CD3 antibody decreased as the number of MQ, became limiting, and in the presence of only 2% MQ, the proliferative response was half-maximal (Fig. 3a). In the cultures consisting of T lymphocytes plus HIV 1-infected M@ the titration curve was strikingly different in the range of 0.5% to 5% M a supplementation. Here supplementation with 2% HIV l-infected MQ, resulted in only 15% of the maximal Tcell response. Similar results were obtained when the antigen 'IT was used as the T cell trigger (Fig. 3a).
._ +*
z:
0
8
5:
-8 rn
+
rn
% ++ " 8
C ~ population U
2
rn
0
+
.-8+
z
Figure2. Accessory activity of M@ after HIV 1 infection as detected in Tcell proliferation or lymphokine secretion assays. On day 21 to 24 after inoculation of Mo with HIV 1the M@ cultures were harvested and PBMC cultures of autologousT lymphocytes supplemented with 20% MQ, were established. Proliferative responses (a) were measured by [3H]dThdincorporation on day 3 after PHA (m) or anti-CD3 mAb (B) stimulation and on day 5 after antigen-induced(IT ,PPD E3) stimulation. Secreted IL 2 in culture SN (b) was measured in the IL2-dependent murine CTL-6 cell assay and activity of IFN-y was evaluated in specific ELISA (c). Controls include LPS supplementation ( ), IL2 addition (B) and medium control (0).IL2 units were calculatedby comparison with an internal laboratory standard and IFN-y levels were compared with an rIFN-y standard. Results are expressed as the average of triplicates f SD. The data are representative of five replicate experiments.
These data suggest that HIV 1-infected MQ, have a reduced ability t o provide accessory function t o T 1ymphocytes.This fact is most obvious when restricted numbers of MQ, are present in the culture. However, at optimal concentrations the uninfected fraction in the HIV 1-infected MQ, population provides sufficient accessory activity to give normal proliferative Tcell responses. There were differences in the effect that infection of MQ, had on lymphokine production in mixed cultures stimulated by mitogen o r antigen. Whereas levels of IL2 decreased in parallel with the lymphoproliferative response (Fig. 3b), there was no such inhibition of I F N y production (Fig. 3c).
3.3 IL1 activity in co-culture SN Since IL 1 is involved in mitogen- and antigen-triggered MQ-dependent enhancement of T lymphocyte proliferation and lymphokine secretion, we screened SN for activity of this monokine. Fig. 4a shows the results for IL1 production by antLCD3 mAb-stimulated PBMC cultures. No differences in the low I L 1 activity in SN from pure HIV 1-infected M a or control M@ were observed. Only in co-cultures of MQ, withTcells could we detect considerable enhancement of I L 1 activity resulting from cell-to-cell interaction. However, IL 1secretion was significantly lower
2454
Eur. J. Immunol. 1990. 20: 2451-2456
J. Ennen, I. Seipp, S. G . Norley and R. Kurth
Tr
20
0
-
Tr I
I
I
I
I
I
I
l
l
l
l
% of macrophages added to T cells
Figure 3. Titration of HIV 1-infected M@ accessory activity onT lymphocyte function. Increasing numbers of HIV 1-infected MO (m) or uninfected control MO ( 0 )were co-cultured with autologous T lymphocytes. On day 3 (anti-CD3 mAb stimulation) or day 5 (IT stimulation) proliferation (a) was measured and activities of IL2 (b) and IFN-y (c) in the SN were determined.The responses of PBMC populations of T lymphocytes plus 20% MQ supplementation were referred to as 100%.
when HIV 1-infected M@ were present in PBMC cultures triggered by anti-CD3 mAb. Similar results we obtained with TT as the T cell trigger (data not shown). To answer the question of whether I L 1 production is generally disturbed in HIV 1-infected M@,werepeated the experiments with the addition of LPS stimulation of the cultures. In fact no cooperative effect of T lymphocytes and M a (with respect to enhanced I L 1 production) was observed and, furthermore, IL 1secretion triggered by LPS was not affected by HIV 1 infection of the M@ (Fig. 4b). Finally, LPS-triggered stimulation of IL 1 release is threeto sevenfold higher than anti-CD3-mediated induction. To evaluate the role of IL1 in the impairment of MQ-T lymphocyte cooperation further we supplemented antiCD3-stimulated cultures with exogenous human rIL 1. As shown inTable 1there were high proliferative responses by T lymphocytes even in the presence of HIV 1-infected M@. We conclude from these data that interaction of mitogenor antigen-stimulated T lymphocytes with M@ is affected by HIV 1infection of the M@ and that a possible reason for this low proliferation is the significantly decreased production of IL 1which can be restored by exogenous addition of this monokine.
4 Discussion In this study we examined the functional capacity of HIV 1-infected M@ to cooperate with T lymphocytes in antigen- or mitogen-triggered T lymphocyte responses. We observed a reduced ability of M a to support the proliferation of T lymphocytes and secretion of IL2 after infection with HIV 1. The inhibition of cooperation-dependent T cell proliferation in the presence of HIV 1-infected M@ in vitro is a novel observation, although the mechanism of inhibition is so far unknown. Proliferation of PBMC in response to soluble antigens involves many steps including processing and presentation, T lymphocyte activation and lymphokine secretion [21]. In antigenic stimulation the processing/presentation of antigen and interaction of the MHC class I1 complex with the CD4 molecules of TI, lymphocytes are crucial steps for activation of specific T lymphocytes [28]. Clearly, the inhibition of M@ accessory activity after HIV 1 infection could occur at any of these stages. Since PHA- or anti-CD3-triggered proliferative T cell responses or IL 2 secretion are also decreased in the presence of infected M@, it is likely that expression of HIV proteins may interfere generally with monocytic accessory capacities (i.e. those common for antigen- and mitogen-triggered res-
Eur. J. Immunol. 1990. 20: 2451-2456
Decreased accessory cell activity of HIV 1-infected MQ,
2455
Table 1. Restoration of T lymphocyte proliferation in the presence of HIV 1-infected MQ, by supplementation with IL 1
100
80
Ccll population
IL 1f3 added (U/ml)
['HIdThd incorpora tion
60
T cells HIV-infected M a a )
500 125
+
40
T cells + control MQ,
125
20 h
F3
v 4
0 0,1
None 500
1
10
100
Tcells alone HIV-infected MQ, alone Control MQ,alone
None None
None None
42 300
+
3700 38700 f 5150 15200 f 2 m 49 200 t 4sw 16400 f 62W 48 300 _t SYOO 8700 k 1100 2100+ Y70 33w+ 700
a) Cultures of T lymphocytes plus 2% MQ, were stimulated for 3 days with anti-CD3 mAb. rIL l p was added as indicated on day 0 of culture. Data are expressed as mean f SD from one of
three representative experiments.
Interestingly, the phenomenon of reduced IL 1 release by HIV-infected M a in mitogen- or antigen-driven T lymphocyte cultures is not a common feature of these cells. In contrast, LPS induction of IL1 secretion is similar in control M a and HIV 1-infected M a (Fig. 4b),whereas the cell-to-cell interaction-dependent enhancement of IL 1 secretion stimulated by anti-CD3 mAb (Fig. 4a) or T T (data not shown) was significantly decreased. In further 0,1 1 10 100 support of a causative role for low IL1 levels in the % macrophages added to culture inhibition of Tcell function in the presence of HIV-infected M a we could restore anti-CD3-triggered proliferative Figure 4. IL 1 activity in culture fluids from cultures of anti-CD3 mAb (a) or LPS-stimulated (b) MQ,or MQ, plus T lymphocytes. responses by supplementation of cultures with exogenous HIV-infected MQ, (M) or control MQ, ( 0 )were either cultured IL1 (Table 1). Similarly, it has been shown that the alone or in co-culturewithTcells (HIV-infectedMQ,plusTcells, A; reduction of Tcell proliferation in the presence of HIVcontrol MQ, plus T cells (A). Detection of IL If3 by ELISA is infected U-937 cells could be restored by addition of IL 1 or described in Sect. 2.4. IL2 [31]. ponses) rather than events specific for antigen processing or presentation. For mitogenic stimulation with antibody specific for the TcR [29], interaction with mononuclear phagocytes via their FcyR is required [30]. This cellular interaction probably provides Ma-derived signals (e.g. IL 1) which are presumably necessary for T lymphocyte activation and subsequent proliferation. We tested expression of the FcyR on the surface membrane of HIV-infected M a and failed t o detect any significant alteration (unpublished observation). In agreement with this result Petit et al. [31] reported normal expression of membrane FcyR on the HIV-infected U-937 promonocytic cell line. However, the normal expression of FcyR on mononuclear phagocytes after HIV infection does not necessarily imply full functional activity. Mann et al. [32] demonstrated a diminished ability of HIV 1-infected peripheral monocytes t o stimulate allogenic cells in comparison with uninfected cells. However, the production or secretion of monokines which could account for defective allogenic stimulation had not been measured. Our results strongly suggest that the decrease in mitogenic or antigenic triggering of T cell responses after HIV infection of M a might be due t o low IL1 production.
We have attributed the effect of inhibition of M a dependent Tcell function to Mo/MQ,HIV infection. Recent evidence suggests that purified native [33] or recombinant [7,34] gp120 may selectively suppress the immune system, presumably through its direct interaction with CD4 determinants onT lymphocytes. However, these studies use high concentrations of viral glycoproteins, whereas the addition of lo5 TCIDSOof HIV 1 to control cultures in our experiments had no significant effect on Ma-dependent enhancement of Tcell function (data not shown). The experiments studying the effect of HIV 1 infection of M a on the response of T lymphocytes to mitogen or antigen suggest that HIV infection may interfere with immune recognition independent of CD4+ T lymphocyte depletion. However, for the following reasons these in v i m observations may not be extrapolated to the situation in vivo without caution: (a) the measurement of HIV proteins or nucleid acids in the circulating Mo of seropositive patients reveals a low frequency of productively infected cells (about 1 in 104, see [35]). M a , however, are the major infected cells in the brain [8] and in BM [9, 361. Our current view is that suppression of T cell activation would most likely occur within the confines of organs or of secondary lymphoid tissues where the putative reservoirs of virally
2456
J. Ennen, I. Seipp, S. G. Norley and R. Kurth
Eur. J. Immunol. 1990. 20: 2451-2456
5 Kurth, R., Intervirology 1990, in press. 6 Shearer, G. M., Bernstein, D. C., Tung, K. S. K.,Via, C. S., Redfield, R., Salahuddin, S. Z. and Gallo, R. C., J. Immunol. 1986. 137: 2514. 7 Gurley, R. J., Ikeuchi, K., Byrn, R. A., Anderson, K. and Groopman, J. E., Proc. Natl. Acad. Sci. USA 1989. 86: 1993. 8 Koenig, S., Gendelman, H. E., Orenstein, J. M., Dal Canto, M. C., Pezeshkpour, G. H., Yungbluth, M., Janotta, F., Aksamit, F., Martin, M. A. and Fauci, A. S., Science 1986.233: 1089. 9 Gartner, S., Markovits, I?, Markovitz, D. M., Kaplan, M. H., Gallo, R. C. and Popovic, M., Science 1986. 233: 215. 10 Ho, D. D., Rota,T. R. and Hirsch, M. S., J. Clin.Invest. 1986. 77: 1712. 11 Gendelman, H. E., Orenstein, J. M., Baca, L. M.,Weiser, B., Burger, H., Kalter, D. C. and Meltzer, M. S., AIDS 1989. 3: 475. There is clear evidence of functional discrepancies of 12 Roy, S. and Wainberg, M. A,, J. Leukocyte Biol. 1988. 43: 91. monocytes from AIDS patients ex vivo and MOM@from healthy donors infected with HIV in vitro. Whereas Mo 13 Smith, F! D., Ohura, K., Masur, H., Lane, H. C., Fauci, A. S. and Wahl, S. M., J. Clin. Invest. 1984. 74: 2121. accessory function from symptomatic HIV+ patients was 14 Poli, G., Bottazzi, B., Acero, R., Bersani, L., Rossi, V., reported to be defective [22,23], we measured low accesIntrona, M., Lazzarin, A., Galli, M. and Mantovani, A., Clin. sory function in mitogen- or antigen-driven assays in vitro Exp. Immunol. 1986. 62: 136. only in conditions of LD, and not when “physiological” 15 Wright, S. C., Jewett, A., Mitsuyasu, R. and Bonavida, B., J. ratios of Mo t o T lymphocytes (e.g. 1to 5) were employed. Immunol. 1988. 141: 99. This observation is most striking in view of the very low 16 Roux-Lombard, P., Modoux, C., Cruchard, A. and Dayer, J. number of infected peripheral Mo in vivo [35] and the M., Clin. Immunol. Immunopathol. 1989. 50: 374. number of 31% to 59% infected Mo/MO in vitro (this 17 Breen, E. C., Rezai, A. R., Nakajima, K., Beall, G. N., Mitsuyasu, R.T., Hirano,T., Kishimoto,T. and Martinez-Maza, report). In addition, high levels of I L 1 are produced by O., J. Immunol. 1990. 144: 480. purified blood Mo from HIV-infected patients [16]. How18 Nathan, C. F., Trans. R. SOC. Trop. Med. Hyg. 1986. 77: 620. ever, levels of LPS-triggered IL l release by purified M 0 D. 0. and Nathan, C. F., Immunol. Today 1983. 4: infected in vitro by HIVare normal in comparison to control 19 Adams, 166. MQ, (Fig. 4b). In addition, while comparing the functional 20 Johnston, R. B, Jr., N . Engl. J. Med. 1988. 318: 747. properties of M@ infected in vitro by HIV, with mono- 21 Unanue, E. R. and Allen, P. M., Science 1987. 236: 551. nuclear phagocytes ex vivo from HIV+ individuals, the 22 Prince, H. E., Moody, D. J., Shubin, B. I. andFahey, J. L., J. experimental treatment should be considered. In our case Clin. Immunoi. 1985. 5: 21. peripheral Mo were treated with GM-CSF before infection 23 Rich, E. A., Toossi, Z., Fujiwara, H., Hanigoski, R., Lederman, M. M. and Ellner, J. J., J. Lab. Clin. Med. 1988. 112: with HIV and cultured for 3 weeks until a high percentage 174. expressed HIV proteins. Taken together, these discrepan24 Ulmer, A. J., Scholz, W., Emst, M., Brandt, E. and Flad, cies clearly indicate the possible role of cofactors (e.g. H.-D., Immunobiology 1984. 166: 238. cytokines) being responsible for dysregulation of MOM@ 25 Ennen, J., Emst, M. and Flad, H.-D., Imrnunobiology 1989. activation in HIV infection in vivo. 179: 17. 26 Koyanagi, Y., OBrien, W. A. Zhao, J. Q., Golde, D. W., These results appear to have important implications for our Gasson, J. C. and Chen, I. S.Y., Science 1988. 241: 1673. understanding of the mechanisms by which a HIV-infected 27 Miiller, C., Moritz, R., Prigge, G., Weitzel, D., Kunze, R., Kage, A. and Kottgen, E., Fresenius Z. Anal. Chem. 1988.330: individual may progress from an asymptomatic state to a 352. state of profound immunosuppression leading finally to 28 Doyle, C. and Strominger, J. L., Nature 1987.330: 256. clinical AIDS. 29 Chang, T. W., Kung, P. C., Gingras, S. P. and Goldstein, G., Proc. Natl. Acad. Sci. USA 1981. 78: 1805. The authors would like to thank Dr. Callati (Bade)for supplying us 30 Smith, K. G. C., Austyn, J. M., Hariri, G., Beverley, P. C. L. with anti-IFN-y ELISA reagents. We also thank Dr. H . Herzbeck and Morris, F! J., Eur. J. Immunol. 1986. 16: 478. (Borstel) for providing the CTL-6 cells. 31 Petit, A. J. C., Tersmette, M.,Terpstra, E G., Ruud, E.Y., De Goede, R. E. Y., Van Lier, R. A. W. and Miedema, F., J. Received April 22, 1990; in revised form July 3, 1990. Immunol. 1988. 140: 1485. 32 Mann, D. L., Gartner, S., LeSane, F., Blattner, W. A. and Popovic, M., Clin. Immunol. Immunopathol. 1990.54: 174. 5 References 33 Mann, D. L., LeSane,F., Popovic, L. O., Arthur, L. O., Robey, W. G., Blattner,W. A. and Newman, M. J., J. Immunol. 1987. 1 Lane, H. C., Depper, J. M., Greene, W. C., Whalen, B. S., 138: 2640. Waldmann,T. A. and Fauci, A. S., N. Engl. J. Med. 1985.313: 34 Shalaby, M. R., Krowka, J. F., Gregory,T. J., Hirabayashi, S. 79. E., McCabe, S. M., Kaufman, D. S., Stites, D. P. and Ammann, 2 Nicholson, J. K. A., McDougal, J. S., Jaffe, H. W., Spira,T. J., A. J., Cell. Immunol. 1987. 110: 140. Haverkos, H. W., Kennedy, M. S., Jones, B. M., Darrow,W.W., 35 Schmittmann, S. M., Psallidopoulos, M. C., Lane, H. C., Morgan,W. M. and Hubbard, M., Ann. Intern. Med. 1985.103: Thompson, L., Baseler, M., Massari, F., Fox, C. H., Salzman, 37. N. P. and Fauci, A. S., Science 1989. 245: 305. 3 Fauci, A. S., Science 1988. 239: 617. 36 Ho, D. D., Pomerantz, R. J. and Kaplan, J. C., N. Engl. J. Med. 4 Sattentau, Q. J. and Weiss, R. A., Cell 1988. 52: 631. 1987. 317: 278.
infected M@ are presumed to be in residence. (b) Definitive studies on the antigen-presenting capacity of M o m 0 from HIV+ and AIDS-related complex/AIDS patients have not been done. LD analysis of the accessory cell activity of HIV-infected MOM@ex vivo would provide information on the M o m 0 function in antigen-specific responses. However, functional inhibition may result from defective regulation of mononuclear phagocyte activity by cytokines or from direct infection of monocytes themselves. (c) Little is known about the critical monokine concentrations in the microenvironments of tissues or organs where MWT cell interactions take place. Clearly, more studies are needed to determine the biological significance of the decreased cooperative cell-to-cell interaction of HIV-infected mononuclear phagocytes with T lymphocytes.
