- Email: [email protected]
Macrophage-colony stimulating factor (M-CSF) stimulation induces cell death in HIV-infected human monocytes
Immunology Letters, 42 (1994) 35-40 Elsevier Science B.V. IMLET 2199
Macrophage-colony stimulating factor (M-CSF) stimulation induces cell death in HIV-infected human monocytes Alberto Bergamini a, Marcella Capozzi
a
and Mauro Piacentini b,.
a Department of Public Health and Cell Biology and b Department of Biology, University o.fRome "Tor Vergata", 00133 Rome, Italy (Received 13 December 1993; revision received 14 June 1994; accepted 15 June 1994) Key words: M-CSF, Monocyte-macrophage; Apoptosis; "Tissue" transglutaminase; HIV-gpl20; HIV-gp41
1. Summary We show here that HIV-infected monocyte-macrophages stimulated by macrophage-colony stimulating factor (M-CSF) undergo massive syncytia formation and die. The M-CSF-stimulated HIV-infected monocyte-macrophages (M/M) destroy themselves by blebbing out particles (resembling apoptotic bodies) which may contain condensed and marginated chromatin. The death of monocyte-macrophages is also characterized by the expression of "Tissue" Transglutaminase (fiG) which is one of the genes specifically expressed and activated in apoptising cells. Noteworthy, when the syncytia formation and consequently death is prevented, infected monocyte-macrophages remain viable and produce large amounts of virus for an extended period. The concentrations of M-CSF (1000 U/ml) used in this work are similar to those that stimulate macrophages in vivo. This suggests that HIV killing of M / M in the presence of M-CSF could lead, in vivo, to a greater than expected loss of immune cells and may contribute to explain the complex derangement of the immune function observed in HIV-infected patients.
2. Introduction HIV infects, in addition to T lymphocytes, other immunocompetent cells bearing on their surface the CD4 molecule, such as cells of the monocyte-macrophage lineage (M/M) [1-2]. These cells are a crucial target of the virus in vivo and account for the majority of HIV-infected cells in the body [3-4]. As a consequence of viral replication, M / M may fuse forming multinucleated giant cells (syncytia) [5]. This latter * Corresponding author: Mauro Piacentini, Dipartimento di Biologia, Via della Ricerca Scientifica 1, 00133 Rome, Italy. SSD1 0165-2478(94)001 15-8
phenomenon requires the expression on the HIV-envelope glycoprotein of 120 kDa (gpl20) and is mediated by gp-120-CD4 interaction [5-6]. Recently, the cytopathic effect of HIV has been connected to its ability to evoke in the target ceils the physiological program of cell death "apoptosis" [7-9]. Apoptosis is the silent (not associated to inflammation) physiological process of cell death leading to the controlled elimination of cells from tissues [10-11]. Independently from their origin cells undergoing apoptosis end in common, well characterized, morphological features suggesting the existence of a general pathway of physiological death regulated by specific genes [10-11]. Although the molecular mechanisms at the basis of apoptosis are not fully characterized, some biochemical features of the apoptotic cells have been established. The chromatin of apoptotic cells undergoes condensation which is often associated to its fragmentation at internucleosomal sites catalyzed by not yet established endonucleases [10-11]. The cytoplasm also becomes condensed and the cell fragments consequently to an active blebbing process that determines the formation of several membrane-bound cellular fragments (apoptotic bodies) [10-11]. A peculiar biochemical feature of these cell remnants is that their proteins are extensively cross-linked by "tissue" transglutaminases (EC 2.3.2.13; tTG), a Ca2+-dependent enzyme which is specifically induced and activated in the apoptotic cells [12-14]. The maturation and the activation of M / M are regulated by cytokine such as Macrophage-Colony Stimulating Factor (M-CSF) [16-18]. We have previously demonstrated that M-CSF (a cytokine that in vivo regulates the maturation and activation of M/M), induces, in vitro, the susceptibility of these cells to the HIV cytopathogenicity [19]. Since apoptosis has been suggested to be the cause of T-lymphocyte loss in HIV-in35
fected individuals [7-9], we investigated whether the M-CSF associated, HIV-induced cytopathogenicity in in vitro infected M / M could be mediated by apoptosis.
3. Materials and Methods
3.1. Cells M / M (from the peripheral blood of HIV-negative donors enriched for mononuclear cells (PBMC) by centrifugation over Ficoll Hypaque) were purified by two hour adherence of PBMC to Petri dishes as previously described [17]. Each experiment was performed using PBMC from a single donor. After purification, 5 × 10 4 M / M were seeded in fiat bottom, 96-wells plastic microtiter trays in 200 /xl of RPMI-1640 medium supplemented with 20% heat-inactivated fetal calf serum, 2 mM L-glutamine, 50 U / m l penicillin and 50 m g / m l streptomycin (complete medium). Ceils were cultured for 7 days with or without 1000 U / m l M-CSF (Medical System, Genova Italy).
3.2. H1V infection of M / M M / M were infected at day 7 of culture with 100 TCIDs0 of HIV-1 (HIV-1 strain HTLV-III Ba-L, which was a kind gift of Drs. S. Gartner, R.C. Gallo and M. Popovic, National Cancer Institute, N.I.H., Bethesda, MD). Corresponding mock-infected cultures were run in each experiment. Two hours after infection, M / M were extensively washed to remove excess virus and cultivated in complete medium at 37°C in a humidified atmosphere of 5% CO 2 in air with the same concentration of cytokine as before. Cells were washed and fed every 7 days. In selected experiments, a monoclonal antibody against the viral linking domain of CD4 (OKT4a) was added to the cultures at day 16. The total number of nuclei was assessed by counting at day 7, 14, 21 and 28 (cells were suspended in lysis buffer containing 10 mM KC1, 2 mM MgCI 2, 0.5% Triton X-100 in 10 mM Tris. HC1, pH 7.5) and the nuclei scored in a counting chamber using a phase contrast microscope (Laboflux D, Leitz). Cell viability was assessed on the culture dishes by Trypan Blue exclusion. The percentage of syncytia (multinucleated cells with more than three nuclei) was evaluated directly on the culture dishes after methanol fixation and Giemsa staining.
3.3. FACS analysis Expression of the CD4 protein was evaluated by flow cytometry using the Leu 3 a + b monoclonal anti36
body (Becton Dickinson). Paired isotype-specific control antibodies (IgG-1-FITC, Becton Dickinson) were run with each sample. More than 99% of the cells analyzed were LeuM3-positive monocytes. The percentage of antigen-positive cells was calculated by straight channel integration, with the integration channel set to that less than 1% of the isotype control cells appeared positive. The density of CD4 on the surface of the cells was calculated comparing the Leu 3 a + b fluorescence with appropriate fluorescein labeled isotype controls.
3.4. Immunohistochemistry Cells were cultured on slides, fixed in 2.5% paraformaldehyde and, after immunostaining with the monospecific anti-tTG antibody, counterstained with Mayer's haemalum. Immunohistochemical staining of M / M cells was performed using an affinity purified monospecific IgG raised in rabbits against human red blood cell soluble tTG (1 : 100) as primary antibody; the anti-tTG antibody was kindly supplied by Dr. L. Fesus (Department of Biochemistry, University Medical School of Debrecen, Debrecen, Hungary). Incubations with the primary antibody were carried out in a wet chamber overnight at 4°C. A biotinylated goat anti-rabbit IgG was used as second antibody, followed by a preformed streptavidin-horseradish peroxidase complex (Biogenex, USA). The reaction was developed using aminoethylcarbazole (AEC), (CRL, USA) as chromogen substrate and 0.01% H202. Endogenous peroxidase activity was blocked by methanol-H202.
3.5. Effect of HIV on M / M viability HIV- or mock-infected M / M were incubated at different time points (see Results) with 200 /zCi Na251CrO4 (20 ml from a 10 mCi stock solution) for 90 min at 37°C in 5% CO 2. The cells were then washed three times in PBS and cultured as described above. After 7 days of incubation the plates were spun at 200 g for 10 min, and 50 ml of the supernatant was removed for counting in a y-counter. The percentage of specific lysis was calculated as: (counts per minute in experiment-counts per minute spontaneous lysis)/ (counts per minute maximum release-counts per minute spontaneous release) × 100.
4. Results and Discussion Infection of M / M by HIV plays an important role in the pathogenesis of AIDS [1-4]. These cells are considered as a reservoir of the virus in the body especially in the late stages of HIV infection [1-2]. Macrophages act in vivo under the stimulation of different cytokines
200
TABLE 1 EFFECT OF HIV ON SYNCYTIA FORMATION AND VIABILITY IN M / M
'
Untreated
g.
Syncytia have been counted in M / M cultures at day 28 of culture. M / M cultures were fixed in methanol and stained with Giemsa; 300 cells were counted for each sample. Cell viability was assessed at the same time point by Trypan Blue exclusion directly on the culture dishes. In parentheses the percentage of dead syncytia is reported.
D
100" HIV
Treatment
0
i 7
0
1'4
i 21
None M-CSF
i 28
% of syncytia Not infected
HIV-infected
11.7 (24) 6.6 (18)
24.6 (42) 92.1 (71)
Days of culture
200"
100
M-CSF
g
g
#
x o
100"
50 o'*
7
.
_=
0
~
FT,
I I
I,I
7
14
21
28
Fig. 1. Effect of HIV infection on the viability of M-CSF-treated M / M . Data represent the arithmetic mean of 4 different experiments each carried out in triplicate with a S.D. less than 15%. The hollow squares represent the not-infected M / M , black squares represented HIV-infected M / M , black triangles represent the OKT4a-treated M / M . The empty bars in lower panel represent the effect of HIV-infection on the viability of M-CSF stimulated M / M cells. Cell viability was estimated by 51CrO4 release after T days of incubation. Data represent the mean of two experiments carried out in triplicate with the S.E.M. less than 10%.
HIV-infected M / M once the infection was already established reduced the rate of cell death (Fig. 1). Moreover, the blocking of the cell fusion, exerted by anti-CD4 treatment, led to the emergence of a chronically infected M / M population that went on to produce large amounts of virus (Fig. 2). In contrast, the viral output (measured as .concentration of p24) in cultures not treated with anti-CD4 decreased progressively (Fig.
2). We wondered whether the syncytia formation elicited by M-CSF could be due to an increase in the percentage of M / M expressing the CD4-receptor. We found that the percentage of CD4 positive cells was variable in unstimulated M / M with a mean of 51%, however, M-CSF treatment induced a marked increase in the number of CD4 positive cells: 95 vs. 51% (data not shown). Taken together these data seem to indicate that there is a correlation between CD4 expression and syncytia formation in M-CSF-treated HIV-infected M/M.
120
100 "
including M-CSF [16-19]. It should be noted that measurements performed on human blood have shown that endogenous M-CSF levels are in the range of 700-1000 U / m l [20]. Moreover, it has been reported that during experimentally induced bacterial infection in mice the blood levels of M-CSF may increase up to 1500 U / m l [21]. We have recently shown that M-CSF (1000 U / m l ) causes HIV to be cytopathic for M / M in vitro [22]. Accordingly, HIV infection affected the growth rate of M-CSF-treated M / M (Fig. 1). Data reported in Table 1 demonstrate that M-CSF-induced cell death in HIV-infected M / M cells (Fig. 1) is associated to massive syncytia formation. This effect is very likely mediated by the interaction between CD4 and HIV envelope glycoprotein gp120 [5]. Indeed, masking CD4 receptor with specific anti-CD4 antibody in M-CSF-stimulated
80" 60"
40"
20" 0
!
4
21
|
28
i
35
i
42
4
Time (days)
Fig. 2. HIV-p24 production in M - C S F - M / M treated or not with OKT4a. OKT4a was added to the cultures on day 16. The hollow squares represent the untreated HIV-infected M / M , black squares represent M-CSF treated HIV-infected M / M cells. Data are the mean of three separate experiments with a + S.E.M. less than 15%.
37
38
We investigated whether the M-CSF-associated death of HIV-infected macrophages could be caused by apoptosis. One characterizing feature of apoptotic cells is that they do not lyse, but shrink and fragment into smaller membrane-limited particles with a characteristic morphology: the apoptotic bodies. The stabilization of the apoptotic bodies has been associated to the activation of tTG [15]. Indeed, the extensive tTG-dependent cross-linking of intracellular proteins, that renders the apoptotic cells resistant even to treatment with detergents [14], seems to be a key event in preventing the lysis of apoptotic cells [15]. When we evaluated our macrophage cultures for tTG expression, we found that M-CSF-treated, HIV-infected syncytia markedly accumulated tTG protein (Fig. 3). In contrast, tTG protein was not found in mononuclear cells in both infected (Fig. 3A) or uninfected cultures (not shown). The expression of tTG in the syncytia was associated with the formation of cytoplasmic blebs which in some percentages contained remnants of condensed chromatin (Fig. 3D). These morphological and biochemical observations suggest that apoptosis may be triggered by M-CSF in HIV-infected human macrophages. To get further molecular evidence of apoptosis we analysed DNAfragmentation; however, under our experimental conditions, we did not find any evidence of DNA laddering (data not shown). A potential explanation could be that in our cultures the rate of cell death is too low to produce a detectable DNA ladder. Indeed, no more than 10-30% of HIV-infected cells died during a 7 days period (Fig. 2); furthermore it is well known that secondary necrosis takes place in the dead cells floating in culture fluid leading to the uncontrolled disruption of intracellular components including DNA [23]. Another explanation could be that these cells undergo apoptosis without digesting their DNA, as already described in hepatocytes [24] and epithelial cells [25]. It is well established that apoptotic cells do not lyse but fragment into membrane sealed apoptotic bodies [11], thus suggesting that apoptosis might have a protective role in reducing the spreading of infectious virions from dying infected cells. (Fig. 3). Interestingly, we found that in M-CSF-treated cultures, in which syncytial ceils express high levels of tTG (Fig. 3B,C), the ratio between intracellular and extracellular HIV-p24 antigen is much lower than in unstimulated cultures in
TABLE 2 EFFECT OF M-CSF ON THE RELEASE OF HIV p24 On day 20 and 27 supernatants were aspirated from the cultures and wells were replenished with fresh complete medium with the same concentration of cytokines. On day 21 and 28 wells were washed, supematants were stocked at - 7 0 ° C and cultures were exposed to a 200 /xl solution containing Triton X 1% for 30 min. After exposure samples were collected and kept at - 7 0 ° C . In parentheses the ratio between the amount of p24 protein detected in the culture medium vs. that found inside the cells is given. Data are the mean ±S.E.M. of three separate experiments. Treatment
Control M-CSF
Days of
HIV-p24 p g / m l / 1 0 0 0 cells
culture
Extracellular
Intracellular
Ratio
21 28 21 28
30 +_ 12 118 + 30 75+31 100 + 23
52 + 15 202 + 33 50+ 6 907 _+99
(0.58) (0.58) (0.50) (0.11)
which only a limited number of syncytia expressing tTG was detected (Table 2). An interesting possibility is whether the virions could be trapped inside the dying cell by a tTG-dependent cross-linking. In keeping with this hypothesis it has recently been shown that the viral transmembrane glycoprotein gp41 as well as gpl20 can very effectively act as in vitro substrates for tTG [26]. In this paper we demonstrate that M-CSF makes human macrophages susceptible to the HIV-induced cytopathic effect. Such effect is associated with the ability of M-CSF to stimulate the overexpression of the CD4 receptor on macrophages and is mediated by syncytia formation. In addition we present evidence suggesting a possible role for apoptosis in mediating HIV cytopathogenicity in macrophages.
Acknowledgements The authors would like to express their gratitude to Prof. F. Autuori and Prof. G. Rocchi for their stimulating discussion and encouragement, and to Prof. L. Fesus for kindly providing the tissue transglutaminase antibody. The work was supported by grants from "AIDS Project", by the Italian Ministry of Health, and Progetto Finalizzato "Prevenzione e Controllo dei Fattori di Malattia (FATMA), by the Italian National Research Council.
Fig. 3. tTG protein expression in multinucleated HIV-infected M / M . A. HIV infected, untreated M / M cells (day 28): unstained mononuclear cells with a positive syncytium (arrow heads); BAR = 25/xm. B, C. HIV-infected, M-CSF treated M / M (day 14 B, day 21 C); BAR = 25/xm. Note the progressive condensation of the syncytia which finally undergo fragmentation. D, F. Higher magnification of HIV-infected M-CSF treated M / M (day 28; BARS: D = 4 /~m; E, F = 10 /zm). The cytoplasm of the giant multinucleated cells is heavily stained with the anti-tTG antibody; the globular cell remnants (apoptotic bodies; note their shrunken cytoplasm and condensed chromatin) originated by the fragmentation of syncytia are indicated by the arrows.
39
References [1] Nicholson, J.K.A., Cross, G.D., Callaway, C.S. and McDougal, J.S. (1986) J. Immunol. 137, 323. [2] Crowe, S., Mills, J. and McGrath, M.S. (1987) AIDS Res. Hum. Retrovir. 3, 135. [3] Pantaleo, G., Graziosi, C., Denmarest, J.F., Butini, L., Montroni, M., Fox, C.H., Orenstein, J.M., Kotler, D.P. and Fauci, A.S. (1993) Nature 362, 355. [4] Embretson, J., Zupancic, M., Ribas, J.L., Burke, A., Racz, P., Tenner-Racz, K. and Haase, A.T. (1993) Nature 362, 359. [5] Lifson, J.D., Feinberg, M.B., Reyes, G.R., Rabbin, L., Banapours, B., Chakrabarti, S., Moss, B., Wong-Staal, F., Steimer, K.S. and Engleman, E.G. (1986) Nature 323, 725. [6] Gartner, S., Markovits, D.V.M., Markovitz, D.M., Betts, R.D. and Popovic, M. (1986) JAMA, 256, 2365. [7] Terai, C., Kornbluth, R.S., Pauza, C.D., Richman, D.D. and Carson, D.A. (1991) J. Clin. Invest., 87, 1710. [8] Groux, H., Torpier, G., Mont6, D., Mouton, Y., Capron, A. and Ameisen, J.C. (1992) J. Exp. Med. 175, 331. [9] Gougeon, M.L., Olivier, R., Garcia, S., Guetard, D., Dragie, T., Dauguet, C. and Montagneir, L. (1991) C.R. Acad. Sci. Paris. Ser. III Sci. Vie 312, 529. [10] Fesus, L., Davies, P.J.A. and Piacentini, M. (1991) Eur. J. Cell Biol. 56, 170. [11] Wyllie, A., Kerr, J.F.K. and Currie, A.R. (1980) Int. Rev. Cytol. 68, 251. [12] Piacentini, M., Fesus, L., Farrace, M.G., Ghibelli, L., Piredda, L. and Melino, G. (1991a) Eur. J. Cell Biol. 54, 246. [13] Piacentini, M., Autuori, F., Dini, L., Farrace, M.G., Ghibelli, L., Piredda, L. and Fesus, L. (1991b) Cell Tissue Res. 263, 227.
40
[14] Fesus, L., Thomazy, V., Autuori, F., Ceru, M.P., Tarcsa, E. and Piacentini, M. (1989) FEBS Lett. 245, 150. [15] Gentile, V., Thomazy, V., Piacentini, M., Fesus, L. and Davies, P.J.A. (1992) J. Cell Biol. 119, 463. [16] Chen, B.D.M., Clark, M. and Ciou, T. (1988) Blood, 4, 997. [17] Ampel, N.M., Wing, E.J., Waheed, A. and Shadduck, R.K. (1986) Cell. Immunol. 97, 334. [18] Grabstein, K.H., Urdal, D.L., Tushinski, R.J., Mochizuchi, D.F., Price, V.L., Cantrell, M.A., Gillis, S. and Conlon, P.J. (1986) Science 232, 506. [19] Young, D.A., Lowe, L.D. and Clark, C.S. (1990) J. Immunol. 145, 607. [20] Furosawa, S., Komatsu, H., Saito, K., Enokihara, H., Hirose, K. and Shishido, H. (1978) J. Clin. Med. 91, 377. [21] Wing, E.J., Barczynski, L.C., Waheed, A. and Shadduck, R.K. (1983) Infect. Immun. 49, 325. [22] Bergamini, A., Perno, C.F., Capozzi, M., Mannella, E., Salanitro, A., Calib, R. and Rocchi, G. (1992) J. Virol. Methods 40, 275. [23] Piacentini, M., Davies, P.J.A. and Fesus, L. (1994) in: Apoptosis: The molecular basis of apoptosis in disease (L.D. Tomei and F.O. Cope, Eds.) Cold Spring Harbor Laboratory Press, pp. 143-163. [24] Oberhammer, F., Fritsch, G., Pavelka, M., Froschl, G., Tiefenbacher, R., Purchio, T. and Schulte-Hermann, R. (1992) Toxicology Lett. 64, 701. [25] Oberhammer, F., Wilson, J.W., Dive, C., Morris, I.D., Hickman, J.A., Wakeling, A.E., Walker, P.R. and Sikorska, M. (1993) EMBO J. 12, 3679. [26] Mariniello, L., Esposito, C., Di Pierro, P., Cozzolino, A., Pucci, P. and Porta, R. (1993) Eur. J. Biocbem. 215, 99.
Macrophage-colony stimulating factor (M-CSF) stimulation induces cell death in HIV-infected human monocytes Alberto Bergamini a, Marcella Capozzi
a
and Mauro Piacentini b,.
a Department of Public Health and Cell Biology and b Department of Biology, University o.fRome "Tor Vergata", 00133 Rome, Italy (Received 13 December 1993; revision received 14 June 1994; accepted 15 June 1994) Key words: M-CSF, Monocyte-macrophage; Apoptosis; "Tissue" transglutaminase; HIV-gpl20; HIV-gp41
1. Summary We show here that HIV-infected monocyte-macrophages stimulated by macrophage-colony stimulating factor (M-CSF) undergo massive syncytia formation and die. The M-CSF-stimulated HIV-infected monocyte-macrophages (M/M) destroy themselves by blebbing out particles (resembling apoptotic bodies) which may contain condensed and marginated chromatin. The death of monocyte-macrophages is also characterized by the expression of "Tissue" Transglutaminase (fiG) which is one of the genes specifically expressed and activated in apoptising cells. Noteworthy, when the syncytia formation and consequently death is prevented, infected monocyte-macrophages remain viable and produce large amounts of virus for an extended period. The concentrations of M-CSF (1000 U/ml) used in this work are similar to those that stimulate macrophages in vivo. This suggests that HIV killing of M / M in the presence of M-CSF could lead, in vivo, to a greater than expected loss of immune cells and may contribute to explain the complex derangement of the immune function observed in HIV-infected patients.
2. Introduction HIV infects, in addition to T lymphocytes, other immunocompetent cells bearing on their surface the CD4 molecule, such as cells of the monocyte-macrophage lineage (M/M) [1-2]. These cells are a crucial target of the virus in vivo and account for the majority of HIV-infected cells in the body [3-4]. As a consequence of viral replication, M / M may fuse forming multinucleated giant cells (syncytia) [5]. This latter * Corresponding author: Mauro Piacentini, Dipartimento di Biologia, Via della Ricerca Scientifica 1, 00133 Rome, Italy. SSD1 0165-2478(94)001 15-8
phenomenon requires the expression on the HIV-envelope glycoprotein of 120 kDa (gpl20) and is mediated by gp-120-CD4 interaction [5-6]. Recently, the cytopathic effect of HIV has been connected to its ability to evoke in the target ceils the physiological program of cell death "apoptosis" [7-9]. Apoptosis is the silent (not associated to inflammation) physiological process of cell death leading to the controlled elimination of cells from tissues [10-11]. Independently from their origin cells undergoing apoptosis end in common, well characterized, morphological features suggesting the existence of a general pathway of physiological death regulated by specific genes [10-11]. Although the molecular mechanisms at the basis of apoptosis are not fully characterized, some biochemical features of the apoptotic cells have been established. The chromatin of apoptotic cells undergoes condensation which is often associated to its fragmentation at internucleosomal sites catalyzed by not yet established endonucleases [10-11]. The cytoplasm also becomes condensed and the cell fragments consequently to an active blebbing process that determines the formation of several membrane-bound cellular fragments (apoptotic bodies) [10-11]. A peculiar biochemical feature of these cell remnants is that their proteins are extensively cross-linked by "tissue" transglutaminases (EC 2.3.2.13; tTG), a Ca2+-dependent enzyme which is specifically induced and activated in the apoptotic cells [12-14]. The maturation and the activation of M / M are regulated by cytokine such as Macrophage-Colony Stimulating Factor (M-CSF) [16-18]. We have previously demonstrated that M-CSF (a cytokine that in vivo regulates the maturation and activation of M/M), induces, in vitro, the susceptibility of these cells to the HIV cytopathogenicity [19]. Since apoptosis has been suggested to be the cause of T-lymphocyte loss in HIV-in35
fected individuals [7-9], we investigated whether the M-CSF associated, HIV-induced cytopathogenicity in in vitro infected M / M could be mediated by apoptosis.
3. Materials and Methods
3.1. Cells M / M (from the peripheral blood of HIV-negative donors enriched for mononuclear cells (PBMC) by centrifugation over Ficoll Hypaque) were purified by two hour adherence of PBMC to Petri dishes as previously described [17]. Each experiment was performed using PBMC from a single donor. After purification, 5 × 10 4 M / M were seeded in fiat bottom, 96-wells plastic microtiter trays in 200 /xl of RPMI-1640 medium supplemented with 20% heat-inactivated fetal calf serum, 2 mM L-glutamine, 50 U / m l penicillin and 50 m g / m l streptomycin (complete medium). Ceils were cultured for 7 days with or without 1000 U / m l M-CSF (Medical System, Genova Italy).
3.2. H1V infection of M / M M / M were infected at day 7 of culture with 100 TCIDs0 of HIV-1 (HIV-1 strain HTLV-III Ba-L, which was a kind gift of Drs. S. Gartner, R.C. Gallo and M. Popovic, National Cancer Institute, N.I.H., Bethesda, MD). Corresponding mock-infected cultures were run in each experiment. Two hours after infection, M / M were extensively washed to remove excess virus and cultivated in complete medium at 37°C in a humidified atmosphere of 5% CO 2 in air with the same concentration of cytokine as before. Cells were washed and fed every 7 days. In selected experiments, a monoclonal antibody against the viral linking domain of CD4 (OKT4a) was added to the cultures at day 16. The total number of nuclei was assessed by counting at day 7, 14, 21 and 28 (cells were suspended in lysis buffer containing 10 mM KC1, 2 mM MgCI 2, 0.5% Triton X-100 in 10 mM Tris. HC1, pH 7.5) and the nuclei scored in a counting chamber using a phase contrast microscope (Laboflux D, Leitz). Cell viability was assessed on the culture dishes by Trypan Blue exclusion. The percentage of syncytia (multinucleated cells with more than three nuclei) was evaluated directly on the culture dishes after methanol fixation and Giemsa staining.
3.3. FACS analysis Expression of the CD4 protein was evaluated by flow cytometry using the Leu 3 a + b monoclonal anti36
body (Becton Dickinson). Paired isotype-specific control antibodies (IgG-1-FITC, Becton Dickinson) were run with each sample. More than 99% of the cells analyzed were LeuM3-positive monocytes. The percentage of antigen-positive cells was calculated by straight channel integration, with the integration channel set to that less than 1% of the isotype control cells appeared positive. The density of CD4 on the surface of the cells was calculated comparing the Leu 3 a + b fluorescence with appropriate fluorescein labeled isotype controls.
3.4. Immunohistochemistry Cells were cultured on slides, fixed in 2.5% paraformaldehyde and, after immunostaining with the monospecific anti-tTG antibody, counterstained with Mayer's haemalum. Immunohistochemical staining of M / M cells was performed using an affinity purified monospecific IgG raised in rabbits against human red blood cell soluble tTG (1 : 100) as primary antibody; the anti-tTG antibody was kindly supplied by Dr. L. Fesus (Department of Biochemistry, University Medical School of Debrecen, Debrecen, Hungary). Incubations with the primary antibody were carried out in a wet chamber overnight at 4°C. A biotinylated goat anti-rabbit IgG was used as second antibody, followed by a preformed streptavidin-horseradish peroxidase complex (Biogenex, USA). The reaction was developed using aminoethylcarbazole (AEC), (CRL, USA) as chromogen substrate and 0.01% H202. Endogenous peroxidase activity was blocked by methanol-H202.
3.5. Effect of HIV on M / M viability HIV- or mock-infected M / M were incubated at different time points (see Results) with 200 /zCi Na251CrO4 (20 ml from a 10 mCi stock solution) for 90 min at 37°C in 5% CO 2. The cells were then washed three times in PBS and cultured as described above. After 7 days of incubation the plates were spun at 200 g for 10 min, and 50 ml of the supernatant was removed for counting in a y-counter. The percentage of specific lysis was calculated as: (counts per minute in experiment-counts per minute spontaneous lysis)/ (counts per minute maximum release-counts per minute spontaneous release) × 100.
4. Results and Discussion Infection of M / M by HIV plays an important role in the pathogenesis of AIDS [1-4]. These cells are considered as a reservoir of the virus in the body especially in the late stages of HIV infection [1-2]. Macrophages act in vivo under the stimulation of different cytokines
200
TABLE 1 EFFECT OF HIV ON SYNCYTIA FORMATION AND VIABILITY IN M / M
'
Untreated
g.
Syncytia have been counted in M / M cultures at day 28 of culture. M / M cultures were fixed in methanol and stained with Giemsa; 300 cells were counted for each sample. Cell viability was assessed at the same time point by Trypan Blue exclusion directly on the culture dishes. In parentheses the percentage of dead syncytia is reported.
D
100" HIV
Treatment
0
i 7
0
1'4
i 21
None M-CSF
i 28
% of syncytia Not infected
HIV-infected
11.7 (24) 6.6 (18)
24.6 (42) 92.1 (71)
Days of culture
200"
100
M-CSF
g
g
#
x o
100"
50 o'*
7
.
_=
0
~
FT,
I I
I,I
7
14
21
28
Fig. 1. Effect of HIV infection on the viability of M-CSF-treated M / M . Data represent the arithmetic mean of 4 different experiments each carried out in triplicate with a S.D. less than 15%. The hollow squares represent the not-infected M / M , black squares represented HIV-infected M / M , black triangles represent the OKT4a-treated M / M . The empty bars in lower panel represent the effect of HIV-infection on the viability of M-CSF stimulated M / M cells. Cell viability was estimated by 51CrO4 release after T days of incubation. Data represent the mean of two experiments carried out in triplicate with the S.E.M. less than 10%.
HIV-infected M / M once the infection was already established reduced the rate of cell death (Fig. 1). Moreover, the blocking of the cell fusion, exerted by anti-CD4 treatment, led to the emergence of a chronically infected M / M population that went on to produce large amounts of virus (Fig. 2). In contrast, the viral output (measured as .concentration of p24) in cultures not treated with anti-CD4 decreased progressively (Fig.
2). We wondered whether the syncytia formation elicited by M-CSF could be due to an increase in the percentage of M / M expressing the CD4-receptor. We found that the percentage of CD4 positive cells was variable in unstimulated M / M with a mean of 51%, however, M-CSF treatment induced a marked increase in the number of CD4 positive cells: 95 vs. 51% (data not shown). Taken together these data seem to indicate that there is a correlation between CD4 expression and syncytia formation in M-CSF-treated HIV-infected M/M.
120
100 "
including M-CSF [16-19]. It should be noted that measurements performed on human blood have shown that endogenous M-CSF levels are in the range of 700-1000 U / m l [20]. Moreover, it has been reported that during experimentally induced bacterial infection in mice the blood levels of M-CSF may increase up to 1500 U / m l [21]. We have recently shown that M-CSF (1000 U / m l ) causes HIV to be cytopathic for M / M in vitro [22]. Accordingly, HIV infection affected the growth rate of M-CSF-treated M / M (Fig. 1). Data reported in Table 1 demonstrate that M-CSF-induced cell death in HIV-infected M / M cells (Fig. 1) is associated to massive syncytia formation. This effect is very likely mediated by the interaction between CD4 and HIV envelope glycoprotein gp120 [5]. Indeed, masking CD4 receptor with specific anti-CD4 antibody in M-CSF-stimulated
80" 60"
40"
20" 0
!
4
21
|
28
i
35
i
42
4
Time (days)
Fig. 2. HIV-p24 production in M - C S F - M / M treated or not with OKT4a. OKT4a was added to the cultures on day 16. The hollow squares represent the untreated HIV-infected M / M , black squares represent M-CSF treated HIV-infected M / M cells. Data are the mean of three separate experiments with a + S.E.M. less than 15%.
37
38
We investigated whether the M-CSF-associated death of HIV-infected macrophages could be caused by apoptosis. One characterizing feature of apoptotic cells is that they do not lyse, but shrink and fragment into smaller membrane-limited particles with a characteristic morphology: the apoptotic bodies. The stabilization of the apoptotic bodies has been associated to the activation of tTG [15]. Indeed, the extensive tTG-dependent cross-linking of intracellular proteins, that renders the apoptotic cells resistant even to treatment with detergents [14], seems to be a key event in preventing the lysis of apoptotic cells [15]. When we evaluated our macrophage cultures for tTG expression, we found that M-CSF-treated, HIV-infected syncytia markedly accumulated tTG protein (Fig. 3). In contrast, tTG protein was not found in mononuclear cells in both infected (Fig. 3A) or uninfected cultures (not shown). The expression of tTG in the syncytia was associated with the formation of cytoplasmic blebs which in some percentages contained remnants of condensed chromatin (Fig. 3D). These morphological and biochemical observations suggest that apoptosis may be triggered by M-CSF in HIV-infected human macrophages. To get further molecular evidence of apoptosis we analysed DNAfragmentation; however, under our experimental conditions, we did not find any evidence of DNA laddering (data not shown). A potential explanation could be that in our cultures the rate of cell death is too low to produce a detectable DNA ladder. Indeed, no more than 10-30% of HIV-infected cells died during a 7 days period (Fig. 2); furthermore it is well known that secondary necrosis takes place in the dead cells floating in culture fluid leading to the uncontrolled disruption of intracellular components including DNA [23]. Another explanation could be that these cells undergo apoptosis without digesting their DNA, as already described in hepatocytes [24] and epithelial cells [25]. It is well established that apoptotic cells do not lyse but fragment into membrane sealed apoptotic bodies [11], thus suggesting that apoptosis might have a protective role in reducing the spreading of infectious virions from dying infected cells. (Fig. 3). Interestingly, we found that in M-CSF-treated cultures, in which syncytial ceils express high levels of tTG (Fig. 3B,C), the ratio between intracellular and extracellular HIV-p24 antigen is much lower than in unstimulated cultures in
TABLE 2 EFFECT OF M-CSF ON THE RELEASE OF HIV p24 On day 20 and 27 supernatants were aspirated from the cultures and wells were replenished with fresh complete medium with the same concentration of cytokines. On day 21 and 28 wells were washed, supematants were stocked at - 7 0 ° C and cultures were exposed to a 200 /xl solution containing Triton X 1% for 30 min. After exposure samples were collected and kept at - 7 0 ° C . In parentheses the ratio between the amount of p24 protein detected in the culture medium vs. that found inside the cells is given. Data are the mean ±S.E.M. of three separate experiments. Treatment
Control M-CSF
Days of
HIV-p24 p g / m l / 1 0 0 0 cells
culture
Extracellular
Intracellular
Ratio
21 28 21 28
30 +_ 12 118 + 30 75+31 100 + 23
52 + 15 202 + 33 50+ 6 907 _+99
(0.58) (0.58) (0.50) (0.11)
which only a limited number of syncytia expressing tTG was detected (Table 2). An interesting possibility is whether the virions could be trapped inside the dying cell by a tTG-dependent cross-linking. In keeping with this hypothesis it has recently been shown that the viral transmembrane glycoprotein gp41 as well as gpl20 can very effectively act as in vitro substrates for tTG [26]. In this paper we demonstrate that M-CSF makes human macrophages susceptible to the HIV-induced cytopathic effect. Such effect is associated with the ability of M-CSF to stimulate the overexpression of the CD4 receptor on macrophages and is mediated by syncytia formation. In addition we present evidence suggesting a possible role for apoptosis in mediating HIV cytopathogenicity in macrophages.
Acknowledgements The authors would like to express their gratitude to Prof. F. Autuori and Prof. G. Rocchi for their stimulating discussion and encouragement, and to Prof. L. Fesus for kindly providing the tissue transglutaminase antibody. The work was supported by grants from "AIDS Project", by the Italian Ministry of Health, and Progetto Finalizzato "Prevenzione e Controllo dei Fattori di Malattia (FATMA), by the Italian National Research Council.
Fig. 3. tTG protein expression in multinucleated HIV-infected M / M . A. HIV infected, untreated M / M cells (day 28): unstained mononuclear cells with a positive syncytium (arrow heads); BAR = 25/xm. B, C. HIV-infected, M-CSF treated M / M (day 14 B, day 21 C); BAR = 25/xm. Note the progressive condensation of the syncytia which finally undergo fragmentation. D, F. Higher magnification of HIV-infected M-CSF treated M / M (day 28; BARS: D = 4 /~m; E, F = 10 /zm). The cytoplasm of the giant multinucleated cells is heavily stained with the anti-tTG antibody; the globular cell remnants (apoptotic bodies; note their shrunken cytoplasm and condensed chromatin) originated by the fragmentation of syncytia are indicated by the arrows.
39
References [1] Nicholson, J.K.A., Cross, G.D., Callaway, C.S. and McDougal, J.S. (1986) J. Immunol. 137, 323. [2] Crowe, S., Mills, J. and McGrath, M.S. (1987) AIDS Res. Hum. Retrovir. 3, 135. [3] Pantaleo, G., Graziosi, C., Denmarest, J.F., Butini, L., Montroni, M., Fox, C.H., Orenstein, J.M., Kotler, D.P. and Fauci, A.S. (1993) Nature 362, 355. [4] Embretson, J., Zupancic, M., Ribas, J.L., Burke, A., Racz, P., Tenner-Racz, K. and Haase, A.T. (1993) Nature 362, 359. [5] Lifson, J.D., Feinberg, M.B., Reyes, G.R., Rabbin, L., Banapours, B., Chakrabarti, S., Moss, B., Wong-Staal, F., Steimer, K.S. and Engleman, E.G. (1986) Nature 323, 725. [6] Gartner, S., Markovits, D.V.M., Markovitz, D.M., Betts, R.D. and Popovic, M. (1986) JAMA, 256, 2365. [7] Terai, C., Kornbluth, R.S., Pauza, C.D., Richman, D.D. and Carson, D.A. (1991) J. Clin. Invest., 87, 1710. [8] Groux, H., Torpier, G., Mont6, D., Mouton, Y., Capron, A. and Ameisen, J.C. (1992) J. Exp. Med. 175, 331. [9] Gougeon, M.L., Olivier, R., Garcia, S., Guetard, D., Dragie, T., Dauguet, C. and Montagneir, L. (1991) C.R. Acad. Sci. Paris. Ser. III Sci. Vie 312, 529. [10] Fesus, L., Davies, P.J.A. and Piacentini, M. (1991) Eur. J. Cell Biol. 56, 170. [11] Wyllie, A., Kerr, J.F.K. and Currie, A.R. (1980) Int. Rev. Cytol. 68, 251. [12] Piacentini, M., Fesus, L., Farrace, M.G., Ghibelli, L., Piredda, L. and Melino, G. (1991a) Eur. J. Cell Biol. 54, 246. [13] Piacentini, M., Autuori, F., Dini, L., Farrace, M.G., Ghibelli, L., Piredda, L. and Fesus, L. (1991b) Cell Tissue Res. 263, 227.
40
[14] Fesus, L., Thomazy, V., Autuori, F., Ceru, M.P., Tarcsa, E. and Piacentini, M. (1989) FEBS Lett. 245, 150. [15] Gentile, V., Thomazy, V., Piacentini, M., Fesus, L. and Davies, P.J.A. (1992) J. Cell Biol. 119, 463. [16] Chen, B.D.M., Clark, M. and Ciou, T. (1988) Blood, 4, 997. [17] Ampel, N.M., Wing, E.J., Waheed, A. and Shadduck, R.K. (1986) Cell. Immunol. 97, 334. [18] Grabstein, K.H., Urdal, D.L., Tushinski, R.J., Mochizuchi, D.F., Price, V.L., Cantrell, M.A., Gillis, S. and Conlon, P.J. (1986) Science 232, 506. [19] Young, D.A., Lowe, L.D. and Clark, C.S. (1990) J. Immunol. 145, 607. [20] Furosawa, S., Komatsu, H., Saito, K., Enokihara, H., Hirose, K. and Shishido, H. (1978) J. Clin. Med. 91, 377. [21] Wing, E.J., Barczynski, L.C., Waheed, A. and Shadduck, R.K. (1983) Infect. Immun. 49, 325. [22] Bergamini, A., Perno, C.F., Capozzi, M., Mannella, E., Salanitro, A., Calib, R. and Rocchi, G. (1992) J. Virol. Methods 40, 275. [23] Piacentini, M., Davies, P.J.A. and Fesus, L. (1994) in: Apoptosis: The molecular basis of apoptosis in disease (L.D. Tomei and F.O. Cope, Eds.) Cold Spring Harbor Laboratory Press, pp. 143-163. [24] Oberhammer, F., Fritsch, G., Pavelka, M., Froschl, G., Tiefenbacher, R., Purchio, T. and Schulte-Hermann, R. (1992) Toxicology Lett. 64, 701. [25] Oberhammer, F., Wilson, J.W., Dive, C., Morris, I.D., Hickman, J.A., Wakeling, A.E., Walker, P.R. and Sikorska, M. (1993) EMBO J. 12, 3679. [26] Mariniello, L., Esposito, C., Di Pierro, P., Cozzolino, A., Pucci, P. and Porta, R. (1993) Eur. J. Biocbem. 215, 99.