819

Infectious Respiratory Syncytial Virus (RSV) Effectively Inhibits the Proliferative T Cell Response to Inactivated RSV In Vitro F. Maxine Preston, Paul L. Beier, and John H. Pope

Sir Albert Sakzewski Virus Research Centre. Royal Children's Hospital. Brisbane. Australia

Respiratory syncytial virus (RSV) is the major viral respiratory pathogen responsible for severe bronchiolitis and pneumonia in infants worldwide [I, 2]. Reinfection is common, in some instances occurring in infants within weeks after recovery from primary infection [3]. However, the severity of disease is decreased by the third infection in childhood [4, 5], and multiple reinfections appear to induce a more substantial immunity [4, 6]. The role of the immune system in protection against RSV infection and reinfection is not well understood. There is some correlation between serum antibody titer and protection against RSV infection in cotton rats [7], mice [8] and humans [9], but reinfection in humans has been documented in the presence of anti-RSV antibody [5, 9]. The occurrence of persistent RSV infection in children with depressed cellmediated immunity [ 10] suggests an important role for cellular immunity in recovery from RSV infection. RSV-specific cytolytic T lymphocytes have been demonstrated in humans [11-15] and mice [16], and their role in protection [16] has also been described. It has been postulated that the well known reinfection with RSV may be in part due to inhibitory effects of the virus itself on the generation of an RSV-specific cellular immune response. Reinfection by RSV may then be the consequence of failure to mount an effective primary or secondary (mem-

Received 18 September 1991; revised 5 December 1991. All donors gave informed consent, and the research proposal was approved by the Ethics Committee ofthe Royal Children's Hospital, Brisbane. Grant support: National Health and Medical Research Council, Canberra, Australia. Reprints or correspondence: Dr. F. Maxine Preston. Sir Albert Sakzewski Virus Research Centre, Royal Children's Hospital, Herston Road, Brisbane 4029, Queensland, Australia.

The Journal of Infectious Diseases 1992;165:819-25 © 1992 by The University of Chicago. All rights reserved. 0022-1899/92/6505-0005$01.00

ory) anti-RSV immune response. The current study was therefore undertaken to examine this possibility. RSV has been shown to infect human mononuclear cells (MNC) in vitro [17, 18] and in vivo [I 7], to inhibit the mitogen-induced proliferation of human MNC [19], and to induce the production of an inhibitor of interleukin (IL )-1 in monocytes/macrophages [20]. In addition, MNC have been shown to proliferate in response to inactivated but not infectious RSV [21]. In this study we have confirmed the inhibitory effect of RSV on mitogen-stimulated MNC and demonstrated a similar depression of the CMI response to EpsteinBarr virus (EBV). Most importantly, we have demonstrated a clear role for RSV in the inhibition in vitro of an anti-RSV T cell response.

Materials and Methods Viral stocks. RSV subgroup A included strains Long (provided by G. A. Tannock, University of Newcastle, Australia) and 8812 (isolated in this laboratory and typed by immunofluorescence with monoclonal antibody [MAb] provided by J. C. Hierholzer, Centers for Disease Control, Atlanta). Subgroup B was represented by strain 18537 (also provided by J. C. Hierholzer). Virus was grown in HEp-2 cells and pools were prepared essentially as described [22], filtered through a 650-nm filter, and stored at .,.-70°C. Titers were determined by assay in HEp-2 cells, and the TCID so per milliliter was calculated by the Karber method. Mock inoculum was prepared similarly from uninfected HEp-2 cultures. We took particular care to ensure that neither viral stocks nor cell lines were contaminated with mycoplasma, as it is well known that some mycoplasmas are highly inhibitory for T cells [23]. Therefore, HEp-2 cells and viral pools were routinely tested for the presence of mycoplasma by nucleic acid hybridization (Gen-Probe, San Diego, CA) and proved negative. Inactivated RSV was prepared by treatment with {j-propiolactone (final concentration 1:4000; Sigma, St. Louis) for 18 h at

Downloaded from http://jid.oxfordjournals.org/ at New York University on September 1, 2015

The effect of respiratory syncytial virus (RSV) on the cellular immune response of human mononuclear cells in vitro was examined. Inhibition by RSV of the lymphocyte response to phytohemagglutinin in vitro was confirmed using cells from human umbilical cord blood. In addition, RSV significantly inhibited both the proliferative and T cell colony responses of human mononuclear cells to Epstein-Barr virus. An RSV-specificcellular immune response was induced in vitro by stimulation of mononuclear cells from RSV-seropositive donors with p-propiolactone-inactivated RSV. This RSV-specific response was significantly inhibited by infectious RSV itself, and the inhibition was mediated by an extracellular factor produced by RSV-infected mononuclear cells. A similar inhibition in vivo of the RSV-induced cellular immune response may contribute significantly to delayed recovery from primary infection and to reduced resistance to subsequent infections.

820

Preston et al.

colonies counted on a dissecting microscope using a 5- X 5-mm grid. Only tight clusters containing ~20 cells were counted as colonies. The results are expressed as the percentage of colony formation relative to the mock-infected culture and represent the mean of four dishes. When required, EBV-specific colonies were harvested and maintained in medium containing recombinant IL-2 (rIL-2) by twice weekly stimulation with irradiated autologous LCL. Transfer of infected culture supernatant. Culture fluid was harvested from MNC incubated with inactivated RSV for 6 days in the presence of infectious virus or mock inoculum. The cells were removed by centrifugation and the supernatant fluid filtered through a 100-nm filter. The filtered supernatant contained no detectable infectious virus as assessed by inoculation of HEp-2 cells. Fresh autologous MNC (10 6/ml in 24-well plates) stimulated with inactivated RSV were immediately exposed to the supernatant (1 ml/well in 24 wells) for 6 days, and T cell colonies were prepared and counted. The effect of the supernatant on colony formation was then directly compared with that seen when infectious virus was used as the inoculum. Flow cytometry. PHA-stimulated cord MNC exposed to RSV or mock inoculum were stained with anti-RSV MAb followed by phycoerythrin-conjugated F (ab'), anti-mouse IgG (Jackson Laboratories, West Grove, PA). The cell sample was divided and further analyzed for expression of CD antigens by staining with fluorescein-conjugated anti-Leu-3a (CD4), antiLeu-2a (CD8), anti-Leu-l Z (CD 19), anti-Leu-l l (CD 16), or anti-Leu-M3 (CD 14) (Becton-Dickinson, Mountain View, CA). The cells were examined for the expression of dual-color immunofluorescence using an Epics flow cytometer (Coulter Electronics, Brookvale, Australia). Colony-derived CD4+ and CD8+ T cell lines (derived from EBV-specific T cell colonies described above) inoculated with RSV were also examined for expression of viral antigen by staining with anti-RSV MAb and analysis by flow cytometry. Statistics. Data were analyzed by Student's t test.

Results Inhibition of the PHA response of cord MNC by RSV. Exposure to RSV (3 days) resulted in a significant reduction in the PHA-induced proliferation of cord MNC (P < .02 for Long, P< .01 for 8812 and 18537) relative to MNC exposed to mock inoculum (figure 1). The decrease, however, was not accompanied by a loss of cell viability as assessed by trypan blue exclusion. The cells from the RSV-inhibited cultures were also examined for the expression of viral antigen by dual-color flow cytometry. The CDI4+ macrophage population demonstrated the highest level of RSV infection (table I) with> 50% of cells infected in some experiments. No infection ofB cells (CD 19+) or NK cells (CD 16+) was detected and negligible viral expression (~I %) was observed in the CD8+ T cell population. Although the CD4+ T cell is the predominant cell in the PHA-stimulated culture, relatively few were found to express viral antigen (table I). The infectivity of the virus strains varied; the A subgroup gave consistently higher levels of viral antigen-positive

Downloaded from http://jid.oxfordjournals.org/ at New York University on September 1, 2015

4°C. The viral suspension was then incubated at 37°C for 2 h to completely hydrolyse the {j-propiolactone [24]. Complete inactivation was demonstrated by the failure of the viral suspension to infect HEp-2 cells. RSV and inactivated viral pools were used at MOl = 1 in all experiments described. Preparation ofhuman MNC. MNC were prepared from umbilical cord blood and normal healthy adult volunteers by sedimentation on Ficoll-Paque (Pharmacia, Uppsala, Sweden) [25]. Due to excessive red cell contamination, cord MNC were subjected to a further Ficoll-Paque separation. No significant change in the relative proportions of the MNC subsets (CD4+ and CD8+ T cells, CD 19+ B cells, CD 16+ natural killer cells, and CD 14+ monocytes) was observed after the second centrifugation step, as assessed by flow cytometry. Cord blood and adult MNC were cultured at 106/ml in RPMI 1640 medium containing 2 mM L-glutamine and 10% or 20% fetal calf serum (10 or 20FCI 1640). All cultures were maintained at 37°C in an atmosphere of 5% CO 2 in air. MNC proliferation assay. Cord MNC were incubated in 96well microtiter plates in I OFCI 1640 containing 2 ~l/ml phytohaemagglutinin-P (PHA; Difco Laboratories, Detroit) and were inoculated with RSV (Long, 8812, and 18537) or mock inoculum. The cultures were incubated for 72 h and pulse-labeled with [3H]thymidine (eH]TdR; Amersham, Sydney, Australia) 8 h before harvesting; tritium incorporation was measured in a liquid scintillation counter. Experiments were done in triplicate, and the results are presented as mean ± SE of multiple experiments. MNC from adult EBV-seropositive donors were stimulated with an irradiated (8000 rad) EBV-transformed autologous lymphoblastoid cell line (LCL; 105/ml) in 96-well plates in 20FCI 1640 to reactivate EBV-specific memory T cells. The cultures were immediately inoculated with RSV (Long, 8812, and 18537) or mock inoculum, and cell proliferation was assessed at 72 h by [3H]TdR uptake. Similarly, the proliferation of MNC from RSV-seropositive adult donors to {j-propiolactone-inactivated RSV (8812) was assessed in the presence of infectious RSV. The stimulated cultures were immediately inoculated with RSV (8812), and cell proliferation was assessed by eH]TdR uptake at 6 days. Parallel cultures, both EBV- and RSV-stimulated, were also examined for T cell colony formation. Colony formation. The ability of MNC stimulated with either EBV or inactivated RSV antigens to form colonies in agarose [26] was assessed in the presence of infectious RSV. MNC, at a concentration of 106/ml in 20FC/1640 in 24-well plates, were stimulated with either irradiated autologous LCL or inactivated RSV and then immediately inoculated with RSV (Long, 8812, or 18537) or mock inoculum. At 3 (EBV-stimulated) and 6 (RSV-stimulated) days, the cultures were harvested and the cells resuspended in 20FCI 1640 containing 0.34% agarose (Sea Plaque; FMC Bioproducts, Rockland, ME) and 20% culture supernatant ofMLA-144, a gibbon ape cell line that constitutively secretes IL-2 and several other cytokines [27]. The cell suspension was then poured into 35-mm Petri dishes (10 5 cells/Z ml) and quick set by use ofa cold table (Lindner and May, Brisbane, Australia). The dishes were incubated for 6 more days and the

lID 1992;165 (May)

lID 1992; 165 (May)

ttJ

I

o

RSV Inhibition of Cellular Immunity

140

..-4

H

120

=-f3

100 80 80

40 20 Control

Long

8812

18537

CD4+ and CDI4+ cells than did the B subgroup. No fluorescence was observed in the mock inoculum control culture. EBV-specific CD4+ and CD8+ T cell colony-derived lines (induced and expanded in agarose as described above) were also exposed to RSV (2 days), but no significant viral antigen fluorescence « 1%) was detected by flow cytometry. Inhibition ofthe EB V-specific T cellresponse. Of the three RSV strains tested, only the Long strain significantly inhibited the proliferative response of adult MNC to EBV-transformed LCL (P < .05 relative to the EBV-stimulated MNC exposed to mock inoculum; figure 2A). In contrast, all three strains markedly inhibited (>60%) the ability of the EBVspecific T cells to form colonies in agarose (P < .001 for Long and 8812; P < .01 for 18537) relative to mock inoculum control cultures (figure 2B). The inhibition of colony formation in the presence of RSV was reflected not only in the number but in the size and viability of colonies as assessed microscopically. RSV inhibited the RSV-specificresponse. In two initial experiments, MNC from adult RSV-seropositive donors that were stimulated with {j-propiolactone-inactivated RSV exhibited a proliferative response 2.1-3.7 times higher than that of the control cultures stimulated with similarly inactivated mock inoculum. This indicated that part of the response was specifically generated by RSV antigens. A further series of five experiments clearly showed that the response to inactivated RSV was markedly inhibited (55% compared with the uninfected control) by immediate exposure to infectious virus (figure 3A). The results were not statistically significant due mainly to the large variation observed within the control group. This variation in proliferative response may be attributed to individual donor differences in RSV memory T cell precursor cell frequencies. The inhibitory effect of RSV was consistently observed with the subgroup A isolate 8812 and was later confirmed in two experiments for the subgroup B virus 18537 (data not shown).

To examine the effect of RSV on colony formation, we first assessed the ability of MNC stimulated by inactivated RSV to form colonies in agarose. An optimal experimental protocol was developed whereby MNC were stimulated with inactivated RSV for 6 days and then allowed to form colonies in agarose for a further 6 days. In five experiments, there was a significantly higher production of T cell colonies in response to stimulation with the inactivated RSV preparation than with the similarly inactivated mock inoculum (P < .01), indicating again that RSV specifically contributed to the generation of the response in seropositive donors. Further studies of the specificity of induction by testing the response of RSV-seronegative adult donors were precluded by the ubiquity ofRSV, and MNC from cord blood were unsuitable as they are so distinct from adult MNC in their immune responsiveness. On harvest and expansion in culture, colonies consisted predominantly ofCD4+ T cells as assessed by flow cytometry. It was significant that the addition ofinfectious RSV at the time of stimulation with inactivated RSV resulted in a dramatic decrease in the number of colonies (P < .001) compared with parallel cultures exposed to mock inoculum (figure 3B), again accompanied by a reduction in viability. The observed inhibition of colony formation by RSV was clearly dependent on the infectivity of the viral inoculum. Heat inactivation (56°C for 30 min), filtration (100 nm) (figure 4A), and neutralizing MAb (pool of 133-1H, 131-2G, and 130-12H from J. C. Hierholzer) (figure 4B) all resulted in almost complete restoration of colony formation, while the control MAb did not. The addition of rIL-I a, - I B, - 3, -4, or -6 (at 5, 5, 10, 30, and 100 units/ml, respectively; Genzyme, Boston) and rIL-2 (40 units/ml; Boehringer Mannheim, Mannheim, Germany) at the time of stimulation, either singly or combined, had no significant effects on the RSV-induced inhibition of colony formation. Inhibition can be transferred by culturefluid. Adult MNC were stimulated with an inactivated preparation ofRSV and immediately inoculated with infectious virus or mock inoculum. At 6 days, the cell- and virus-free supernatant was then

Table 1. Proportion ofCD4+ and CD14+ mononuclear cells expressing respiratory syncytial virus (RSV) antigen. RSV strain Phenotype

Long*

8812*

18537 t

8.3 ± 2.0 26.4 ± 6.0

5.6 ± 0.3 38.8 ± 6.2

2.2 ± 0.4 18.4 ± 6.2

NOTE. Mitogen-stimulated cord blood mononuclear cells were inoculated with RSV and examined for expression ofviral antigen by flow cytometry. Data are mean percentage ± SE of five experiments. * Subgroup A. t Subgroup B.

Downloaded from http://jid.oxfordjournals.org/ at New York University on September 1, 2015

Figure 1. Effect ofrespiratory syncytial virus (RSV) on mitogeninduced blast response of cord blood mononuclear cells (MNC) inoculated with RSV (Long, 8812, or 18537) or mock inoculum (control). Results are mean and SE of five experiments based on uptake of [3H]thymidine ([3H]TdR). CPM, counts per minute.

821

822

Preston et al.

B

A

M

~ 100

virus (RSV) on reactivation of Epstein-Barr virus (EBV)-specific T cells. Adult mononuclear cells were stimulated with autologous irradiated EBV-transformed lymphoblastoid cell line and immediately inoculated with RSV (Long, 8812, or 18537) or mock inoculum (control). Proliferation based on (A) uptake ofeH]thymidine ([3H]TdR) and (B) colony formation were measured. Colony formation is expressed as percentage relative to control. Results are mean and SE of five experiments. CPM, counts per minute.

c:I o

>

Infectious respiratory syncytial virus (RSV) effectively inhibits the proliferative T cell response to inactivated RSV in vitro.

The effect of respiratory syncytial virus (RSV) on the cellular immune response of human mononuclear cells in vitro was examined. Inhibition by RSV of...
652KB Sizes 0 Downloads 0 Views