Persistent Ross River Virus Infection of Murine ... - Science Direct

47 downloads 72374 Views 445KB Size Report
... Division of Science and Design, University of Canberra, Canberra 2601, Australia ...... ing Adobe Photoshop 6.0.1 (Adobe Systems Inc.) and quantitated using Image Guage V3.3 (Fuji PhotoFilm Co. Ltd. & Kohshin Graphic Systems). Briefly ...
Virology 301, 281–292 (2002) doi:10.1006/viro.2002.1587

Persistent Ross River Virus Infection of Murine Macrophages: An in Vitro Model for the Study of Viral Relapse and Immune Modulation during Long-Term Infection Samantha J. R. Way,* ,† Brett A. Lidbury,† and Joanne L. Banyer* ,1 *Division of Immunology and Cell Biology, John Curtin School of Medical Research, Australian National University, Canberra 2601, Australia; and †Gadi Research Center for Health and Medical Sciences, Division of Science and Design, University of Canberra, Canberra 2601, Australia Received March 4, 2002; returned to author for revision April 18, 2002; accepted May 24, 2002 A clinical feature of Ross River virus disease (RRVD) is the periodic relapse of symptoms months after the initial onset of disease. The underlying mechanisms responsible for this relapse have not been determined. In a long-term (148 days) in vitro study of persistently infected murine macrophages we established that RRV infection periodically fell to undetectable biological levels that required genetic detection. However, the virus concentration spontaneously relapsed to biologically detectable levels that corresponded with enhanced viral mRNA expression, cellular detachment, and cytopathic effect. By altering the cell culture conditions we found that relapse could also be induced. We propose that the periodic relapse of symptoms in RRVD may be associated with spontaneous or stress-induced increases in RRV within persistently infected macrophages. This study also established that RRV enhanced macrophage phagocytic activity and dysregulated the immunoregulatory molecules CD80, IFN-␥, and TNF-␣ that may facilitate persistence of RRV and avoidance of immune responses. © 2002 Elsevier Science (USA) Key Words: Ross River virus; macrophage; persistence; immunity; phagocytosis.

murine model that was characterized by hind-leg paresis/paralysis and a corresponding massive infiltrate of F4/80 ⫹ macrophages into the gastrocnemus muscle of the hind-leg post subcutaneous infection (Lidbury et al., 2000). In this model disease symptoms were completely ameliorated in RRV-infected mice by pretreatment with macrophage toxic agents. Furthermore, another study of a murine macrophage cell line, RAW 264.7, showed an increased expression of the macrophage monocyte chemoattractant protein-1 (MCP-1) in response to RRV infection (Mateo et al., 2000), providing additional evidence that the recruitment of macrophages to the site of infection is a primary factor in the disease. In general, disease symptoms diminish over time with relapse of the symptoms sporadically occurring sometimes years after disease onset (Hawkes et al., 1985). The underlying mechanism responsible for this relapse has not been established. It has been suggested that the virus may persist for extended periods of time, as serological findings have identified a delayed isotype switch from IgM to occur in EPA patients (Mackenzie and Smith, 1996), which may be synonymous with persistent antigenic challenge by the virus. Until recently the majority of human clinical studies have only detected the virus serologically for the first week after disease onset (Fraser et al., 1981). However, a more sensitive molecular RTPCR approach has now been successful in detecting the virus in patient synovial tissue samples up to 5 weeks after the onset of symptoms (Soden et al., 2000). Also, an in vitro study (Linn et al., 1996) detected RRV in 5–10% of

INTRODUCTION Ross River virus (RRV) is an enveloped positive-sense RNA alphavirus in the family Togaviridae (Strauss and Strauss, 1994). Transmission is via a horizontal cycle of mosquitoes, marsupials, and humans, with marsupials maintaining the virus in the environment. RRV is found predominantly in Australia, the Solomon Islands, and Papua New Guinea and is the major etiological agent of epidemic polyarthritis (EPA) in Australia (Mackenzie and Smith, 1996). EPA is characterized by a predominance of arthritis and arthralgia, accompanied by myalgia, fever, rash, and fatigue (Flexman et al., 1998). Investigation into the etiology of this disease has established that the rash, usually occurring first, is cleared rapidly by a RRV-specific CD8 T lymphocyte cell-mediated immune response (CMI) (Fraser et al., 1983). Development of arthritis in this disease, however, is thought to result from an inability of the CMI response to completely clear the virus. Instead a local nonspecific immune response predominates in arthritis involving a massive influx of mononuclear macrophages (Clarris et al., 1975; Fraser et al., 1981; Hazelton et al., 1985; Soden et al., 2000). The importance of macrophages in RRV-induced disease has been supported by detailed analyses of a

1 To whom correspondence and reprint requests should be addressed at John Curtin School of Medical Research, P.O. Box 334, Canberra, A.C.T., Australia 2601. Fax: ⫹61 2 61252595. E-mail: [email protected].

281

0042-6822/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.

282

WAY, LIDBURY, AND BANYER

FIG. 1. Immunofluorescent staining of RRV infection of RAW-F3 cells. (A) Noninfected RAW-F3 cells. Infected RAW-F3 cells at (B) 1 day p.i., (C) 14 days p.i., (D) 39 days p.i., (E) 60 days p.i., and (F) 62 days p.i. RRV-infected (m.o.i. ⫽ 0.1) RAW-F3 cultures from day 14 p.i. onward represent RAW-F3R (RAW-F3 cells that have recovered from initial RRV-induced CPE).

macrophage RAW 264.7 cells up to day 50 p.i. using immunofluorescence RRV antibody staining. It has not yet been established whether virus persists long-term in patients or is associated with clinical relapse of symptoms. The clinical observation that relapse of RRVD symptoms can occur several months after clearance of RRV infection, associated with the experimental finding that RRV can persist in macrophages for up to six weeks (Lidbury et al., 2000), prompted a long-term investigation into the effects of RRV persistence on macrophages and their immunoregulatory properties. This study established that RRV infection persisted for 170 days in macrophages. Additionally, immunofluorescence (IFA)-detected infection was found to spontaneously relapse to enhanced levels, accompanied by macrophage cytopathic effect (CPE). Also, persistent RRV infection induced dysregulation of macrophage immune function associated molecules, which may aid the survival of RRV by facilitating the evasion of specific adaptive immune responses. RESULTS Persistence and relapse of RRV infection during long-term culture RAW cells were examined for RRV shortly after infection and at sequential time points over 5 months p.i. This was achieved using IFA staining and a RT-PCR approach

specific for the negative RNA strand of the RRV-E2 gene that is only present during replication of the virus. IFA analysis showed detectable RRV infectivity in RAW-F3 cells shortly after infection (1 day p.i., Fig. 1B) and at 14 days p.i. (Fig. 1C). No fluorescence was detected for noninfected RAW-F3 cells (Fig. 1A). During this culture period approximately 95% of the cells became nonadherent due to virus-induced CPE. The remaining 5% of cells recovered (RAW-F3R), maintained adherence, and cleared the virus as shown by IFA on days 39 and 60 p.i. (Figs. 1D and 1E). Surprisingly, although the virus had apparently cleared, RRV was again detected by IFA on day 62 p.i. (Fig. 1F), corresponding with CPE and detachment of the cells. With continued culture these cells again recovered, and the virus again became undetectable by IFA. This observation of viral relapse followed by recovery occurred spontaneously throughout the months of culturing. Semi-quantitative RT-PCR analysis detected little RRV expression shortly after infection (days 1–14 p.i.) where 95% of the RAW-F3 cells underwent viral CPE (Figs. 2A and 2B, ⫺2 and ⫺3), and this is likely to reflect the poor state of health of these cells. Upon recovery from CPE (post 14 days), the level of RRV mRNA increased and continued to be detected even during the time that IFA was unable to detect RRV protein (Figs. 2A and 2B, 4–10) (days 39–148 p.i.), indicating the virus had not been cleared from the cells. The level of RRV mRNA in healthy RAW-F3R cells (Figs. 2A and 2B, 4–10) was also found to

RRV PERSISTENT INFECTION OF MURINE MACROPHAGES

283

3F and 3G, ⫺2). Interestingly, RRV in the room temperature cultured RAW-F3R cells localized to intracellular vesicles (Fig. 3D) instead of only residing within the cytoplasm like the other persistently infected cells. These cells were also different to the FCS-depleted cells, having decreased expression of RRV by RT-PCR analysis (Figs. 3F and 3G, ⫺4) compared to the unstressed control (Figs. 3F and 3G, ⫺2). Morphological comparison between RAW-F3 and RAW-F3R cells

FIG. 2. Semi-quantitative RT-PCR analysis of RRV-E2 gene expression post RRV infection. (A) RT-PCR showing RRV and GAPDH expression in RAW cells. Lanes: 1, noninfected RAW-F3-negative control cells; 2, RAW-F3R 1 day p.i.; 3, 14 days p.i.; 4, 39 days p.i.; 5, 41 days p.i.; 6, 72 days p.i.; 7, 86 days p.i.; 8, 91 days p.i.; 9, 133 days p.i.; 10, 148 days p.i. (B) Quantitation of RT-PCR samples showing the intensity of the RRV PCR product relative to GAPDH.

vary, but unlike the fluctuation of IFA-detectable RRV protein, this did not correspond with CPE of the cells and was likely due to changes in culture/medium conditions and stage of cell cycle that has been reported by others (Igarashi et al., 1977). Stress induced relapse of RRV load RAW-F3R cells (120 days p.i.) were stressed by either depleting the serum (1% FCS instead of 10%) for 4 days or culturing at room temperature for 3 days to verify whether RRV relapse, as detected by IFA, could be induced and to test whether mRNA expression of RRV also increases with relapse of RRV. In response to altering the culture conditions the cells became detached and underwent CPE. Cells were analyzed for the presence of virus protein by IFA and viral RNA expression by semiquantitative RT-PCR. Similar to the RRV-infected RAW-F3 cells (positive control, day 3 p.i. Fig. 3E), immunofluorescent staining detected increased levels of RRV in the FCS-depleted RAW-F3R cells (Fig. 3C) compared to nonstressed RAW-F3R cells (Fig. 3B) and noninfected RAW-F3 cells (negative control, Fig. 3A). This result was supported by RT-PCR analysis showing corresponding higher levels of RRV mRNA in these cells (Figs. 3F and 3G, ⫺3) compared with unstressed RAW-F3R cells (Figs.

Detailed morphological analysis using transmission EM found that the intracellular structure of the noninfected RAW-F3 and persistently RRV infected RAW-F3R (day 84 p.i.) cells in a nonadherent state was identical (Fig. 4A). These cells were approximately 8–12 ␮m in diameter, with characteristic features of macrophages (Elgert, 1996; Lowe, 1997) including a vacuolated cytoplasm with an irregular-shape nucleus, numerous small lysosomal granules, and pseudopodia extending from the cells. High-scale magnification of the persistently infected RAW-F3R cells detected low levels of viral production demonstrated by small numbers of viral particles within intracellular vacuoles (Fig. 4B), being released from the cell (Fig. 4C) and budding from the cell membrane (Fig. 4D), that were absent in the noninfected RAW-F3 cells. RRV infection increases the phagocytic potential of the host RAW-F3R cell To assess whether macrophage function was affected by persistent RRV infection, Fc␥R-mediated phagocytosis of SRBCs was examined. Noninfected RAW-F3 and infected RAW-F3R cells from days 80 and 84 p.i. were cocultured briefly with SRBCs for 30 and 60 min and then stained and visualized by light microscopy and transmission EM. RRV infection was found to enhance the phagocytic activity of the host cell with 20% of the RAW-F3R cells from days 80 and 84 p.i. having engulfed SRBCs at 30 and 60 mins (Fig. 5D). The noninfected RAW-F3 cells did not engulf the SRBCs as rapidly; no phagocytosis was detected at 30 min and only 5% of the cells had phagocytosed SRBCs after 60 min (Fig. 5D). The RAWF3R cells also displayed a different morphology to the noninfected RAW-F3 cells. Those RAW-F3R cells that were actively engulfing SRBCs appeared more elongated (Fig. 5B), with SRBCs clustered around the cells and within the cells (depicted by arrows in Figs. 5B and 5C). The noninfected RAW-F3 cells, however, appeared spherical/ovoid-shaped and did not have SRBCs clustered around the cells (Fig. 5A). It is likely that the elongated morphology of cells actively phagocytosing SRBCs is due to an increased surface area that occurs in phagocytic cells prior to phagocytosis and enables them

284

WAY, LIDBURY, AND BANYER

FIG. 3. The effect of stress on the infectivity of persistent RRV in RAW-F3R cells at 120 days p.i. The level of RRV was examined by immunofluorescence staining of (A) noninfected RAW-F3 cells, (B) unstressed RAW-F3R cells, (C) FCS-depleted RAW-F3R cells, (D) reduced culture temperature (room temperature) of RAW-F3R cells, and (E) RAW-F3R cells day 3 p.i. (positive control). The level of RRV expression was also examined by semi-quantitative RT-PCR of (F1) noninfected RAW-F3 cells (negative control), (F2) unstressed RAW-F3R cells, (F3) FCS-depleted RAW-F3R cells, and (F4) reduced culture temperature RAW-F3R cells. Corresponding quantitation of the RRV RT-PCR samples relative to GAPDH is shown in (G).

to extend the dendritic protrusions and engulf extracellular material (Hackam et al., 1998). RRV infection causes dysregulation of expression of costimulatory molecules CD80 and CD86 To determine whether prolonged RRV infection altered the expression of immune function associated cell surface molecules, RAW-F3 and persistently infected RAWF3R cells, collected at various times p.i., were analyzed by FACS. Expression of the costimulatory molecules CD80 and CD86 were examined as these molecules are

normally up regulated in response to viral infection (Johnston et al., 1996; Mcadam et al., 2000). RRV infection resulted in changes to the level of expression of CD80 (Fig. 6) but did not affect CD86 expression (data not shown) in the RAW cells. RRV infection enhanced expression of CD80 in a proportion of the cells at 1 day p.i. but then led to a significant decrease in the level of expression of CD80 during persistent infection when compared with the mean level of expression in the noninfected RAW-F3 cells (Figs. 6A and 6B). Other cell surface markers that were studied, but not affected by RRV infection

RRV PERSISTENT INFECTION OF MURINE MACROPHAGES

285

FIG. 4. Intracellular examination of RAW-F3 and RAW-F3R cells by transmission EM. (A) Low-scale magnification showing morphology representative of both RAW-F3 and RAW-F3R (84 days p.i.) cells. (B–D) High-scale magnification showing localization of RRV in persistently infected RAW-F3R cells at 84 days p.i.

included CD11b, CD44, F4/80, CD40, CD45R (B220), CD69, and MHC class II 1-A b (data not shown). RRV persistent infection affects the expression of host cell immune-associated cytokine and chemokine genes Expression of the virally induced cytokines IFN-␥ (Czarniecki et al., 1984; Kohonen-Corish et al., 1990) and TNF-␣ (Czarniecki et al., 1984; Sambhi et al., 1991) in RAW-F3R cells increased during the first 14 days of infection (Figs. 7A3, 7B3, and 7A4, 7B4), compared with noninfected RAW-F3 cells (Figs. 7A2 and B2) by semiquantitative RT-PCR analysis. This corresponds with the stage of RRV-induced CPE. Upon recovery and during persistent infection, however, the level of expression of these cytokines declined to, or below, the basal level of expression detected in noninfected cells (Figs. 7A and 7B, 5–8). Dysregulation of TNF-␣ was also detected at the protein level by ELISA analysis (Fig. 8). IFN-␥ protein

production, however, was not examined in these cells as macrophages are generally considered poor producers of this cytokine, and therefore small changes in the level of IFN-␥ production may go undetected by this technique. The expression of MCP-1, a monocyte chemoattractant previously found to be induced by RRV infection (Mateo et al., 2000), was similarly up regulated early after infection (Figs. 7A and 7B, ⫺3); however, unlike IFN-␥ and TNF-␣, it continued to be expressed at higher levels than detected in noninfected cells during prolonged persistent infection (Figs. 7A and 7B, 4–8). DISCUSSION This investigation has utilized an in vitro cultured macrophage cell line to examine both the molecular and the cellular properties of persistent RRV infection and how persistent RRV affects the expression of immune-associated molecules. A key observation was that RRV infection can relapse after apparent clearance (assessed by

286

WAY, LIDBURY, AND BANYER

FIG. 5. RAW-F3 and RAW-F3R phagocytosis of SRBCs. Duplicate samples of RAW-F3 and RAW-F3R (days 80 and 84 p.i.) cells were cocultured with SRBCs for 30 and 60 min. A representative photograph of these cells analyzed by light microscopy is shown in (A) RAW-F3 and (B) RAW-F3R cells. Higher scale electron magnification showing internalization of SRBCs (indicated by an arrow) in the RAW-F3R cells is

IFA) either spontaneously or in response to stress, and this corresponded with macrophage nonadherence and CPE. After a relapse episode the macrophages again recovered and remained persistently infected. This observation may provide insights into the mechanisms underlying clinical relapse of disease symptoms that is known to occur months or years after the initial diagnosis of RRV disease. In addition, this in vitro culture system may be useful as a model to study factors that enable viral relapse in other persistent viral infections. Exactly what triggers the relapse of infection leading to CPE requires further examination. However, as described by this study, exposing macrophages to stress during culture established that vigorous viral replication can be reinduced in persistently infected cells. Also, of particular interest was the localization of virus to intracellular vesicles as a result of reducing the culture temperature. RRV is accustomed to fluctuations in temperature due to its ability to infect several species, for example, mosquito cells cultured at 28°C (Richardson et al., 1980) and bird cells that can reach temperatures of 40°C (Strauss and Strauss, 1994). It is possible that localization of virus to the intracellular vesicles and the corresponding low level of RRV mRNA detected by RT-PCR were due to an alteration in pH that prevented fusion of the virus with endosomal membranes and hindered the replication process, as previous studies have shown that a low pH is required in the endosomes for fusion to occur (Kielian and Jungerwirth, 1990; Marsh, 1984). Alphavirus persistence is not a novel phenomenon and was first recognized for Sindbis virus in mouse fibroblast cultures (Inglot et al., 1973). Of relevance to our study, the authors also noted the periodic appearance of CPE in their cultures that coincides with increased multiplication of virus. Persistent RRV infection was first demonstrated in confluent primary mouse (striated) muscle cultures; however, unlike in the earlier Sindbis virus study, periodic CPE was not detected up to 42 days p.i. (Eaton and Hapel, 1976). RRV persistence has also been shown in macrophages previously (Linn et al., 1996) and we have demonstrated that, similar to the infection of muscle cells, RRV can be detected in RAW 264.7 cultures by plaque assay up to day 42 p.i. (Lidbury et al., 2000). Detection of viral mRNA and virus particles in intracellular vesicles of the persistently infected RAW-F3R cells presented here demonstrated that RRV can persist for much longer durations of time (5 months) and does not completely clear from the cells. This finding is similar to that of persistence of RRV in Aedes albopictus cells of mosquitoes (Richardson et al., 1980) which also demonstrated persistence of RRV within intracellular vesicles.

shown in (C). Quantitation indicating the percentage of RAW cells that have engulfed SRBCs by 30 and 60 min is shown in (D).

RRV PERSISTENT INFECTION OF MURINE MACROPHAGES

287

For survival it is important for the virus to be able to both persist within host cells and manipulate the cellular environment to avoid or neutralize specific adaptive immune responses triggered by infection. Our studies showed that RRV persistence (post 14 days of infection) altered immunoregulatory properties normally induced in macrophages by the presence of virus. This included dysregulation of antiviral cytokines, TNF-␣, and IFN-␥. Apart from antiviral properties, IFN-␥ also has a role at the early stages of an immune response, prior to adaptive T cell immune engagement. Low-level production from innate immune cells in response to virus serves to activate monocyte/macrophages, as well as induce expression of the costimulatory molecule CD80 on these cells (Freedman et al., 1991). In this study RRV, like other viruses, induced expression of these cytokines at the early cytopathic stage of infection; however, as persistence was established the expression of these cytokines reduced despite the continued presence of the virus. This suggests that RRV may utilize an immune avoidance mechanism that dysregulates the expression of these cytokines. Apart from dysregulation of cytokines, expression of the chemoattractant MCP-1 was also dysregulated, but unlike for the cytokines studied, MCP-1 continued to be up regulated during both the cytopathic and the persistent stages of infection. This may contribute to the continued recruitment of mononuclear cells to the site of arthritis that is commonly observed in RRVD patients (Soden et al., 2000). Of importance to the activation of adaptive immune responses to virus infections is the interaction of costimulatory molecules, CD80 and CD86, on antigen-presenting cells such as macrophages, with the T lymphocyte receptors CD28 and cytotoxic T lymphocyte-associated-4 (CTLA-4) (Fleischer et al., 1996). Although the role of CD80 in RRV pathogenesis is unknown, the importance of this molecule in anti-viral immune responses is becoming increasingly apparent. For instance, CD80 is directly involved in the development of protective primary T cell and B cell immune responses against herpes simplex virus (Thebeau and Morrison, 2002) and anti-HIV CD8 responses triggered by autologous macrophages (Barker et al., 1999). In this study RRV infection stimulated a greater proportion of the cells to express CD80 during the cytopathic phase of infection that corresponds with a stage of cellular activation. This result is not unexpected as macrophages express CD80 upon activation (Freedman et al., 1991), in response to viral infection (Lumsden et al., 2000), and with IFN-␥ stimulation (Freedman et al.,

FIG. 6. Persistent RRV infection affects expression of the costimulatory molecule, CD80. RAW-F3R cells that had been stored periodically p.i. were recovered in culture and then used for FACS analysis. In (A) the average of three samples mean fluorescent intensities of unstained

controls (black) and stained samples (white) are shown for noninfected RAW-F3, at the cytopathic stage of infection (day 1 p.i.), and RAW-F3R cells during persistent infection at days 31, 58, 98, and 143 p.i. Error bars and significant differences (*P ⬍ 0.05) comparing noninfected RAW-F3 and infected RAW-F3R cells are also shown. In (B) representative histograms of the fluorescent intensity for each cell sample are shown.

288

WAY, LIDBURY, AND BANYER

FIG. 8. TNF-␣ response to Ross River virus (RRV) infection of RAW 264.7 macrophage cultures from days 2 to 21 p.i. RAW 264.7 murine macrophages were cultured in DMEM–5% FCS for 18–24 h prior to infection with RRV at a m.o.i. of 0.1 PFU. Culture supernatants were collected at days 2, 7, 15, and 21 p.i. from control noninfected wells (n ⫽ 3), as well as RRV-infected wells (n ⫽ 3). TNF-␣ protein concentration in culture supernatants was measured by sandwich ELISA. Error bars are shown and the level of TNF-␣ was found to be significantly different in the infected compared with noninfected cell samples on days 2, 7, and 15 (P ⬍ 0.05).

1991). Importantly, IFN-␥ was up regulated in the RAW cells at the same time as CD80. During the persistent phase of infection, however, the level of expression of CD80 on these cells fell below the basal level found in noninfected cells. At low levels of expression this molecule may effect T cell responsiveness to this virus, as blocking CD80 function in mixed lymphocyte cultures has been shown to generate alternatively activated macrophages that suppress T cell responses (Tzachanis et al., 2002). This type of immune avoidance mechanism would support persistence of the viral infection by inhibiting clearance by T cells. In this study the expression of CD86 was not effected by RRV, which is unexpected as this molecule is also usually induced by viral infection (Lumsden et al., 2000) and has been shown to be the key costimulatory molecule on macrophages that enhances CD8 cell suppression of HIV replication (Barker et al., 1999). Together, these results suggest that RRV, like the other viruses described here, may utilize immune avoidance mechanisms that directly effect anti-viral adaptive T cell responses by dysregulation of the costimulatory molecules CD80 and CD86. Whether this dysregulation leads to alternatively activated macrophages that suppress T cell responses awaits further investigation. One macrophage immune function that was not compromised by RRV was phagocytosis. In fact, RRV infecFIG. 7. Semi-quantitative RT-PCR of cytokine/chemokine expression by RAW-F3 and RAW-F3R cells. (A) Shows RT-PCR analysis of the expression of IFN-␥, TNF-␣, MCP-1, and GAPDH. Lanes: 1, no DNA template for negative control; 2, noninfected RAW-F3; 3, RAW-F3 1 day p.i.; 4, RAW-F3R 14 days p.i.; 5, RAW-F3R 27 days p.i.; 6, RAW-F3R 39 days p.i.; 7, RAW-F3R 48 days p.i.; 8, RAW-F3R 72 days p.i. (B) Quantitation

of the RT-PCR samples showing fold change in the level of cytokine and chemokine expression in RAW-F3R cells with respect to the level of expression in noninfected RAW-F3 cells.

RRV PERSISTENT INFECTION OF MURINE MACROPHAGES

tion up regulated the level of phagocytic activity in these cells. In agreement with this finding, a previous clinical study also detected enhanced phagocytic activity of macrophages isolated from the synovial fluid of RRVD (polyarthritis) patients (Clarris et al., 1975). Although phagocytosis usually assists with the elimination of extracellular pathogens, there is also a role for phagocytosis in clearing damaged cells that result from viral infection. Enhanced phagocytic activity may be advantageous to RRV, enabling the virus to more readily target macrophages where it can establish a persistent infection. In conclusion, these studies have made substantial steps in further understanding the effects of persistent RRV infection on macrophage activities that may underpin the etiology of RRVD symptoms such as epidemic polyarthritis. We have established that RRV can persist in macrophages for long periods of in vitro culture without leading to clearance, which was characterized by phases of viral dormancy (IFA negative, RRV RNA positive, cellular adherence, and virions observed in intracellular vesicles) and periodic relapse (IFA-positive, enhanced levels of RRV RNA, cellular detachment, and CPE). In addition, RRV persistence altered the expression of macrophage immunoregulatory molecules that may act to both exacerbate disease symptoms and allow RRV to avoid both immune detection and response. Whether relapse of RRV infection effects the expression of these molecules in persistently infected cells is the subject of continued investigation. The results described herein will be beneficial to studies of other persistent viral infections and mechanisms by which such viruses affect the immunoregulatory properties of the host cell. MATERIALS AND METHODS Ross River virus Virus stocks for these studies were derived from the full-length cDNA clone of T48, designated RR64 (Kuhn et al., 1991). To produce infectious virus, the clone was linearized by SacI digestion, transcribed in vitro from the cDNA using SP6 RNA polymerase, and the infectious RNA was transfected into BHK-21 cells as previously described (Kuhn et al., 1991). Viral stocks were propagated in Vero cells (ATCC CCL-81), and no virus stocks exceeded two Vero cell passages prior to experimental use. Viral titers were determined by plaque assay on Vero cells (Lidbury and Mahalingam, 2000). Murine macrophage cell lines RAW-F3 and RAW-F3R The mouse macrophage cell line, RAW 264.7 (ATCC TIB-71), was maintained in IMEM (Gibco/BRL) supplemented with 10% heat-inactivated fetal calf serum (HIFCS) (Gibco/BRL), 1% penicillin/streptomycin (Sigma, St. Louis, MO), and 2 mM L-glutamine (Sigma). Individual RAW 264.7 cells were cloned by seeding 96-well trays

289

(Nunc, Roskilde, Denmark) at 0.5 cells per well, followed by their sequential expansion. These cloned cells were screened for susceptibility to RRV at a multiplicity of infection (m.o.i.) of 0.1. A cell clone that showed RRV growth of 4.1 (⫾0.17) log 10 PFU/mL at day 2 p.i. was selected for further study and designated RAW-F3. RAW-F3 cells were seeded at a density of 5 ⫻ 10 5 cells/well in 24-well trays and RRV-infected at a m.o.i. of 0.1 (in PBS ⫹ 1% HI-FCS) for 1 h at 37°C. The virus inoculum was removed and 1.0 mL of fresh medium was added to each well. By 3 days p.i. approximately 95% of the cells had became nonadherent and died due to virus-induced CPE. The remaining 5% of cells recovered from infection approximately 10 to 14 days p.i. and were termed RAW-F3R (RAW-F3 cells “recovered” from RRV infection). The RAW-F3R cells were cultured continuously for 5 months p.i. and subcultured every 3 days, and samples were periodically stored at ⫺70°C in medium containing 10% dimethyl sulfoxide for future molecular and biological analyses. The recovery of macrophages after primary RRV infection has also been noted in a previous study (Linn et al., 1996). IFA detection of RRV in infected and stressed RAW-F3R cells Noninfected RAW-F3 and RAW-F3R cells at 1, 14, 39, 60, and 62 days p.i., as well as RAW-F3R cells (120 days p.i.) that had been either stressed by depleting the serum (1% FCS instead of 10%) for 4 days or cultured at room temperature for 3 days were used for this analysis. These cells were plated at a density of 6 ⫻ 10 4 cells/ml, fixed in methanol:acetone (1:1) for 1 min, rehydrated in 1⫻ PBS for 1 h at room temperature, and then incubated for 1 h at 37°C with a polyclonal mouse anti-RRV antibody (kindly provided by Dr. E. Lee, ANU, Australia) (1:100) in PBS–1% FCS. Cell samples were subsequently washed twice in PBS and exposed to FITC-conjugated anti-mouse IgG (Fab 2, 1:100) (Chemicon International, Temecula, CA) in PBS–1% FCS for 1 h at 37°C. Cell samples were then washed three times in PBS, mounted with a 10% glycerol in PBS (pH 8) solution, and viewed under a ZEISS Axiphot photomicroscope. To determine infectivity, the quantity of cell fluorescence was compared between RRV infected and noninfected control cells. Semi-quantitative RT-PCR analysis of IFN-␥, MCP-1, TNF-␣, and RRV gene expression in RAW-F3 and RAW-F3R cells mRNA was extracted from RAW-F3 (not infected), persistently infected RAW-F3R cells, and stressed RAW-F3R cells with the QuickPrep Micro mRNA purification system (Amersham Pharmacia Biotech). This extraction system utilized oligo-dT to target the poly-A tail of the negative strand of RRV RNA that is only present during replication of the virus (Ou et al., 1981; Strauss and

290

WAY, LIDBURY, AND BANYER

Strauss, 1994). The purified mRNA sample was used to prepare first-strand cDNA using a “Ready-To-Go” firststrand cDNA synthesis kit (Amersham Pharmacia Biotech). Gene expression was analyzed by hot-start semiquantitative RT-PCR. Briefly, the first-strand cDNAs were PCR-amplified in various experiments using Taq polymerase (Gibco/BRL), the supplied 10⫻ buffer, MgCl 2 (2 mM), dNTPs (0.2 mM) (Pharmacia Biotech), and oligonucleotide primers (2 ␮M). Primers included GAPDH 5⬘TCCACCACCCTGTTGCTGTA and 3⬘ACCACAGTCCATGCCATCAC which amplified a 500-bp PCR product. IFN-␥ primers 5⬘AGGAACTGGCAAAAGGATGGT and 3⬘GTGCTGGCAGAATTATTCTTATTG amplified a 400-bp PCR product. TNF-␣ primers 5⬘TCCCTCCAGAAAAGACACCA and 3⬘GGATGAACACCCATTCCCTT amplified a 500-bp PCR product. MCP-1 primers 5⬘ATGCAGGTCCCTGTCATGCT and 3⬘GACCAGTGTGACAGTGAACT amplified a 444-bp PCR product. uRRV-E2 primers 5⬘GCGCGAATTCGTAGTGTAACAGAGCACTT and 3⬘GCGCGAATTCCTGCGTTCGCCCTCGGTGCG amplified a 1300-bp PCR product. All cDNA PCR programs included an initial denaturation step at 94°C for 20 s (hot-start) prior to the addition of enzyme to prevent mispriming. All programs also included an extension step at 72°C for 20 s. All primers were annealed for 20 s using the following temperatures for each primer pair and amplified with the following number of cycles: GAPDH, 65°C, 28 cycles; IFN-␥, 60°C, 35 cycles; TNF-␣, 58°C, 28 cycles; MCP-1, 58°C, 28 cycles; uRRV-E2 gene-specific primers, 65°C, 35 cycles. RT-PCR samples were viewed by standard electrophoresis of 2% agarose gels and visualized using a Novaline gel documentation system. Prior to the actual assays, the thermal conditions and number of cycles used were optimized for both primer pairs so that the PCR products obtained did not proceed past the exponential phase of the PCR reaction nor exceed analysis limits. All semiquantitative RT-PCR experiments were carried out more than once. For semi-quantitation RT-PCR gels were scanned using Adobe Photoshop 6.0.1 (Adobe Systems Inc.) and quantitated using Image Guage V3.3 (Fuji PhotoFilm Co. Ltd. & Kohshin Graphic Systems). Briefly, the intensity of GAPDH PCR products was normalized, which then enabled cross-comparison of the intensity of the gene of interest. The intensity of these genes was presented either as an intensity value relative to GAPDH or as a fold change in the intensity found in infected RAW-F3R cells with respect to a basal level of intensity found in noninfected RAW-F3 cells. Transmission EM analysis of RAW-F3 and RAW-F3R cells RAW-F3 and RAW-F3R (day 84 p.i.) cells from culture or during phagocytic analysis, at a density of 1 ⫻ 10 7 cells/ml, were washed with 1% PBS, pelleted, and resus-

pended in 1 ml of PBS. Cells were then fixed with 2% glutaraldehyde (LADD Research Industries, Inc., VT) in 0.1 M sodium cacodylate (BDH) buffer, pH 7.4, for 2 h, followed by two 30-min washes in fresh sodium cacodylate buffer. A secondary fixation was done with 1% osmium tetroxide (BDH) in 0.1 M sodium cacodylate buffer, pH 7.4, for 1.5 h, followed by two 30-min washes in fresh sodium cacodylate buffer. The samples were then washed three times in distilled water, stained en bloc in 2% uranyl acetate (BDH) for 2 h, and rinsed three additional times in distilled water. Samples were dehydrated in a graded alcohol series, embedded in Spurrs’ resin (Agar Scientific), and sectioned to 80 nm using a Reichhert Ultracut. Sections were stained with Reynold’s lead citrate (BDH) for 15 min and examined using a Phillips EM 301 electron microscope at 60 kV. Phagocytic ability of RAW-F3 and RAW-F3R cells RAW-F3 and RAW-F3R (days 80 and 84 p.i.) cells were cocultured with 1% hemolysin-sensitized sheep red blood cell (SRBCs) for 30 and 60 min at 37°C. Cell samples were stained in May–Gru¨newald–Giemsa, dried, mounted, and analyzed by light microscopy using a ZEISS Axiophot photomicroscope or processed for transmission EM as described above. The degree of phagocytic activity was assessed and described as the percentage of cells that had phagocytosed SRBCs by 30 and 60 min. FACS analysis of RAW-F3 and RAW-F3R cell surface markers RAW-F3 and RAW-F3R (1, 31, 58, 98, and 143 days p.i.) cells at a density of 1 ⫻ 10 5 cells/ml were stained with the following antibodies according to the manufacturer’s instructions: CD11b (Caltag), CD44 (PharMingen/Becton–Dickinson), CD40 (Southern Biotechnology Associated, Inc.), CD45R (B220) (Caltag), CD69 (PharMingen/ Becton–Dickinson), CD80 (PharMingen/Becton–Dickinson), CD86 (PharMingen/Becton–Dickinson), I-A b MHC class II alloantigen (PharMingen/Becton–Dickinson), F4/ 80, (kindly provided by Dr. A. Hapel, JCSMR, ANU, Australia). Samples were analyzed using the Becton–Dickinson FACS scan with a standard filter set-up and Cell Quest software. The data for unstained controls and stained samples were presented as the average of three samples medium relative fluorescent intensities. The standard deviation, shown as error bars, and the statistical (t test) significant difference (*P ⬍ 0.05) between the medium relative fluorescent intensities of noninfected RAW-F3 and infected RAW-F3R cells was calculated. FACS analysis was repeated three times; the data shown, including histograms of the fluorescence intensity, are representative of the results obtained.

RRV PERSISTENT INFECTION OF MURINE MACROPHAGES

Assessment of the effect of RRV infection on RAW cell production of TNF-␣ protein RAW cells were seeded into 24-well culture trays (Nunc, Roskilde, Denmark) at a density of 3.0 ⫻ 10 5 cells per well in 1 mL of complete DMEM with 5% heatinactivated fetal calf serum (Trace, Melbourne Australia). Cultures were incubated for 18–24 h at 37°C in a humidified atmosphere (95%) containing 5% CO 2. The cultures were then infected with RRV at a m.o.i. of 0.1 PFU in PBS containing 1% HI-FCS. Noninfected controls were treated with PBS–FCS alone. After incubation for 1 h at 37°C, the viral inoculum was removed and fresh DMEM–FCS added to each well (1 mL per well). Culture supernatants were collected at days 2, 7, 15, and 21 p.i.; 0.5 mL fresh DMEM–FCS per well was added after each collection. The concentration of TNF-␣ protein was measured in RAW cell supernatants via sandwich ELISA techniques. Briefly, 96-well Nunc polysorp plates were coated with a monoclonal antibody (TN3-19.12, hamster anti-mouse) to murine TNF-␣ in pH 9.6 bicarbonate coating buffer. After overnight incubation at 4°C, wells were blocked with PBS containing 3% bovine serum albumin (BSA) and 0.1% Tween 20 and incubated at room temperature for 2 h. Supernatant samples were then added undiluted to the appropriate wells, with a recombinant murine TNF-␣ standard also applied to wells after first being serially diluted 1:2 in PBS containing 1% BSA and 0.1% Tween 20 (50–0.05 ng/mL). Plates were then incubated overnight at 4°C, after which wells were probed with rabbit antimouse TNF-␣ polyclonal antibodies. Color was developed via the addition of a sheep anti-rabbit alkaline phosphatase conjugate (incubated for 90 min at room temperature) followed by the addition of substrate in an alkaline buffer (room temperature for 30–60 min). Well absorbance was read at 410 nm (reference 630 nm). Positive samples for TNF-␣ were taken as those with an absorbance at two standard deviations above background absorbance. Between each step of this procedure, wells were washed six times with PBS containing 0.1% Tween 20. ACKNOWLEDGMENTS We thank Duncan Adams and Kavitha Kugathas (University of Canberra), the EM and FACS service departments at JCSMR, and Leah Nolan for assistance in preparation of this manuscript. This research was supported by Prof. Ian Ramshaw, via block funding of the John Curtin School of Medical Research, ANU.

REFERENCES Barker, E., Bossart, K. N., Fujimura, S. H., and Levy, J. A. (1999). The role of CD80 and CD86 in enhancing CD8 ⫹ cell suppression of HIV replication. Cell. Immunol. 196, 95–103. Clarris, B. J., Doherty, R. L., Fraser, J. R., French, E. L., and Muirden, K. D. (1975). Epidemic polyarthritis: A cytological, virological and immunochemical study. Aust. N. Z. J. Med. 5(5), 450–457.

291

Czarniecki, C. W., Fennie, C. W., Powers, D. B., and Estell, D. A. (1984). Synergistic antiviral and antiproliferative activities of Escherichia coli-derived human alpha, beta, and gamma interferons. J. Virol. 49(2), 490–496. Eaton, B. T., and Hapel, A. J. (1976). Persistent noncytolytic togavirus infection of primary mouse muscle cells. Virology 72(1), 266–271. Elgert, K. D. (1996). “Immunology. Understanding the Immune System,” Wiley–Liss, New York. Fleischer, J., Soeth, E., Reiling, N., Grage-Griebenow, E., Flad, H. D., and Ernst, M. (1996). Differential expression and function of CD80 (B7–1) and CD86 (B7–2) on human peripheral blood monocytes. Immunology 89(4), 592–598. Flexman, J. P., Smith, D. W., Mackenzie, J. S., Fraser, J. R., Bass, S. P., Hueston, L., Lindsay, M. D., and Cunningham, A. L. (1998). A comparison of the diseases caused by Ross River virus and Barmah Forest virus. Med. J. Aust. 169(3), 159–163. Fraser, J. R., Cunningham, A. L., Clarris, B. J., Aaskov, J. G., and Leach, R. (1981). Cytology of synovial effusions in epidemic polyarthritis. Aust. N. Z. J. Med. 11(2), 168–173. Fraser, J. R., Ratnamohan, V. M., Dowling, J. P., Becker, G. J., and Varigos, G. A. (1983). The exanthem of Ross River virus infection: Histology, location of virus antigen and nature of inflammatory infiltrate. J. Clin. Pathol. 36(11), 1256–1263. Freedman, A. S., Freeman, G. J., Rhynhart, K., and Nadler, L. M. (1991). Selective induction of B7/BB-1 on interferon-gamma stimulated monocytes: A potential mechanism for amplification of T cell activation through the CD28 pathway. Cell. Immunol. 137(2), 429–437. Hackam, D. J., Rotstein, O. D., Sjolin, C., Schreiber, A. D., Trimble, W. S., and Grinstein, S. (1998). v-SNARE-dependent secretion is required for phagocytosis. Proc. Natl. Acad. Sci. USA 95, 11691–11696. Hawkes, R. A., Boughton, C. R., Naim, H. M., and Stallman, N. D. (1985). A major outbreak of epidemic polyarthritis in New South Wales during the summer of 1983/1984. Med. J. Aust. 143(8), 330–333. Hazelton, R. A., Hughes, C., and Aaskov, J. G. (1985). The inflammatory response in the synovium of a patient with Ross River arbovirus infection. Aust. N. Z. J. Med. 15(3), 336–339. Igarashi, A., Koo, R., and Stollar, V. (1977). Evolution and properties of Aedes albopictus cell cultures persistently infected with sindbis virus. Virology 82(1), 69–83. Inglot, A. D., Albin, M., and Chudzio, T. (1973). Persistent infection of mouse cells with Sindbis virus: Role of virulence of strains, autointerfering particles and interferon. J. Gen. Virol. 20(1), 105–110. Johnston, L. J., Halliday, G. M., and King, N. J. (1996). Phenotypic changes in Langerhans’ cells after infection with arboviruses: A role in the immune response to epidermally acquired viral infection? J. Virol. 70(7), 4761–4766. Kielian, M., and Jungerwirth, S. (1990). Mechanisms of enveloped virus entry into cells. Mol. Biol. Med. 7(1), 17–31. Kohonen-Corish, M. R., King, N. J., Woodhams, C. E., and Ramshaw, I. A. (1990). Immunodeficient mice recover from infection with vaccinia virus expressing interferon-gamma. Eur. J. Immunol. 20(1), 157–161. Kuhn, R. J., Niesters, H. G., Hong, Z., and Strauss, J. H. (1991). Infectious RNA transcripts from Ross River virus cDNA clones and the construction and characterization of defined chimeras with Sindbis virus. Virology 182(2), 430–441. Lidbury, B. A., and Mahalingam, S. (2000). Specific ablation of antiviral gene expression in macrophages by antibody-dependent enhancement of Ross River virus infection. J. Virol. 74(18), 8376–8381. Lidbury, B. A., Simeonovic, C., Maxwell, G. E., Marshall, I. D., and Hapel, A. J. (2000). Macrophage-induced muscle pathology results in morbidity and mortality for Ross River virus-infected mice. J. Infect. Dis. 181(1), 27–34. Linn, M. L., Aaskov, J. G., and Suhrbier, A. (1996). Antibody-dependent enhancement and persistence in macrophages of an arbovirus associated with arthritis. J. Gen. Virol. 77(Pt. 3), 407–411. Lowe, J. A. A. S. (1997). “Human Histology,” 2nd ed., Times Mirror, Barcelona.

292

WAY, LIDBURY, AND BANYER

Lumsden, J. M., Roberts, J. M., Harris, N. L., Peach, R. J., and Ronchese, F. (2000). Differential requirement for CD80 and CD80/CD86-dependent costimulation in the lung immune response to an Influenza virus infection. J. Immunol. 164, 79–85. Mackenzie, J. S., and Smith, D. W. (1996). Mosquito-borne viruses and epidemic polyarthritis. Med. J. Aust. 164(2), 90–93. Marsh, M. (1984). The entry of enveloped viruses into cells by endocytosis. Biochem. J. 218(1), 1–10. Mateo, L., La Linn, M., Mccoll, S. R., Cross, S., Gardner, J., and Suhrbier, A. (2000). An arthrogenic alphavirus induces monocyte chemoattractant protein-1 and interleukin-8. Intervirology 43(1), 55–60. Mcadam, A. J., Farkash, E. A., Gewurz, B. E., and Sharpe, A. H. (2000). B7 costimulation is critical for antibody class switching and CD8(⫹) cytotoxic T-lymphocyte generation in the host response to vesicular stomatitis virus. J. Virol. 74(1), 203–208. Ou, J. H., Strauss, E. G., and Strauss, J. H. (1981). Comparative studies of the 3⬘-terminal sequences of several alpha virus RNAs. Virology 109(2), 281–289. Richardson, M. A., Boulton, R. W., Raghow, R. S., and Dalgarno, L. (1980). Polypeptide synthesis in alphavirus-infected Aedes albopictus cells

during the establishment of persistent infection. Arch. Virol. 63(3–4), 263–274. Sambhi, S. K., Kohonen-Corish, M. R., and Ramshaw, I. A. (1991). Local production of tumor necrosis factor encoded by recombinant vaccinia virus is effective in controlling viral replication in vivo. Proc. Natl. Acad. Sci. USA 88(9), 4025–4029. Soden, M., Vasudevan, H., Roberts, B., Coelen, R., Hamlin, G., Vasudevan, S., and La Brooy, J. (2000). Detection of viral ribonucleic acid and histologic analysis of inflamed synovium in Ross River virus infection. Arthritis Rheum. 43(2), 365–369. Strauss, J. H., and Strauss, E. G. (1994). The alphaviruses: Gene expression, replication, and evolution. Microbiol. Rev. 58(3), 491– 562. Thebeau, L. G., and Morrison, L. A. (2002). B7 costimulation plays an important role in protection from herpes simplex virus type 2-mediated pathology. J. Virol. 76(5), 2563–2566. Tzachanis, D., Berezovskaya, A., Nadler, L. M., and Boussiotis, V. A. (2002). Blockade of B7/CD28 in mixed lymphocyte reaction cultures results in the generation of alternatively activated macrophages, which suppress T-cell responses. Blood 99(4), 1465–1473.