PRRSV pathogenesis. Initial pathogenesis studies with the VR-2332 North American prototypes of PRRS were done collaboratively by investigators at MN and SD (Benfield et al., 1992, Collins et al., 1992, Rossow et al, 1995). The PRRS virus replicates in an unknown primary site that results in a viremia as early as 12 hours after intranasal inoculation of three-day old gnotobiotic pigs (Rossow et al., 1995). The viremia distributes the virus in a variety of tissues especially lung and lymph nodes (Rossow et al., 1996; Lawson et al, 1997). Macrophages appear to be the major target cell in most tissues. Lesions in young pigs consist primarily of interstitial pneumonia and lymph node hypertrophy (see review by Rossow, 1998) and the severity of lung lesions is apparently virus isolate specific (Halbur et al, 1995). Although there are sequence differences between high and low virulent isolates of PRRSV, no correlation between viral gene sequences and virulence have been established (Meng et al, 1995).

The reproductive form of PRRS is seen after the late term infection of pregnant gilts and sows. The consequences of intrauterine PRRSV infection include 1) infertility of sows, 2) dead fetuses, 3) piglets that die or perform poorly, and 4) persistently infected pigs (Lager and Ackerman, 1994; Lager and Mengeling, 1995; Mengeling et al, 1996). During acute infection of the sow the PRRSV-infected macrophage acts a Trojan Horse, disseminating PRRSV throughout the pig (Cafruny and Bradley, 1996, Rossow et al., 1995). The associations between macrophage-tropism, persistent infection, and transplacental infection are not unique to the arteriviruses, since other macrophage-tropic viruses, such as HIV, cross the placenta and infect fetuses (Kesson et al., 1994). How infected macrophages contribute to infection of fetuses is not known. One possibility is that infected macrophages traffic to the placenta and establish localized regions containing large concentrations of virus. Another possibility is the direct transfer of virus from maternal to fetal macrophages, perhaps via macrophage-like cells in the placenta. Infected macrophages may also produce inflammatory cytokines, which increase the permeability of the placenta to viruses.

In addition to the direct effect of virus replication on tissues, several viral-mediated reproductive disorders are frequently associated with changes in immune regulatory cytokines. IL-1, TNF, IL-6, CSF-1, IL-8, and interferon (IFN), typically associated with macrophage function, also participate in a two-way communication between the maternal decidua and placental trophoblast cells during pregnancy. During disease this communication system becomes dysregulated, or overloaded, with pro-inflammatory antiviral cytokines (Raghupathy, 1997). One cytokine of particular interest is TNF-a, an autocrine factor produced by macrophages (Murtaugh, 1994), trophoblast cells (Hunt et al, 1990), maternal decidual cells (Casey et al., 1989), and Hofbauer cells (Berkowitz et al., 1990). Levels of TNF-a are elevated in amniotic fluid and fetal blood during intrauterine growth retardation and inflammation (Stallmach et al, 1995). TNF-a may also affect virus load. For example, TNF-a given to mice increases the number of permissive macrophages that support LDV replication (Broen et al, 1992; Haven et al, 1996). A similar situation could exist for PRRSV and investigators at IA, IL, MN and SD plan to determine how cytokines regulate the permissiveness of macrophages for PRRSV. Proposed research in this area is directed at characterizing PRRSV replication during late gestation and specifically focuses on how virus-related events at the fetal-maternal interface correlate with infection of fetuses, induction of abortagenic cytokines, and fetal death.

Finally, apoptosis may also have a role in the pathogenesis of PRRSV. In vivo experiments done in pigs or tissue culture at IA and NE suggest the existence of an indirect mechanism for the induction of apoptosis in non- infected, bystander cells. In both studies, a majority of the apoptotic cells were not infected with the PRRSV, indicating that PRRS infection induces apoptosis in neighboring cells (Sirinarumitr et al, 1998; Sur et al. 1998). A direct effect of the whole virus or its apoptogenic p25 glycoprotein on the cell membrane cannot be ruled out at this time. One proposed mechanism is that apoptogenic cytokines (i.e. tumor-necrosis-factor, TNF) are released into the environment from PRRSV infected cells, such as, PRRSV-infected macrophages, which besides being the most prominent cell in PRRSV-infected tissues, are known to enhance the TNF secretion when infected by viruses.

Viral persistence. IA, MN, NE and SD have been actively engaged in experiments to determine how virus persists in young pigs and breeding age swine. Viral persistence is defined as a period of virus replication beyond the acute symptomatic phase of infection. The prototypic model for arterivirus persistence is LDV infection of mice (Plagemann et al., 1995). LDV is cytopathic for a small subpopulation of macrophages, whose only known function is the elimination of excess circulating LDH enzyme from the blood. Virus replication is maintained by the infection of new permissive macrophages derived from a non-permissive progenitor population. The presence of a weak neutralizing antibody response combined with clonal exhaustion of CTLs explains how LDV escapes humoral and cell-mediated immune responses (Plagemann et al., 1995; Even et al., 1995). Studies at IA, MN, NE and SD indicate that PRRSV establishes a long-term asymptomatic infection (Benfield et al, 1998; Christopher-Hennings et al, 1995; Swenson et al, 1994; Wills et al, 1997).

SD conducted a long-term study of PRRSV replication in pigs exposed to PRRSV in utero as a result of inoculating sows with PRRSV at 90 days of gestation (Benfield et al., 1998; Lawson et al. 1998). These results indicated that for days 1 to 21 post-parturm, cells supporting PRRSV replication were observed in all organs and tissues, which is consistent with the macrophage- tropism of PRSV replication. Pigs that survived longer than 21 days eventually recovered and showed no clinical signs of PRRS. Beginning at 63 days post-parturition (PP), PRRSV was almost exclusively found in lymphoid tissues (tonsil and lymph nodes), but not in lung tissue, which is the main site of replication in the acute stage of the disease. The absence of virus in lung and serum correlated with the absence of detectable virus in other non-lymphoid organs, including salivary gland, liver, and kidneys. In contrast, PRRSV was isolated from tonsil and lymph nodes for up to 132 days PP. The biological relevance of low-level PRRSV replication in the lymphoid organs was evaluated by determining if PRRSV was transmitted to uninfected pigs. Sentinel pigs were introduced into the group of infected pigs at 64 days PP (2 sentinels), 84 (2 sentinels), 98 (one sentinel) and 112 days PP (1 sentinel). All pigs became infected within one week after introduction demonstrating that virus was readily shed to naive pigs.

There are several strategies that PRRSV may utilize to evade host defenses during persistence. Similar to LDV, PRRSV replication may be restricted to a small, non-essential population of macrophages. Another evasion strategy is the replication of PRRSV in immunologically privileged sites, such as the male reproductive tract and the eye (Molitor 1999). Several studies at IA, MN, NE and SD have documented the long-term replication of PRRSV in the male testes and accessory organs and presence of virus in semen (Christopher-Hennings et al, 1995, 1998; Shin et al, 1997; Swenson et al, 1994; Sur et al, 1997). The mechanism of persistence of PRRSV in the boar reproductive tract is still unknown. Perhaps the most prominent finding regarding cell tropism in the NE study is the notion that PRRSV can infect cells other than macrophages during the pathogenesis of testicular infection. Generally, testicular infection by PRRSV centers on two types of cells: (i) epithelial germ cells of the seminiferous tubules, primarily spermatids and spermatocytes, and (ii) macrophages, which are located in the interstitium of the testis (Sur et al. 1997). Formation of multi-nucleated giant cells (MGCs) and abundant germ cell depletion and death are observed. Importantly, we obtained evidence that such germ cell death occurs by apoptosis.(Sur et al. 1997). As a result there was a significant increase in the number of PRRSV-infected immature sperm cells (mainly MGCs, spermatids and spermatocytes) in the ejaculates of PPRSV-inoculated boars and these cells are likely responsible for the venereal transmission of PRRSV. In a separate study, scientists at NE also found there is a frank infection by PRRSV in macrophages of the atretic follicles of the ovarium, and some minor involvement of stromal and granulosa cells. (Sur et al.. Vet Pathol submitted). This data may circumstantially explain infertility and estrous alterations reported in PRRSV-infected and vaccinated gilts. However, the female gonad is not a likely site of PRRS persistence, because there was no evidence of ova infection and/or perpetuation of PRRSV in reproductive tissues or embryos.

Another possible mechanism to explain PRRSV persistence is antigenic drift, i.e. a change in the ammo acid sequence of epitopes that interact with neutralizing antibody. Scientists at SD determined changes in the ORF5 ectodomain in pigs that were persistently infected as a result of in utero exposure to PRRSV. Fifty percent of the GP sequences, obtained by direct RT-PCR amplification of PRRSV from tissues contained a Asp to Asn at amino acid 34. These results demonstrate that mutations appear as a result of selection (possibly antibody), but the mechanism driving this selection remains to be determined (Rowland et al, 1999, Virology, in press).

PRRSV Immunity. The best evidence for protective immunity against PRRSV is that herds return to normal production levels following clinical episodes of PRRSV. Presumably, humoral antibody, cell-mediated immunity and cytokines contribute to the resolution of PRRSV infections. Serum antibodies can be detected early after infection, but their role in protection is not known (Yoon et al, 1995). The role of neutralizing antibodies is suspect, because virus can be isolated from the serum and tissues of pigs in the presence of high levels of neutralizing antibodies and infectious virus can still be transmitted to susceptible (Wills et al, 1997). Antibody may also enhance the pathogenesis of the disease by a antibody-dependent enhancement (ADE). ADE occurs when antibody-virus complexes attached to a receptor on macrophages facilitating uptake of the virus by the macrophage. In PRRS, this may actually result in a greater number of macrophages being infected, because the antibody may facilitate the uptake of PRRSV by permissive macrophages (Yoon et al, 1996). IA, KS, NE and SD are collaborating to determine the role of antibodies in protection against PRRSV infections.

The role of cell-mediated immunity (CMI) in PRRSV is less defined than the role of antibodies. Antigen specific T-cell responses have been described in pigs following infection (Bautista and Molitor, 1997). These studies done at MN indicate that PRRSV induces an antigen-specific, CMI response in infected pigs, which was blocked by anti-CD4 and anti-MHC class II antibodies. This CMI response was first detected at 4 weeks after infection and continued to be detected up to 11 weeks. Likewise, through collaborative studies between IL and NE, we now know that there is a distinct contrast in the quality and kinetics of the cellular immune response induced following either vaccination or infection with PRRSV and similar exposure to PRV (Meier et al., 1997). In general, the cellular immune response to a virulent field strain of PRRSV is more pronounced and quicker to develop than similar responses elicited from the commercial modified-live vaccine viruses. All in all, it appears that the PRRSV modified-live vaccines have a lower efficiency in stimulating T-cell responsiveness to the challenge strain, when compared with a pseudorabies modified-live vaccine (taken as a gold-standard) (Zuckermann et al, 1997; Osorio et al, 1998). Interferon (IFN), the most important anti-viral cytokine, may also play a role in modulating PRRSV replication.

Interferon functions by upregulating several antiviral proteins, which make cells nonpermissive for virus infection (Samuel, 1991; Kerr and Stark, 1992). Investigators at SD found that IFN-y decreased the number of PRRSV-positive cells, virus yield, and viral RNA content in cell culture. This inhibitory effect was reversed by adding 2-aminopurine (2-AP), an inhibitor of dsRNA inducible protein kinase (PKR), suggesting that PRRSV is sensitive to IFN and that IFN-y inhibition of PRRSV replication occurs primarily through activation of a single antiviral protein, PKR. In pigs, IFN-y was measured using an ELISPOT assay developed by IL (Zuckermann et al, 1998). This test detects the production of IFN-y by individual antigen- stimulated T cells and is highly specific and sensitive for the detection of active CMI responses and immune memory. NE and IL followed the IFN-y and humoral immunity to PRRSV in a group of 12 animals. Antibodies detected by the PRRSV ELISA were absent in some animals at 7 months PI. Importantly, animals that exhibited no antibodies on the whole-virus ELISA were still positive by Western blot to the N and E viral proteins (Kwang et al, 1999). The ELISPOT (IFN-y) was positive at several times ranging from 318 to 522 days PI, even in animals that had become seronegative by conventional serologic tests. These studies indicate that PRRSV may persist in animals for an extended period of time and stimulate T-cell responses.

PRRSV may also modulate the host immune system through infection of the primary target cell, the macrophage. PRRSV replicates in and destroys macrophages (see review by Rossow, 1998) thereby decreasing the number of alveolar macropahges present in lung lavages during acute infection (Molitor et al, 1992; Zhou et al, 1992). Infection of macrophages with PRRSV also increases production of inflammatory cytokines (IL-1 and TNF) (Zhou et al, 1992) and suppresses antibacterial activity of these cells (Zhou et al, 1992; Thanawongnuwech et al, 1997; 1998). This reduced antibacteriacidal activity of macrophages in PRRSV infected pigs may explain the increase incidence of secondary bacterial infections observed in natural PRRSV infections (see review Rossow, 1998; Zimmerman et al, 1998). Furthermore, PRRSV may also modulate lymphocyte populations in pigs infected in utero and increase susceptibility to secondary bacterial infections (Feng et al, 1998).

Overall, information on the immune response to PRRSV and the mechanism of protective immunity to PRRSV is lacking. Additional studies on the role of neutralizing antibodies, CMI and cytokines in the pathogenesis and immunity of PRRSV need to be investigated and several stations IA, IL, MN, NE, KS and SD are collaborating to address this issue.

Control and prevention of PRRS. Progress in understanding the epidemiology of PRRSV and improving the accuracy of diagnostic testing procedures has enabled us to develop management protocols for the control and prevention of PRRS. There approaches were developed at IA, MN and NC including: gilt development, isolation and acclimation: partial depopulation; test and removal; segregated early weaning; all-in/all-out pig flow; vaccination with modified-live or killed vaccines; mass vaccination and unidirectional pig flow; serological monitoring; McREBEL and others (Zimmerman et al, 1998). Even with our present knowledge, the prevention, control and eradication of PRRSV using current control strategies, diagnostic tests and vaccines is not easily accomplished. Most of the management protocols are costly, time consuming and generally ineffective over time. The ability of PRRS virus to produce persistent infections is the main reason many management protocols fail. In addition, diagnostic assays cannot reliably identify persistently infected animals. Vaccines do not appear to induce long- term protective immunity and simultaneously eliminate or reduce virus shedding (Zimmerman et al, 1998). While eradication is an option with PRRSV, this may increase vulnerability of herds to re-infection with PRRS through the introduction of infected animals or other means. Improved methods to control, prevent and/or eradicate PRRSV are still needed.