Immunologic and histologic observations in reovirus-induced otitis media in the mouse.

The goals of this study were to develop a mouse model for virally induced otitis media, and to study the immune response to infection. Intranasal inoculation of mice by reovirus was used to induce otitis media. Immunohistochemical evidence for the presence of reovirus in the nasopharynx, eustachian tubes, and middle ears and the amount of infiltrating B-cells and T-cells in those sites were serially evaluated by painlessly sacrificing animals over a 21 -day period. Reovirus antigen was detected in the middle ear mucosa by day 4 in 75% of infected animals, and histologic evidence for otitis media was found in 54% of all infected animals. A significant increase in B-cells in the nasopharynx and eustachian tubes was noted 7 to 10 days following infection. The number of infiltrating T-cells did not vary significantly from that in the control animals at any of the sites. These results provide a basis for further investigations of the immune response in otitis media.

T-Helper 1 and T-helper 2 cytokine responses in gut-associated lymphoid tissue following enteric reovirus infection.

Enteric infection of mice with reovirus serotype 1 elicits antibody and cytotoxic T-lymphocytes in gut-associated lymphoid tissue (GALT). This led to the hypothesis that T-helper 1 (Th1) and T-helper 2 (Th2) responses develop in GALT. Reverse transcriptase-polymerase chain reactions on RNA from Peyer's patches (PP), intraepithelial lymphocytes (IEL), and lamina propria (LP) lymphocytes demonstrated that interferon (IFN)-gamma message was increased in PP and IEL, but not in LP following infection. No increase in mRNA for interleukin (IL)-4, IL-5, or IL-6 was detected. IFN-gamma, IL-5, and IL-6 were produced in in vitro cultures of PP 4-10 days postinfection. PP and spleen lymphocytes from infected mice produced IFN-gamma, but no IL-5 following in vitro restimulation. Infection also induced production of mRNA for the beta2 chain of the IL-12 receptor in PP. We conclude that reovirus induces robust Th1 and weak Th2 cell responses in GALT.

Enteric reovirus infection as a probe to study immunotoxicity of the gastrointestinal tract.

The gastrointestinal (GI) tract contains a complex immune system that defends the host against a wide range of pathogens and toxins. The GI tract is also exposed to many environmental toxins that could adversely affect intestinal immunity, and few systems to study immunotoxicity of the GI tract have been described. We demonstrate that intestinal reovirus infection can be used as a system to assess the effects of toxins on intestinal and systemic immunity. Mice were given various doses of cyclophosphamide (CY) for 5 days at doses ranging from 100 to 500 mg/kg by the oral route or 200 mg/kg by the intraperitoneal route. On day 3 of dosing, mice were orally infected with reovirus serotype 1, strain Lang. The effects of CY on viral clearance, intestinal and systemic immune responses, and distribution of intestinal lymphocytes were assessed. Mice treated with CY failed to clear the virus in a dose-dependent manner, and serum anti-reovirus antibody titers were suppressed. Virus-specific IgA in cultures of intestinal tissue from CY-treated mice was significantly reduced compared to controls, although total IgA production was not affected. The virus-specific cytotoxic T-cell response in spleen was also suppressed in CY-treated animals. Cyclophosphamide treatment reduced the number and percentage of B-cells in Peyer's patches. Reovirus infection did not increase cellularity of Peyer's patches in CY-treated mice. Cyclophosphamide treatment also had little effect on the phenotype of intestinal intraepithelial lymphocytes. These data demonstrate that intestinal reovirus infection is useful in studying exposure of the GI tract to immunotoxic agents.

Enhanced mucosal and systemic immune responses to intestinal reovirus infection in beta2-microglobulin-deficient mice.

Enteric infection of mice with respiratory enteric orphan virus (reovirus) type 1, strain Lang elicits both humoral and cellular immune responses. To investigate the role of CD8+, alpha/beta T-cell receptor (TCR)+ T cells in mucosal immunity to an enteric pathogen, we examined immune responses and viral clearance following enteric reovirus infection in C57BL/6, B6129F2, and beta2-microglobulin-deficient (beta2m-/-) mice. Analysis of Peyer's patch and lamina propria culture supernatants revealed a two- to threefold increase in levels of reovirus-specific immunoglobulin A in beta2m-/- mice compared to normal controls. These data corresponded to a similar increase in the frequency of virus-specific immunoglobulin A-producing cells in Peyer's patches and lamina propria and an increase in immunoglobulin G-producing cells in spleens from beta2m-/- mice compared to controls. These increased humoral immune responses were not due to a difference in B-cell populations because cell counts and flow cytometric analyses showed that beta2m-/- and control mice had similar numbers and percentages of B cells in mucosal and systemic tissues. Analysis of cytokine message by reverse transcriptase-PCR 5 and 10 days after infection revealed no difference in message level for transforming growth factor beta, gamma interferon, interleukin-4, interleukin-5, or interleukin-6 for all mouse strains. Virus tissue titers determined by plaque assay at 5 and 10 days after infection demonstrated that beta2m-/- mice cleared reovirus from the small intestines with the same efficiency as control mice. Collectively, these data suggest that CD8+, alpha/beta TCR+ T cells may regulate mucosal and systemic humoral immune responses to oral infection with reovirus.

Effects of the route of infection on immunoglobulin G subclasses and specificity of the reovirus-specific humoral immune response.

Reovirus serotype 1, strain Lang (T1/L), a well characterized enteric virus, elicits humoral and cellular immune responses in mice. Although orally and intradermally induced infections generate comparable reovirus-specific serum antibody titers, little is known about the effects of the route of infection on the systemic immunoglobulin G (IgG) response. To assess whether the route of exposure affects virus-specific humoral immunity, we infected various strains of mice with reovirus T1/L by the oral or intradermal routes. At day 10 following infection, virus-specific serum antibody titers and IgG subclasses were determined by enzyme-linked immunosorbent assay. Serum IgG2a and IgG2b antibodies were detected in all mouse strains independent of the route of infection. Mice of the H-2d haplotype that received an intradermal infection also had high levels of reovirus-specific serum IgG1. This dichotomy of responses was not associated with differences in the types of cytokine produced by draining peripheral lymph nodes. However, peripheral lymph node lymphocytes from C3H mice produced significantly higher levels of gamma interferon than did BALB/c, C57BL/6, and B10.D2 mice. Additionally, peripheral lymph node lymphocytes from all strains of mice produced only low levels of interleukin-5, with no detectable level of interleukin-4 or interleukin-6. Analysis of specific antibody at inductive sites of the immune response showed that orally infected Peyer's patches produced predominantly IgA and intradermally infected peripheral lymph nodes produced predominantly IgG2a. Western blot (immunoblot) analysis showed that virus-specific IgA, IgG1, and IgG2a reacted with reovirus structural proteins. These data suggest that the route of infection affects the isotype and IgG subclasses, but not the antigen specificity, of the local antibody response. In addition, virus-specific IgG1 generated following an intradermally induced infection is linked to the H-2d major histocompatibility complex haplotype.

3,4-Dichloropropionanilide-induced atrophy of the thymus: mechanisms of toxicity and recovery.

The herbicide 3,4-dichloropropionanilide (propanil) has several well-documented neurotoxic and immunotoxic effects on mice. We report here a detailed characterization of the effects of propanil exposure on the thymus. We found that at doses of 100-200 mg/kg, propanil induces significant thymic atrophy between 2 and 7 days postexposure. This atrophy is characterized by a decrease in thymus/body ratio and a decrease in cellularity. Flow cytometric analyses of thymuses from propanil- and vehicle-treated mice indicate that the CD4(+) CD8(+) population of immature cells, is most significantly decreased in propanil-exposed mice. We performed cell cycle analysis of thymocyte populations using two-color surface staining and the DNA binding dye 7-aminoactinomycin D to determine whether thymic atrophy was associated with changes in the percentages of cells in the S, G2, and M phases of the cell cycle. We found a high percentage of proliferating CD4(+)CD8(+) thymocytes 4 days after exposure. Thus, recovery of the thymus occurs following increases in thymocyte proliferation, most notably the immature CD4(+) CD8(+) thymocytes. We tested the hypothesis that glucocorticoids play a role in the observed atrophy by examining thymuses in adrenalectomized, propanil-treated mice. No atrophy was observed in those animals. These results suggest that propanil has an immunotoxic effect on the thymus that appears to be mediated, in part, by endogenous glucocorticoids.

Changes in primary and secondary lymphoid organ T-cell subpopulations resulting from acute in vivo exposure to propanil.

Acute exposure to the herbicide propanil is immunotoxic for selected immune functions, as well as causing changes in the weights of the thymus and spleen. Although spleen cellularity and weight increase with propanil exposure, the thymus: body weight ratio decreases with increasing doses of propanil. The present study analyzes the thymocyte subpopulations in the thymus, spleen, and mesenteric lymph nodes. C57Bl/6 mice were treated with either 0, 100, 150, or 200 mg/kg propanil, and 7 d later thymocyte populations were analyzed by flow cytometry. In the thymus, propanil exposure resulted in a dose-dependent decrease in total numbers of T cells, as would be expected with its reduced weight. Determination of the thymocyte subpopulation distribution in the thymus showed a significant reduction in the number of CD3+CD4+CD8- (CD3+4+8-), CD3+CD4-CD8+ (CD3+4-8+), and CD3+CD4+CD8+ (CD3+4+8+) cells. Percent distribution of these thymic cell subpopulations showed similar decreases only with the highest dose. Apparent dose-related decreases in the numbers of CD3-CD4+CD8+ (CD3-4+8+) cells were also noted and were attributed to the general decrease in total thymus cells. The percentage of CD3- subpopulations showed an increasing trend with dose, which suggests that at 7 d postpropanil exposure there may be a specific effect on this most immature population. Although the size and cellularity of the spleen were increased, no change in CD4+ or CD8+ cell distribution was observed. Similarly, mesenteric lymph nodes showed no changes in the cell subpopulation distribution between propanil-treated and control animals.

Morphologic and functional alterations of mucosal T cells by cholera toxin and its B subunit.

Despite the mucosal immunogenicity and adjuvanticity in vivo of cholera toxin (CT), both CT and CT B subunit are strong inhibitors of T cell activation in vitro. This study asked whether such T cell inhibition is relevant to the mucosal effects of CT in vivo. The activation of T cells pulsed in vitro for only 15 to 120 min with CT or CT B subunit, respectively, was inhibited, consistent with the expected short exposure times in vivo. Although both CD8+ and CD4+ T cells were inhibited in vitro, CD8+ T cells bound more toxin and were inhibited to a greater degree than were CD4+ T cells. Intestinal gavage of mice with 10 micrograms CT did not alter the overall composition of Peyer's Patch, mesenteric lymph node, or spleen but did cause a marked depletion of intraepithelial lymphocytes, mainly CD8+ T cells, and of lymphocytes in the dome epithelium over Peyer's Patch. To determine whether such inhibition of T cells was functionally relevant in vivo, T cells from mice fed keyhole limpet hemocyanin (KLH) were adoptively transferred into naive recipients, who were then parenterally immunized. T cells from mice fed KLH alone inhibited both the systemic IgG and secretory IgA anti-KLH response, but T cells from mice fed KLH plus CT did not, indicating that mucosally applied CT was able to abrogate the induction of this suppressor T cell. We conclude that one of the mechanisms of CT's mucosal effects in vivo is the inhibition of certain mucosal T cell functions and alteration of the regulatory T cell environment in gut-associated lymphoid tissue.

Developmental relationship between cytotoxic alpha/beta T cell receptor-positive intraepithelial lymphocytes and Peyer's patch lymphocytes.

Following intraduodenal priming of mice with reovirus, precursor cytotoxic T lymphocytes (pCTL) rapidly appear in intraepithelial lymphocytes (IEL) and Peyer's patches. These cells express CTL activity after secondary in vitro stimulation with reovirus-infected cells. Adoptive transfer of Peyer's patch lymphocytes from normal BALB/c mice into reovirus-infected CB.17 severe combined immunodeficiency mice results in the infection-dependent appearance of large numbers of both CD8+Thy-1+ and CD8-Thy-1+, IEL that express the alpha/beta T cell receptor (TcR). Phenotypic and functional characterization of IEL derived from conventionally reared, reovirus-infected mice also points to extensive similarities in the pCTL derived from Peyer's patches and IEL. As in the Peyer's patches, pCTL are persistent in the IEL compartment for up to 4 weeks after infection. A large percentage of IEL that are recovered from reovirus-primed mice after in vitro culture are CD8+Thy-1+ cells that express alpha/beta TcR. Furthermore, depletion experiments demonstrate that the CD8+Thy-1+ population mediates the virus-specific CTL activity. Using limiting dilution analyses, it was estimated that 7 days after intraduodenal infection the average frequency of virus-specific pCTL was 197/10(6) CD8+Thy-1+ IEL and 190/10(6) CD8+Thy-1+ Peyer's patch lymphocytes. Taken together, these observations provide evidence that specific cellular immunity to reovirus in IEL is mediated at least in part, by conventional cytotoxic T lymphocytes and that these cells are functionally and phenotypically similar to the pCTL derived from the Peyer's patches.

Viral gastrointestinal infections.

Numerous disease-causing viruses of the gastrointestinal tract have been described. The diseases that they induce range from subclinical to fatal gastrointestinal or systemic organ infection. Recent advances in molecular biology and virus culture techniques, as well as improvements in animal models of these infections, have increased our understanding of pathologic processes associated with these viruses. Although our overall understanding of these viruses has improved to the extent that many more viruses can be identified as pathogens, we are still a long way from achieving a thorough understanding of the pathogenesis of viral infection and the mechanisms that mediate immunity.

Passive immunity to fatal reovirus serotype 3-induced meningoencephalitis mediated by both secretory and transplacental factors in neonatal mice.

The role of passively acquired immunity to reovirus-induced meningoencephalitis in neonatal mice was examined. It was determined that female mice were capable of conferring protection against viral infection and meningoencephalitis in neonates depending on the route by which the dams were immunized and the serotype of the immunizing virus. Female mice immunized with homotypic virus via the oral route developed the most potent response. Infected neonates born and nursed by these females developed no signs of disease, and no virus was recoverable from their small intestines, livers, or brains following infection. Neonates born to females immunized with homotypic virus by the subcutaneous route manifested no evidence of meningoencephalitis or virus dissemination, yet virus was recovered from neonatal intestines. Mice immunized with heterotypic virus by either the subcutaneous or the oral route also conferred protection against disease; however, virus was recovered in small intestines and livers of infected neonates. Based on results from foster-nursing experiments, it appears that factors obtained both during suckling and by transplacental transfer contribute to protection. Passive transfer of reovirus-immune mouse serum also protected neonates from disease. These results demonstrate that passive immune mechanisms can mediate the protection of neonates against reovirus infection and provide further evidence of the importance of the mucosal immune response in protection against pathogens that invade the host via mucosal tissues.

A further characterization of Candida albicans-induced suppressor B-cell activity.

We have previously established that components of the organism Candida albicans are capable of inducing suppressive activity in a population of B lymphocytes. The activity of this population is antigen non-specific. The proliferative response to T-cell, but not B-cell, specific mitogens is inhibited. In addition, the antibody response in vitro is suppressed. Since little is known about this relatively unique regulatory population, we have attempted to characterize both the expression and induction of activity of the Candida-primed cells. Our results show that both primary and secondary T-cell-dependent antibody responses are inhibited, whereas both type I and type II T-cell-independent antibody responses are not affected by the suppressor cell population. T-cell populations responsible for both interleukin-2 (IL-2) and cytolytic activity are also unaffected. These results suggest that while there is no antigen specificity for this population, the suppressive activity is extended to limited target cell populations. Results also suggest that both T cells and accessory cells are required for the induction of the suppressor cell activity, indicating that the Candida organism acts as a typical T-dependent antigen in the induction of regulatory cell activity.

The induction of T-suppressor cells with a soluble extract of Candida albicans.

We have previously shown that whole cell preparations of Candida albicans are capable of inducing immunosuppressive B-cell activity both in vivo and in vitro. In an effort to characterize the components of the yeast which manifest this immunomodulatory activity, we have successfully generated a soluble extract with dithiothreitol which exerts immunosuppressive activity. This extract is capable of inducing antigen-nonspecific suppressor cells which inhibit the antibody response of normal cells in coculture. Both primary and secondary antibody responses are suppressed by these cells. Our results also show that the suppressor cell population is a member of the L3T4+ Ly-1+ Lyt-2- T-cell lineage. These results provide evidence that Candida extracts may possess clinically significant immunomodulatory activities.

Reoviruses as probes of the gut mucosal T cell population.

Reovirus, serotype 1, causes a transient, asymptomatic infection of the murine intestine when given intraduodenally or orally. However, this infection markedly perturbs both B- and T- cell populations in Peyer's patches (PP) resulting in: 1) a rapid and persistent increase in specific precursors for cytotoxic T cells (pCTL) and a gradient of frequencies highest in PP and lowest in distal lymphoid tissue; 2) a similar increase in memory B cells committed to IgA; 3) the transient appearance of a subset of germinal center B cells identified by MAb, GC-T; 4) the appearance of pCTL among intraepithelial lymphocytes; and 5) the antigen non-specific alteration in Ig isotype potential of B cells previously primed and found in PP. The pCTL appearing upon acute gut mucosal infection with reovirus are Thyl+, Lyt2+, virus-specific, viral serotype non-specific, class I MHC haplotype restricted and occur within the subset of T cells which newly appears also identified by MAb GC-T. Infections of both neonatal and severe-combined immunodeficient mice indicate that the elements of the immune system may operate at many levels to resist, limit, contain, and resolve viral infection.