Our project is to analyse how a virus may modulate the immune responses of its host and which consequences these modifications may have on diseases independent from the infection, but concomitant with it.
Viral infections cause many diseases affecting patients of every age, targetting every organ, and leading to various symptomes, high morbidity and sometimes death. In many cases, the disease is linked to viral tropism, which is the ability of the virus to selectively infect a cell population. Thus, a virus infecting cells in the repiratory system may cause bronchitis or pneumonia, a virus that will attack liver cells will be responsible for hepatitis.
However, in some cases, the mechanisms leading to disease as a consequence of a viral infection are more complex. For instance, anti-viral immune responses may have deleterious side effects for the host. Moreover, many viruses have developed the ability to modulate the immune responses of their host, either by suppressing parts of these defenses that are detrimental to them, or by activating other responses that will divert the immune system from its antiviral tasks. The purpose of this modulation of the immune system is to allow the virus to escape to these defenses. However, as the modifications are not limited to antiviral responses, they also affect the whole immune system. Thus, all the pathologies that are developing into the host at the time of infection and whose mechanisms involve the immune system may be modulated by the infection, although initially they were independent from it.
A consequence of this distinct pathogenic mechanism is for the prevention of diseases linked to viral infections. In the case of diseases specifically linked to a particular virus, such as smallpox, poliomyelitis or flu viruses, it is possible to protect the population by vaccinations that will elicit defenses specifically directed against the causative virus. In contrast, when there is modulation of the immune system, many different viruses may have a similar effect on the host immune system. Vaccination against one of them will therefore not be sufficient to protect against all the causative agents that may trigger the same disease. In this situation, it is more important to understand which mechanisms are affected by infections and to act on those mechanisms.
Interestingly, some diseases may be improved by a concomitant viral infection as well, when the modifications of the immune system have a beneficial effect for the host. In these cases, to understand those mechanisms may allow to define new therapeuthic approaches that might be used also in patients who have not developed viral infections.
From a practical point of view, a fundamental study of the consequences of a viral infection on the immune system of its host requires the use of laboratory animals. It would indeed be unethical to voluntarily infect human subjects to determine the effects of viruses on their immune responses, and the complexity of this system, the numerous interactions between involved cells make impossible a complete in vitro analysis.
In our laboratory, we use a model of mouse infection, since the immune system of this species is well characterized and close to the human immune system. The virus that we use is called lactate dehydrogenase-elevating virus (LDV). It is benign for most mice, but it is able to deeply, but transiently modify the immune responses of its host, which makes it very appropriate to analyse pathologies indirectly caused by an infectious agent that modulates the immune microenvironment. Infection with LDV results in activation of many immune cells, such as macrophages, naturel killer cells or B lymphocytes that secrete antibodies, but also in a qualitative modification of the responses of other cells, including T helper lymphocytes, and in a suppression of some responses.
By using this virus, we have observed dramatic effects on some diseases that were simultaneously induced in the infected mice. For instance, natural killer cell activation protects infected animals against the growth of some cancers naturally sensitive to this type of cells. Similarly, LDV-infected mice are protected against the development of a disease close to multiple sclerosis or against some dangerous side effects of the graft of immune cells, a leukemia treatment. To understand the complex mechanisms triggered by the infection may allow to better understand the mechanisms responsible for the severity of these diseases, and therefore to define new therapeutic approaches for their treatment.
In contrast, LDV infection strongly sensitizes mice to other diseases. For instance, infected animals quickly develop septic shock, a very severe complication of bacterial infections. Moreover, they present anemia and platelet destruction much more important in the presence of antibodies directed against red cells and platelets. It is thus possible to induce in mice infected with LDV a clinical syndrome close to that observed in young children after infection with benign viruses, that is characterized by bleedings, often of limited importance, but sometimes affecting the central nervous system, with much more severity. Mouse analyses have shown that a central mechanism in the sensitization of infected animals to these diseases is macrophage activation. An excessive activation triggered by the virus induces those cells to secrete large amounts of toxic molecules in response to bacterial compounds and to destroy red cells and platelets bound by autoantibodies. The analysis of mechanisms leading to macrophage activation after LDV infection may allow to determine how to limit this excessive activation. If transposed to human diseases, these knowledges may thus also open the way to new therapeutic approaches.
The possibility for evoluted organisms to survive viral infections depends on the ability of their immune system to eliminate the infectious agent. Therefore, numerous mechanisms, involving different types of immune cells such as cytolytic lymphocytes, T helper and B lymphocytes and macrophages, the molecules that allow those cells to communicate, namely the lymphokines, and the products of those interactions, including antibodies, have been elaborated. On the other hand, viral infections strongly modulate the immune microenvironment of the host which often leads to alterations of responses elicited against non-viral antigens and of concomitant diseases with an immune component. Our project is to analyse, in murine models, some aspects of these relations between viruses and the immune system, with a special interest for the consequences of virally-induced modulations of the immune micro-environment on the course of diseases developing concomitantly to the infection in the host.
Viral infections result in a dramatic increase in the proportion of IgG2a
Of particular interest is the fact that all antibody responses are not equal. Indeed, depending on their isotype, immunoglobulins display various properties, such as differential affinity for receptors expressed on phagocytes. During the last years, we found that the isotype of antibody responses was influenced by concomitant viral infections. The effect of the virus resulted in a dramatic increase in the proportion of IgG2a, not only in antiviral antibodies, but also in immunoglobulins with an antigenic target unrelated to viral proteins. The modulations of antibody responses was analysed with more details by using a model of infection with lactate dehydrogenase-elevating virus (LDV), a common mouse nidovirus that induces strong and early immune responses. We could demonstrate that a dual regulation of antibody responses by gamma-interferon (IFN-γ) and interleukin-6 explains this isotypic bias. IgG2a anti-LDV antibodies were found to be more efficient than other isotypes to protect mice against a fatal polioencephalomyelitis induced by the virus. However, the modification of the isotype of antibodies reacting with self antigens could potentially lead to more deleterious autoimmune reactions.
T helper lymphocyte differentiation
This property of viruses to enhance selectively the production of one immunoglobulin isotype could depend on the preferential activation of a subset of T helper lymphocytes. Indeed, different subpopulations of those cells, called Th1 and Th2, respectively, are distinguished in particular by their capability of producing selectively IFN-γ or interleukin-4, which can selectively trigger B lymphocytes to produce IgG2a or IgG1, respectively. We have found that LDV infection results in a suppression of Th2 responses elicited by immunization with an antigen unrelated to the virus. More recently, other populations of Th lymphocytes, such as Th17 cells that are involved in some autoimmune responses, as well as T regulatory lymphocytes that inhibit ongoing responses have been described. Preliminary observations in our group show a dramatic prevention of diseases such as autoimmune encephalitis and graft-versus-host disease in mice acutely infected with LDV. Whether this protective effect of the virus results from a modulation of T helper/ T regulatory cells remains to be determined.
Activation of natural killer cells
Many of the influences that viruses may have on diverse immune responses can be explained by the production of pro-inflammatory cytokines, including IFN-γ. Therefore, our analysis of the relationship between viruses and the immune system has focused on the activation, by LDV, of cells from the innate immune system that are able to secrete this cytokine, namely the natural killer (NK) cells. Within a few days after infection, a strong and transient NK cell activation, characterized by accumulation of this cell population in the spleen, by enhanced IFN-γ message expression and production, as well as by cytolysis of target cell lines was observed. Two pathways of IFN-γ production have been observed that both involve NK cells. The first pathway, found in normal mice, is independent from type I IFN and from interleukin-12. The second pathway involves interleukin-12, but is suppressed by type I IFN. Because NK cells and IFN-γ may participate in the defense against viral infection, we analyzed their possible role in the control of LDV titers, with a new agglutination assay. Our results indicate that neither the cytolytic activity of NK cells nor the IFN-γ secretion affect the early and rapid viral replication that follows LDV inoculation.
Interestingly, NK cell activation results in an increased expression of CD66a (CEACAM-1), an adhesion molecule that display immunoregulatory function on activated T lymphocytes. However, this enhanced expression, that is also found on immature NK cells, results from NK cell stimulation with IL-12 and IL-18, but not with LDV. Therefore, different pathways of NK cell activation, leading to various phenotypes and, probably various functions, may be observed.
Activation of macrophages and enhanced susceptibility to endotoxin shock
Activation of cells of the innate immune system by LDV includes also macrophages and leads to an enhanced response to lipopolysaccharide (LPS), and to an exacerbate susceptibility to endotoxin shock. A synergistic effect of LDV and LPS triggered dramatic production of tumor necrosis factor (TNF) and IFN-γ. Susceptibility to LPS shock was completely mediated by TNF, and partially by IFN-γ. This increased susceptibility of LDV-infected mice to endotoxin shock was not mediated by modulation of the expression of membrane receptors for LPS, but was correlated with increased levels of soluble LPS receptors. In this context, the production of type I IFNs may protect the host against exacerbated pathology by controling the production of IFN-γ.
Blood autoimmune diseases
Virally-induced macrophage activation leads also to an enhanced phagocytic activity, with potential detrimental consequences for ongoing autoimmune diseases. LDV infection resulted in moderate thrombocytopenia in normal animals through enhanced spontaneaous platelet phagocytosis. Our analysis was then focused on autoantibody-mediated blood autoimmune diseases. A new experimental model of anti- platelet response was developed in the mouse. Immunization of CBA/Ht mice with rat platelets was followed by a transient thrombocytopenia and production of autoantibodies that react with epitope(s) shared by rat and mouse platelets. Two IgM anti-platelet monoclonal autoantibodies were further analyzed. They recognized mouse platelet antigens and could induce both platelet destruction and impairment of their function. This response was found to depend on CD4+ T helper lymphocytes reacting with rat, but not with mouse platelets. These anti-rat platelet T helper cells were mainly of the Th1 phenotype. When transferred into naive mice, they enhanced the anti-mouse platelet antibody response induced by subsequent immunization with rat platelets. In addition, depletion of CD25+ cells enhanced the thrombocytopenia induced by immunization with rat platelets whereas adoptive transfer of CD4+CD25+ cells from immunized mice suppressed it. Our results suggest therefore that activation of anti-rat platelet T helper cells can bypass the mechanism of tolerance and result in the secretion of autoreactive antibodies, but this response is still controlled by regulatory T cells that progressively develop after immunization.
We have analysed whether a viral infection could modulate such an autoantibody-mediated autoimmune disease. In mice treated with anti-platelet antibodies at a dose insufficient to induce clinical disease by themselves, infection with LDV or mouse hepatitis virus was followed by severe thrombocytopenia, whereas infection alone, without autoantibody administration led to a moderate disease. Similarly, administration of anti-erythrocyte monoclonal autoantibody to mice resulted in the development of a transient hemolytic anemia that was dramatically enhanced by a simultaneous infection with LDV, leading to the death of most animals. This viral infection induced an increase in the ability of macrophages to phagocytose in vitro autoantibody-coated red cells, and an enhancement of erythrophagocytosis in the liver.
Treatment of thrombopenic or anemic mice with clodronate-containing liposomes and with total IgG indicated that opsonized platelets and erythrocytes were cleared by macrophages. Administration of clodronate-containing liposomes decreased also the in vitro phagocytosis of autoantibody-coated red cells by macrophages from LDV-infected animals. The increase of thrombocytopenia triggered by LDV after administration of anti-platelet antibodies was largely suppressed in animals deficient for IFN-γ receptor. Moreover, LDV infection resulted in an increased expression of receptors recognizing the Fc portion of antibodies, which may at least partially leads towards the enhanced phagocytic activity of macrophages. Together, these results suggest that viruses may exacerbate autoantibody-mediated thrombocytopenia and anemia by activating macrophages through IFN-γ production, a mechanism that may account for the pathogenic similarities of multiple infectious agents. Regulation of macrophage activation results in modulation of autoantibody-mediated cell destruction and may be considered as a possible treatment for autoimmune diseases that involve phagocytosis as a pathogenic mechanism.
Together, these two models may correspond to the development of some auto-immune diseases: a first stimulus triggers the production of autoantibodies, through molecular mimicry. A second stimulus, such as a viral infection, leads to the activation of macrophages and results in the destruction of opsonized target cells.
VIRUSES AND IMMUNE MICROENVIRONMENT