Upon inoculation into tissues of the mammalian host, T. cruzi infects local cells where it multiplies intracellularly as amastigotes. At this early point in infection, the host depends on its innate immune response to microorganisms, which is primarily mediated by macro-phages and natural killer (NK) cells. Unac-tivated (resting) macrophages, however, have little ability to kill T. cruzi and serve as important host cells for parasite repli-cation. Within two hours of macrophage invasion, the majority of T. cruzi have escaped the phagosome and exist free in the host cytoplasm. It is assumed that by rapidly escaping the phagolysosome, T. cruzi is able to avoid degradation by intracellular microbicidal enzymes in the unprimed macrophage. A variety of stim-uli are capable of priming macrophages for antitrypanosomal activity, with IFN-γ being particularly potent. The addition of a complex mixture of cytokines, including interleukins (IL-2, IL-3, IL-4, and IL-5) to IFN-γ leads to even greater macrophage trypanocidal activity. Anti-IL-4 antibodies added to the mixture were capable of neu-tralizing the effect, but the combination of IFN-γ and purified IL-4 was no better than IFN-γalone.
Lipopolysaccharide significantly in-creases the trypanocidal effect of IFN-γ on macrophages, apparently through a mechanism not solely mediated by the stimulation of TNF production. Another cytokine that is effective at priming macro-phages for trypanocidal activity in vitro is granulocyte-macrophage colony-stimulat-ing factor.
Serum taken from an uninfected human does not lyse trypomastigotes, but serum taken from infected patients does lyse trypomastigotes. Thus, the comple-ment system requires the participation of the acquired immune response to T. cruzi to become operational. Parasite lysis by immune serum occurs primarily via the alternative complement pathway. This means that the complement cascade is ini-tiated by direct association of complement components (C3) with the parasite surface, rather than by fixation of complement by antibody Fc receptors. In fact, Fab and F(ab)2′ prepared from chagasic patients’ sera (which are incapable of fixing com-plement) were nearly as efficient as intact immunoglobulin G (IgG) in complement lysis assays with trypomastigotes. This implied that the effect of immune anti-bodies on complement lysis of trypomas-tigotes was a function of the specificity of the antigen-binding sites, rather than complement fixation by bound antibod-ies. In other words, there was a surface component on the parasites that, when neutralized by immune sera, rendered the parasites susceptible to complement lysis.
Antibodies directed to a 160-kDa pro-tein correlated with the capacity of the serum for complement-mediated lysis of trypomastigotes. Subsequent work led to the [purification and characterization of the specific parasite product, gp160. This glyco-protein bound the complement component C3b and inhibited C3 convertase formation, thus inhibiting activation of the alterna-tive complement pathway. It is membrane bound and shares genetic and functional similarities to the human complement reg-ulatory protein, decay accelerating factor. Further work indicated that this protein also bound human C4b, a component of the classical pathway C3 convertase, and therefore may restrict classic complement activation. The T. cruzi complement regu-latory protein is stage specific in that it is only expressed by mammalian forms of the parasite. A recent study reported that stable transfection ofT. cruzi epimastigotes (which are susceptible to complement lysis by normal serum) with the complementary DNA for the T. cruzi complement regula-tory protein conferred complement resis-tance.
Why does T. cruzi preferentially para-sitize muscle and nerve tissue? Arguing teleologically, the host immune response needs to control the parasite infection but not at the cost of destroying vital organs (i.e., heart and peripheral nerves). The down-regulatory immune responses may be more exuberant in these critical tissues where even minor injury can be fatal and regenerative capacity is limited (e.g., the cardiac conduction system). T. cruzi may have evolved to preferentially infect these tissues as relatively immune-privileged sites. Several other parasites that preferen-tially infect muscle tissue or nervous tissue include toxoplasmosis, trichinella, Taenia species (beef and pork tapeworms), and sar-cocystis. Each of these species depends on a carnivorous-definitive host ingesting the meat of an intermediate host to complete its life cycle. There is no evidence, how-ever, that consumption of T. cruzi–infected tissues plays a role in this parasite’s life cycle. Thus, the predilection for growth in muscle and nervous tissue is more likely to be related to a strategy for long-term parasite survival in the host, rather than a strategy related to direct transmission to secondary hosts.
IFN-γ induction of macrophage try-panocidal activity is associated with the production of hydrogen peroxide. How-ever, treatment of activated macrophages with catalase, superoxide dismutase, or sodium benzoate to scavenge respira-tory burst metabolites failed to inhibit trypanocidal activity in vitro, suggesting an oxygen-independent mechanism of T. cruzi destruction. Subsequent studies have revealed the importance of nitric oxide (NO) production in the killing mechanism in murine systems. Macrophages primed with IFN-γ produce NO, and the addition of an inhibitor to NO synthase blocks the trypanocidal activity. Levels of extracellu-lar l-arginine (the substrate for NO produc-tion) modulate the trypanocidal activity of macrophages. NO may be directly toxic to T. cruzi; however, it reacts with superoxide (O2−) to yield peroxynitrite (ONDO−), which is highly toxic to T. cruzi. In vivo, inducible nitric oxide synthase (iNOS) is induced at the protein and messenger RNA levels, and NO is released during acute infection in coincidence with secretion of IFN-γ and TNF. The administration of inhibitors of NO production leads to greater suscepti-bility of T. cruzi–infected mice. Mice carry-ing disruption of the iNOS genes are highly susceptible to T. cruzi infection. The IFN-γ–induced, NO-dependent mechanism of macrophage killing of T. cruzi can be inhib-ited by the addition of interleukin-10 or transforming growth factor-β in vitro.
NK cells participate in the innate immune response to T. cruzi infection, and NK cells secrete IFN-γ after incubation with T. cruzi in vitro. This was shown by culturing splenocytes from athymic nude mice with T. cruzi and detecting the secre-tion of IFN-γ, then demonstrating that the pretreatment of the splenocytes with anti-NK1.1 monoclonal antibody blocked the IFN-γ production. Since IFN-γ acti-vates macrophages to kill T. cruzi, the role of NK cells in T. cruzi infection would be expected to be protective. In fact, a rela-tively T. cruzi–resistant strain of mouse was rendered highly susceptible to infec-tion by pretreatment with anti-NK1.1 anti-bodies. Thus, NK cells appear to be an early source of IFN-γ that helps control parasite replication before the acquired immune response becomes predominant. NK cells may also be involved in the immune response later in infection, since they are present in inflammatory lesions of muscle in experimental mice 270 days after infection.
Trypanosoma cruzi multiplies and dis-seminates throughout the host before a specific antiparasitic immune response is mounted. In mice, parasites first become apparent in blood approximately five to seven days after infection and rise in numbers until three or four weeks into infection when the mice either die or the infection is controlled by the immune response (Figure 12.1). RAG knockout mice, which are deficient in both B- and T-cell function, have similar levels of par-asitemia compared with wild-type mice until day 13 of infection, at which point the parasitemia level becomes higher in the RAG knockouts. This indicates that the acquired immune response has little effect until about two weeks into the infec-tion. Although anti–T. cruzi antibodies are detectable at about day 7 in murine mod-els, protective antibodies are not present until several weeks later. This was shown in experiments in which antibodies taken from acutely infected mice were not protec-tive against T. cruzi infection in immune-naïve mice. However, experimental mice or rats that received sera from animals that survived acute infection experience a sig-nificant decrease in parasitemia level and mortality following a challenge with viru-lent parasites.
The importance of IgG in the humoral immune response was first demonstrated by Castelo Branco. The protective compo-nent of serum from chronically infected mice could be removed by staphylococcal protein A (which absorbs IgG). Further-more, the purified IgG component from whole serum was capable of conferring protection. Both IgG subclasses 1 and 2 are capable of clearing T. cruzi. Antibodies mediate protection from T. cruziby opsonization, complement activation, and antibody-dependent cellular cytotoxicity.
T cells have a protective role in the acquired immune response to T. cruzi. Experiments in mice showed that T-cell activation correlated with resistance to infection. Passive transfer of T cells from mice immunized against T. cruzi conferred resistance in mice challenged with T. cruzi. Deficient T-cell function is associated with increased sensitivity to infection. This has been shown in nude mice; thymec-tomized mice; mice treated with cyclo-sporine A, anti-CD4, and anti-CD8; and mice genomically deleted of CD4, CD8, β2-microglobulin, TAP-1, or MHC mol-ecules. The increased susceptibility associ-ated with deficient T-cell function is most apparent during acute infection. Mice that are depleted of T cells (CD4+ or CD8+) after they have survived the acute infection do not have altered parasitemia or longevity. The declining role of T cells and the impor-tance of humoral immunity in controlling infection after the acute stage are schemati-cally illustrated (Figure 12.1).
The presence of parasite-specific T cells has been demonstrated for both CD4+ and CD8+ T cells. Nickell and colleagues iso-lated CD4+ T cells from spleens of infected mice that proliferate to T. cruzi antigen in a HLA-restricted fashion. The CD4+ T-cell line recognized an undefined trypomasti-gote antigen(s), did not cross-react with Leishmania spp. or Toxoplasma gondii, and was able to passively protect syngeneic recipients from lethal T. cruzichallenge infection. Kahn and Wleklinski isolated and cloned CD4+ T cells from T. cruzi–infected mice that proliferate and secrete cytokines in response to the surface protein, SA-85, a member of the sialidase superfamily. The MHC class II epitope in this protein was mapped to a 20-amino acid sequence. Nickell and associates were able to isolate CD8+ T cells from infected mice that lysed parasite-infected target cells in an MHC-restricted manner. The parasite antigens involved in the stimulation of CD8+ T cells were not characterized. However, Wizel and colleagues detected class I–restricted CD8+ T cells from spleens of infected mice that lyse target cells, presenting epitopes from the trans-sialidase family of proteins. These cytotoxic lymphocytes passively transferred protection against challenge infection.
The inflammatory infiltrates in T. cruzi–infected experimental animals have been analyzed for cellular surface mark-ers. Mirkin and colleagues studied tis-sues from C3H/HeN mice during acute, early chronic, and late chronic infection. The inflammatory infiltrates consisted mainly of lymphocytes (60–90 percent) and macrophages (10–40 percent). The lymphocytes primarily carried the T-cell marker, Thy1.2. Both CD4+ and CD8+ T cells were present in the infiltrates of skeletal muscle, sciatic nerve, and spinal cord. The general trend was for a slight predominance of CD8+ cells over CD4+, particularly during the early chronic and late chronic stages. Sun and Tarleton also found that Thy1.2+ cells were the major lymphocyte population in tissues (cardiac and skeletal muscle) during acute infec-tion. CD8+ T cells (47–59 percent) domi-nated over CD4+ T cells (9–19 percent). B cells and macrophages each repre-sented less than 1 percent of the cells in the inflammatory infiltrates. Tissues from humans with chronic chagasic cardiomy-opathy were similarly analyzed. As with mice, T cells were the primary cell type in the inflammatory infiltrates, with a greater proportion of CD8+cells than CD4+. Many of the CD8+ T cells expressed granzyme A. The extent that the tissue lymphocytes are directed toward parasites and para-site antigen as opposed to self-targets (i.e., autoimmunity) has not been fully clari-fied (see “Pathogenesis and Modulation of Immune Function” section).
T cells in T. cruzi infection perform a variety of antiparasitic functions. They provide helper T-cell function by stimu-lating B cells to produce parasite-specific antibody. Nickell and colleagues showed that a T-cell line derived from the spleens of T. cruzi–infected mice was able to induce normal spleen cells to produce parasitic-directed antibodies when stimu-lated in vitro with T. cruzi antigen. The activation of helper T-cell function is sup-ported by the predominance of IgG2a and IgG2b antibodies in T. cruzi infection, which are typical of CD4+ T-cell-depen-dent responses. T cells produce cytokines in T. cruzi infection, which mediate impor-tant antiparasitic functions. CD8+ T cells lyse T. cruzi–infected host cells, which pre-sumably interrupts the parasite life cycle, thus limiting its replication. Interestingly, mice genetically deficient for genes con-trolling perforin or granzyme B–mediated cytolytic pathways had parasitemia and mortality rates similar to wild-type mice, suggesting that cytolytic function, in fact, may not be the protective effector mecha-nism of CD8+ T cells. Wizel and cowork-ers showed that T. cruzi–specific CD8+ T cells produce IFN-γ and TNF-α upon stimulation, thus raising the possibility that these cells mediate their effects by cytokine release.