Toxoplasmosis

Toxoplasmosis is the most frequent and best known of the various opportunistic parasitic diseases. In immunocompetent patients, infection is normally asymptomatic or unremarkable. Acquired toxoplasmosis normally becomes latent and parasite persistence in the form of cysts leads to the build-up of a protective immunity. In contrast, congenital toxoplasmosis may cause intra-uterine death, neonatal growth retardation, mental retardation, ocular defects and blindness. Similar degrees of severity are found in immunocompromised patients (Ambroise-Thomas & Pelloux, 1993), notably in AIDS patients with cerebral, often life-threatening, manifestations (Leport & Remington, 1992). Multiple visceral localisations (lung, liver, kidney) are also observed (Rabaud et al. 1996) while retinochoroiditis is less common than in congenital toxoplasmosis, for reasons that are not well understood. These cases of toxoplasmosis in AIDS patients almost always result from the reactivation of an old and unrecognised chronic toxoplasmosis, when the CD4+ lymphocyte count is under 100 μl. The same mechanism may occur in life-threatening toxoplasmosis following immunodepressant therapy following organ transplantation (Couvreur et al. 1992; Derouin et al. 1992; Gallino et al. 1996). Occasionally, infection may not be caused by a reactivation of latent toxoplasmosis (as in AIDS patients) but rather from a primary infection resulting from an organ or tissue transplantation and aggravated by immunodepressant agents (Mayes et al. 1995). Without the currently used preventive measures, this risk would reach approximately 50% of heart transplant cases, 20% of liver transplants, and only 2% of kidney or bone marrow transplants (Joynson, 1999), which well illustrates the preferential sites of Toxoplasma gondii cysts or tachyzoites.

Cryptosporidium and Cyclospora infections

In AIDS patients the Cryptosporidium parvum and Cyclospora cayetanensis parasite load is particularly high. The number of parasites increases through merozoite auto-reinfections emerging from type 1 schizonts or from zygotes (Bonnin et al. 1998). Consequently clinical manifestations are particularly severe, with abundant diarrhoea (up to 17 l/day) leading to serious nutritional and hydroelectric consequences. The most prominent feature of cryptosporidiosis in AIDS is the risk of (possibly haematogenous) dissemination with secondary multiple localisation in the liver, pancreas, stomach, and respiratory tract (Pinel et al. 1998). Nevertheless besides serious complications in patients with severe immunodepression (CD4+<50 μl), less serious (paucisymptomatic) manifestations occur in patients with CD4+ counts >180–200 μl. Lastly, as well as in AIDS, severe disseminated cryptosporidiosis may occur in other cases of immunodepression, particularly those due to cancer, transplants, dialyses (Gascon, Zabala & Iglesias, 1998) as well as in patients presenting with congenital IgG or IgA deficiencies.

Isosporosis and sarcocystosis

Of all the species belonging to the genus Isospora, only Isospora belli may be considered as a truly opportunistic parasite. Very rarely found in immunocompetent subjects (less than 0.2% in Southeast Asian refugees) it is however very frequent in AIDS patients, especially in tropical countries (up to 85% of AIDS patients in Haiti) and causes chronic diarrhoea leading to serious absorption problems (Dei-Cas, 1994). By contrast Sarcocystis bovihominis and S. suihominis, which cause moderate digestive problems in immunocompetent subjects, do not present any particular pathogenicity in patients presenting with AIDS or other immunodeficiencies.

Microsporidioses

Microsporidia are found world-wide and infect vertebrates and invertebrates alike. Approximately 100 different genera and 1000 species are known; in man only 12 species belonging to 7 genera have been described: Enterocytozoon, Encephalitozoon, Nosema, Microsporidium, Trachipleistophora, Vittaforma and Brachiola.

Until 1985 microsporidioses were virtually unknown in human pathology since they were only rarely observed and, for the most part, diagnosis was questionable. It was with the emergence of AIDS that this parasitosis became fully recognised. In immunocompetent patients, microsporidioses are mainly asymptomatic or paucisymptomatic, only rarely causing diarrhoea in travellers or exceptionally ocular or cerebral effects. In AIDS patients, these opportunistic parasitoses are normally severe (Bryan, 1995; VanGool & Dankert, 1995). Localisation and clinical presentation vary according to the parasite species: intestinal or hepatic (Encephalitozoon bieneusi, Encephalitozoon intestinalis), disseminated/systemic with polyvisceral localisation (Encephalitozoon hellem, E. cuniculi and Vittaforma corneae), ocular localisation has also been reported for several species including Microsporidium ceylonensis and M. africanum (Schwartz et al. 1996).

However, even in immunodepressed patients, microsporidiosis could be clinically mild. Unlike toxoplasmosis or cryptosporidiosis, CD4+ lymphocyte counts have no prognostic value in AIDS. Besides AIDS, other severe immuno-depressions before organ transplant (renal, allogenic marrow, heart-lung, liver) can cause severe microsporidioses (Sax et al. 1995; Rabodonirina et al. 1996; Kelkar et al. 1997; Gumbo et al. 1999; Metge et al. 2000).

Leishmaniasis and Chagas Disease

In endemic areas, leishmaniasis is the most frequent opportunistic parasitosis in AIDS after toxoplasmosis and cryptosporidiosis (Desjeux, 1996; Dedet & Pratlong, 2000). In fact AIDS multiplies the risk of visceral leishmaniasis 100 to 1000 fold and increases its severity, particularly in L. infantum infections in south-western Europe (WHO, 1999). Visceral leishmaniasis also increases immunodepression and accelerates viral replication and transition from HIV infection to overt AIDS, at the same time increasing the risk of other opportunistic infections.

Among the AIDS cases reported by the WHO, 90.4% of patients had less than 200 CD4+/μl. Even in AIDS patients, asymptomatic leishmaniasis may be observed in approximately 10% of the cases along with the classical forms comparable to those in immunocompetent patients combining fever, hepatosplenomegaly and pancytopenia (82.4% of the cases) (Marlier et al. 1999). Atypical forms have also been described. These are either paucisymptomatic, corresponding most often to an early diagnosis, or by contrast they feature localisations which are unusual in the immunocompetent host: intestinal localisations with erosive or more rarely tumoral lesions, cutaneous lesions which may be indicative of concomitant visceral dissemination, pleuropulmonary (rare), laryngeal or neuromeningeal forms (extremely rare).

Mortality due to leishmaniasis is high (10–20% during the first episode), nevertheless newer combination antiviral drugs (tri-therapy) are likely to modify the course of the disease by improving the immune response and thus increasing the efficacy of anti-Leishmania treatments (Behre et al. 1999).

No change in frequency and presentation has been reported for African trypanosomiasis in HIV patients, while reactivation of Chagas disease is observed in patients dually infected with HIV and Trypanosoma cruzi (Dedet & Pratlong, 2000).

Strongyloidiasis

In immunocompromised patients, Strongyloides stercoralis may cause hyperinfections due to a considerable increase of the auto-reinfection cycle with a dissemination of L₃ larvae throughout the whole host, and of adult females, eggs, and L₁ larvae outside of the digestive tract. Clinically, digestive symptoms are particularly important (severe diarrhoea, emesis, abdominal pain) with various visceral localisations: intestinal (ileum necrosis, ulcerative colitis, peritonitis), pulmonary, often over-infected with Gram negative bacteria (Gulletta et al. 1998), or neurological. These disseminated malignant strongyloidiases are often lethal. Observed for the first time in mentally retarded children with congenital immunodeficiencies (Yoeli et al. 1963), they are mainly caused by intensive and long-term corticotherapy when treating auto-immune diseases, lymphomas etc. The mechanism of action on the parasites is not well understood, but corticotherapy may accelerate extra-intestinal transformation of L₁ into L₃ larvae. This probably explains only part of the phenomenon because disseminated strongyloidiasis also occurs after reactivation of chronic strongyloidiasis due to the immunosuppressive therapy in transplant recipients, particularly in kidney transplantation (De Vault et al. 1990). This risk is frequent and justifies screening for strongyloidiasis in all patients from endemic strongyloidiasis areas and, if necessary, to cure it before organ transplantation. In some very exceptional cases, the graft itself could be the source of contamination (Palau & Pankey, 1997).

The influence of AIDS is still controversial. Nevertheless, disseminated malignant strongyloidiases have been observed more frequently in patients infected by HTLV1 which, for unknown reasons, seems to facilitate these disseminated strongyloidiasis more than HIV (Neisson-Vernant & Edouard, 1990).

Scabies

In immunodeficient patients and particularly in AIDS patients, scabies has a special clinical picture, with widespread or localised hyperkeratotic generally non-pruritic skin lesions. In some cases these crusted lesions are pruritic and resemble Darier's disease or psoriasis (Schlesinger, Oerlich & Tyring, 1994). This ‘Norwegian’ scabies is extremely contagious and undiagnosed cases can be a source of nosocomial infections (Guggisberg et al. 1998). Apart from AIDS, other immunodeficiencies, particularly those expected in old patients, can also induce this type of complicated scabies.

Other parasitic diseases and opportunistic mycoses

This already long and varied list could be completed with other parasitic diseases which are rarely truly opportunistic, but the clinical symptoms of which could be modified during major immunodeficiencies of pathological or therapeutic origin. This is the case of Giardia duodenalis infection (which is worsened by IgA deficiency), of Blastocystis hominis infection (which could be the cause of severe diarrhoea in AIDS patients), or of Babesia divergens or B. microti infections (which only occur in splenectomised patients).

One could also add numerous opportunistic mycoses: pneumocytosis, crypotococcosis, invasive aspergillosis, oro-pharyngeal or disseminated candidiasis. Although in several countries, including France, mycology and parasitology are linked, we will not consider these fungal infections within this review.

The various immunodeficiencies and their consequences

Specific anti-parasitic immunity is produced by numerous and complicated mechanisms mainly implicating three specific cellular populations, phagocytes, and both B and T lymphocytes. For each parasitosis, it is usually possible to demonstrate that one of these factors plays an essential role. Conversely, it has been shown that a given immunodepression would promote a specific type of opportunistic parasitosis. Nevertheless, this is not universal since anti-parasitic immunity almost always results from a combination of various specific and non-specific factors.

Immunodeficiencies act in a predictable way, notwithstanding the diversity of opportunistic parasitoses. They do not really increase the intrinsic pathogenicity of parasites (except maybe in the case of cryptosporidiosis), but they promote their replication, either at the onset (strongyloidiasis for example), or after reactivation of a quiescent infection (toxoplasmosis).

In immunocompetent patients, toxoplasmosis becomes and remains latent due to several complex immunological mechanisms, some of which are not well understood. Humoral immunity plays a role since antibodies specific to the major surface protein of tachyzoites (SAG1) inhibit both the attachment of the parasite to the host cell receptors and then its infection. Nevertheless cellular immunity remains the main component of this immunity. CD4+ lymphocytes play an important but still controversial part even if it is their count, in AIDS patients, which permits the definition of the alert threshold (<100 CD4+/μl). But CD4+ lymphocytes act synergistically with CD8+ lymphocytes which are the key elements of anti-toxoplasmic immuno-protection (Kasper & Buzoni-Gatel, 1998). In the course of AIDS, severe toxoplasmosis only develops after several years, specifically when CD8+ lymphocyte counts begin to decrease. After stimulation by toxoplasmic antigens and with IL-12 acting as a mediator, these cells secrete γ Interferon (IFN-γ) which acts simultaneously with α Tumour Necrosis Factor (TNF-α) in the production of NO. Other cytokines (IL-4, IL-6, IL-7, IL-15) may also be implicated in these protective mechanisms involving other cell populations (dendritic cells). Besides this specific immunity, other phenomena are essential for the protection against toxoplasmosis such as natural killer cells (NK) which, when activated by IL-12 secreted by activated macrophages, are one of the main sources of IFN-γ. HIV-1 also has a direct effect, by reducing the phagocytic and parasiticidal function of macrophages (Biggs et al. 1995).

These mechanisms, independent of CD4+ or CD8+ lymphocyte action, may account for the very paradoxical case of malaria (Butcher, 1992). Indeed, Plasmodium falciparum is a sporozoan like Toxoplasma gondii and Cryptosporidium parvum. Nevertheless, malaria is seldom worsened by AIDS, in contrast to toxoplasmosis and cryptosporidiosis (Leaver, Haile & Watters, 1990). This is all the more unexpected since other viral co-infections induce very severe complications in the course of malaria (Plasmodium falciparum and Epstein-Barr virus co-infections in Burkitt's lymphoma). In fact, apart from humoral and mainly cellular immunity (CD4+ Th₁), protection against malaria is due to macrophages which, after having been activated by plasmodial antigens, produce H₂O₂, NO, and TNF-α (which is a cytotoxic factor acting on P. falciparum). This may explain the lack of evidence for an interaction between HIV infection and malaria, although the risk of malaria during pregnancy was higher in HIV positive than in HIV negative women in Malawi (Verhoeff et al. 1999). Nevertheless most studies did not show this interaction between HIV and malaria, but they had several methodological limitations. Conversely, recent data shown that P. falciparum malaria may increase the HIV-1 viral load (Hoffmann et al. 1999). This may at least partially account for the higher number of AIDS cases in HIV infected patients, in tropical or sub-tropical countries more than in countries free of malaria (Chandramohan & Greenwood, 1998).

For cryptosporidiosis, circulating antibodies are partially responsible for the control of the infection, as shown by the severity of this parasitosis during congenital IgA and IgG deficiencies. Nevertheless, this is also the case for other opportunistic protozoan infections; cellular immunity is the most effective protective mechanism, particularly through CD4+ T lymphocytes. Below a determined value (50 CD4+/μl) the CD4+ count is also one of the main alarm signals concerning the risk of severe cryptosporidiosis during AIDS. In immunocompetent patients, CD4+ T lymphocytes in Peyer's patches or in the intestinal lymph nodes are activated by parasite antigens trapped by cells presenting antigens. These lymphocytes then migrate to the intestinal mucosae where they synthesise various cytokines, particularly IFN-γ which acts directly on enterocytes and can activate macrophages and lymphocytes of the mucosae. Apart from IFN-γ, other cytokines such as TNF-α and IL-8, along with chemokines, play an important part in the protective immunity against Cryptosporidium parvum. The role of CD8+ T lymphocytes is more controversial and is probably less important than in toxoplasmosis. In cryptosporidiosis, a decrease of humoral immunity may also cause important consequences with a drop in circulating specific antibody titres which may neutralise a cryptosporidial toxin with a pathophysiological role.

The problem is even more complicated in the leishmaniases. Indeed, as shown in several studies, there is a vicious circle in which visceral leishmaniasis and AIDS mutually reinforce their effects. These interactions are probably partially explained by the fact that Leishmania and HIV can infect and multiply in the same type of cells (macrophages). Visceral leishmaniases increase the HIV trans-activation in chronically infected monocytes and in CD4+ T lymphocytes. These parasitoses decrease the cytotoxic HIV-specific effect in CD8+ T and promote the transition from HIV-infection to AIDS-disease, with an increased risk of other opportunistic infections, particularly by endocellular bacteriae (mycobacteriae). Conversely, HIV deactivates the macrophage functions, depresses the activity of Th₁ cells, enhances the activity of Th₂ cells, and inhibits phagosome–lysosome fusion (Wolday et al. 1999). Thus, HIV can enhance intracellular growth then the dissemination of Leishmania, and considerably increase the severity of visceral leishmaniasis.

The exact mechanisms by which malignant disseminated strongyloidiases are caused during various types of immunodeficiencies are not well understood. In particular, the role of AIDS is still controversial and so are the reasons why HTLV-1 may facilitate the Strongyloides stercoralis multiplication and dissemination more than HIV (Neisson-Vernant & Edouard, 1990). Apart from various immunological factors, some direct mechanisms may also cause severe disseminated strongyloidiasis, as may be the case during massive corticotherapy with a direct influence of corticoids on the parasite.

Finally, in Norwegian scabies, the data are rare, often incomplete, and mainly correspond to very variable immunological contexts, including physiological, gerontological and mild immunodeficiencies. We know practically nothing about the immune mechanisms which permit proliferation of an ectoparasite such as Sarcoptes scabiei and which considerably increase the severity and the transmissibility of scabies.

Genetic mechanisms

Immunological mechanisms have been pointed out as having a key role in triggering opportunistic parasitoses, especially since it is possible to determine risk circumstances in some cases (AIDS for example) and even to assess that risk with an alert threshold (CD4+ lymphocyte count).

Even if they are very important, these immunological causes are not the only ones. In the case of toxoplasmosis in patients with a latent toxoplasmosis (positive toxoplasmic serology), the onset of cerebral toxoplasmosis in AIDS patients may be determined by a CD4+ lymphocyte count less than 100 CD4+/μl. Nevertheless, this alert threshold has a relative value and should not be wrongly interpreted since very severe toxoplasmic reactivations are observed only in 30% of AIDS patients with a positive toxoplasmic serology, even with a CD4+ count less than 100 μl. These differences have been tentatively explained by pathogenicity specific to some strains of Toxoplasma gondii (Gross et al. 1997). Indeed, genomic virulence markers have been identified (Guo, Gross & Johnson, 1997) and for some, located on chromosome VIII (Howe, Summers & Sibley, 1996), but these genotype features may be modified by various mechanisms (Frenkel & Ambroise-Thomas, 1997). Isoenzyme studies have completed these data by identifying particular zymodemes which, associated with genotypes, allow better characterisation of the strains (Cristina et al. 1995). Some profiles are particularly associated with severe toxoplasmoses, but this relationship is especially true for experimental infections in the murine model. The situation is not as clear concerning human opportunistic toxoplasmoses for which neither the features of the parasitic strain nor the intensity of immunodepression can explain the severity or the mildness of the disease. The host's genetic predisposition features also play a role (may be essential?) (Brown et al. 1995). These features are gradually becoming better understood and are related to genes belonging to the class II HLA group (Mack et al. 1999) or to genes independent of the Major Histocompatibility Complex (Suzuki et al. 1996).

In cryptosporidiosis as well as in toxoplasmosis, features directly linked to the parasite may determine the severity of the disease. Indeed, thanks to the techniques of molecular biology, various genotypes able to infect man have been identified in Cryptosporidium parvum. Four of these genotypes are found in AIDS patients (Pieniazek et al. 1999). Three of these are also found in animals (zoonotic cryptosporidiosis) whereas the type 1 genotype is strictly human (anthroponotic cryptosporidiosis) and could be linked to the most severe human infections. The fact that this ‘‘human’’ genotype exists shows that C. parvum has gone through a recent biological evolution, with a loss or modification of its host specificity allowing it to reproduce only in the human host. This phenomenon probably comes from parasite mutations and, in humans, from selection due to adaptative pressure, may be facilitated by major immunodeficiencies. The emergence of human cryptosporidiosis has come about after the appearance of AIDS and the most active immunosuppressant treatments. By neutralising the immunological mechanisms partially responsible for the species barrier, these immunodepressions have allowed the parasite to reproduce intensively in humans, notably due to a sexual cycle of auto-reinfestation. This is what may have promoted the emergence and the reproduction of variants specifically (?) adapted to parasitism in man. Of course, the mechanisms of this emergence are not only immunological and the great number of human cryptosporidiosis cases reported during various epidemics (400000 cases in Wisconsin, in the USA, in 1993) must also have promoted this evolution. Finally, among the factors of clinical severity, host related genetic factors also play a part. This was demonstrated by experimental infections in the murine model (Davami, Bancroft & McDonald, 1997) and very probably takes place in man, even if the genes responsible have not been identified yet.

The same genetic predisposition factors, in the parasite and in the host, also play a role in the leishmaniases in immunodepressed patients. Thus, isoenzymatic analysis of more than 250 isolates of Leishmania infantum has identified 18 different parasitic zymodemes, 10 of which are found only in HIV co-infections. Thus genetic predisposition is probably a common mechanism in the onset of opportunistic parasitoses, even if it has not been proved in all the cases, especially for opportunistic parasitoses with helminths (strongyloidiasis) or Acaridae (Norwegian scabies).