Pathophysiology of amoebiasis

Few organisms are more aptly named than Entamoeba histolytica, an intestinal protozoan parasite that can lyse and destroy human tissue. Within the past four years, new models of E. histolytica infection have begun to illuminate how amoebic trophozoites cause intestinal disease and liver abscess, and have expanded our understanding of the remarkable killing ability of this parasite. Here, I summarize recent work on the interactions between E. histolytica and human intestine, and between E. histolytica and hepatocytes, and discuss what these studies tell us about the role of inflammation and programmed cell death in the pathogenesis of amoebiasis.     

Patients treated for amebic liver abscess develop cell-mediated immune responses effective in vitro against Entamoeba histolytica

We studied the afferent and efferent cell-mediated immune response in 15 patients treated for amebic liver abscess. Patients had a lower T4 to T8 ratio (1.25 +/- 0.65) compared with age- and sex-matched controls (1.89 +/- 0.44, p less than 0.01) due to a decrease in T4-"helper" cells and an increase in T8-"suppressor" cells (p less than 0.01). The in vitro proliferative response of patient T lymphocytes to the plant mitogen concanavalin A (Con A) was depressed; responses to phytohemagglutinin were not. The proliferative response of patient lymphocytes to an amebic soluble protein preparation (SPP) was greater than the mitogenic response seen in control lymphocytes (mean of 68,300 delta cpm and 22,300 delta cpm, respectively, p less than 0.001), correlated with the T4 to T8 ratio (p less than 0.05) and the duration of time from initiation of antiamebic therapy (p less than 0.01). Supernatants from patient lymphocytes exposed to the amebic SPP activated normal monocyte-derived macrophages to kill virulent axenic E. histolytica trophozoites (p less than 0.001); patient monocyte-derived macrophages activated by Con A-elicited lymphokine could also kill amebae. Finally, when incubated with the amebic SPP for 5 days, T lymphocytes from patients were able to kill virulent amebae (p less than 0.005); patient T lymphocytes not exposed to the amebic SPP or control T lymphocytes incubated for 5 days with the amebic SPP were not cytotoxic to E. histolytica trophozoites. In summary, after cure of amebic liver abscess, specific cell-mediated immune mechanisms develop that are effective in vitro against the parasite.     

An experimental model for amoebic abscess production in the cheek pouch of the Syrian golden hamster, Mesocricetus auratus

A new experimental model was developed in hamsters for amoebic abscess caused by Entamoeba histolytica. E. histolytica trophozoites were cultured in a liquid axenic medium, and then injected intradermally into the cheek pouch of the Syrian golden hamster, Mesocricetus auratus. Inoculation consistently resulted in abscess formation at the site in 20 of 22 (91%) study animals. The amoebic nature of the abscesses was confirmed by light microscopy and histopathologic examination. Abscess formation was maximal at day 12 post-inoculation. Potential applications of this simple and reliable model include further elucidation of the pathogenesis of invasive amoebiasis, studies of the host response to amoebae, and in vivo evaluation of chemotherapeutic agents that show in vitro efficacy against E. histolytica.     

Comparative proteomic analysis of two Entamoeba histolytica strains with different virulence phenotypes identifies peroxiredoxin as an important component of amoebic virulence

Entamoeba histolytica is a protozoan intestinal parasite that causes amoebic colitis and amoebic liver abscess. To identify virulence factors of E. histolytica, we first defined the phenotypes of two E. histolytica strains, HM-1:IMSS, the prototype virulent strain, and E. histolytica Rahman, a strain that was reportedly less virulent than HM-1:IMSS. We found that compared with HM-1:IMSS, Rahman has a defect in erythrophagocytosis and the ability to cause amoebic colitis in human colonic xenografts. We used differential in-gel 2D electrophoresis to compare the proteome of Rahman and HM-1:IMSS, and identified six proteins that were differentially expressed above a fivefold level between the two organisms. These included two proteins with antioxidative properties (peroxiredoxin and superoxide dismutase), and three proteins of unknown function, grainin 1, grainin 2 and a protein containing a LIM-domain. Overexpression of peroxiredoxin in Rahman rendered the transgenic trophozoites more resistant to killing by H2O2 in vitro, and infection with Rahman trophozoites expressing higher levels of peroxiredoxin was associated with higher levels of intestinal inflammation in human colonic xenografts, and more severe disease based on histology. In contrast, higher levels of grainin appear to be associated with a reduced virulence phenotype, and E. histolytica HM-1:IMSS trophozoites infecting human intestinal xenografts show marked decreases in grainin expression. Our data indicate that there are definable molecular differences between Rahman and HM-1:IMSS that may explain the phenotypic differences, and identify peroxiredoxin as an important component of virulence in amoebic colitis.     

Cytopathogenicity of Entamoeba histolytica to human intestinal epithelial cells: inhibition by monoclonal antibodies and sera from amoebic patients.

Interactions of pathogenic Entamoeba histolytica (HM 1) with human intestinal epithelial cells (Henle-407) were investigated. The E. histolytica trophozoites adhered and cytolysed 87% of cultured epithelial cell monolayers. A significant (P less than 0.001) inhibition of cytopathic effect of amoebic trophozoites pretreated with monoclonal antibodies to a 29 kDa surface associated protein suggested utilization of the 29 kDa surface protein in recognition and cytolysis of epithelial target cells. The polyclonal sera from treated patients of amoebic liver abscess and anti-amoebic hyperimmune serum inhibited cytopathogenicity to a greater degree (P less than 0.001) than did the monoclonal antibodies. The data thus suggest involvement of several amoebic molecules in exercising cytopathogenicity to epithelial cells.     

Entamoeba histolytica: induction of cyclooxygenase-2 expression during amoebic liver abscess formation in hamsters (Mesocricetus auratus)

Experimental amoebic liver abscess in hamsters curses with an increase in both, systemic levels of prostaglandin E2 (PGE(2)) and local cyclooxygenase activity in liver microsomes. The cellular source of PGE(2) and the isoform of cyclooxygenase responsible are not completely evidenced. Cyclooxygenase-2 (COX-2) protein and gene expression were demonstrated on macrophages and polymorphonuclear cells as a result of Entamoeba histolytica infection in hamsters at 2, 4, and 7 days postinfection by immunohistochemistry and RT-PCR. E. histolytica trophozoites located in the lesion showed a strong positive signal for COX-2, however the enzyme was not detected in cultured trophozoites by Western blot. Our results indicate that the increment in PGE(2) is the result of COX-2 activity from cells of the reticuloendothelial system and reinforce the possibility that PGE(2) production by enzyme induction in macrophages may be a mechanism by which E. histolytica modulates the host immune response in this parasitic infection.     

Human liver sinusoidal endothelial cells respond to interaction with Entamoeba histolytica by changes in morphology, integrin signalling and cell death

Invasive infection with Entamoeba histolytica causes intestinal and hepatic amoebiasis. In liver, parasites cross the endothelial barrier before abscess formation in the parenchyma. We focussed on amoebae interactions with human hepatic endothelial cells, the latter potentially playing a dual role in the infection process: as a barrier and as modulators of host defence responses. We characterized early responses of a human liver sinusoidal endothelial cell line to virulent and virulence-attenuated E. histolytica. Within the first minutes human cells start to retract, enter into apoptosis and die. In the presence of virulent amoebae, expression of genes related to cell cycle, cell death and integrin-mediated adhesion signalling was modulated, and actin fibre, focal adhesion kinase and paxillin localizations changed. Effects of inhibitors and amoeba strains not expressing pathogenic factors amoebapore A and cysteine protease A5 indicated that cell death and cytoskeleton disorganization depend upon parasite adhesion and amoebic cysteine proteinase activities. The data establish a relation between cytotoxic effects of E. histolytica and altered human target cell adhesion and suggest that interference with adhesion signalling triggers endothelial cell retraction and death. Understanding the roles of integrin signalling in endothelial cells will provide clues to unravel host-pathogen interactions during amoebic liver infection.     

Antisense inhibition of amoebapore expression in Entamoeba histolytica causes a decrease in amoebic virulence

Amoebapores have been proposed to be a major pathogenicity factor of the protozoan parasite Entamoeba histolytica, which is responsible for the killing of target cells. These 77-residue peptides are structural and functional analogues of NK-lysin and granulysin of porcine and human cytotoxic lymphocytes. Inhibition of amoebapore gene expression in amoebae was obtained following transfection with a hybrid plasmid construct (pAP-R2) containing the Neo resistance gene and the gene coding for amoebapore A, including its 5' and 3' untranslated region (UTR) sequences, in reverse orientation under a promoter (g34) taken from one of the E. histolytica ribosomal protein (RP-L21) gene copies. Transfectants of virulent E. histolytica strain HM-1:IMSS, in which the expression of amoebapore was inhibited by approximately 60%, were significantly less pathogenic. Cytopathic and cytolytic activities of viable trophozoites against mammalian nucleated cells, as well as lysis of red blood cells, were markedly inhibited. Moreover, trophozoite extracts of pAP-R2 transfectant displayed lower pore-forming activity and were less potent in inhibiting bacterial growth compared with controls. Notably, liver abscess formation in hamsters by the pAP-R2 transfectant was substantially impaired. These results demonstrate for the first time that amoebapore is one of the pathogenicity factors by which trophozoites of E. histolytica exert their remarkable cytolytic and tissue destructive activity.     

Early interactions of Entamoeba histolytica trophozoites with parenchymal and inflammatory cells in the hamster liver: an immunocytochemical study

We studied the early in situ interactions of live and fixed Entamoeba histolytica trophozoites with hamster hepatic parenchymal and inflammatory cells using immunoperoxidase and immunoelectronmicroscopy. Close contact between trophozoites and endothelial cells and the diffusion of amoebic molecules from trophozoites towards nearby endothelial cells and distant hepatocytes were observed. The inflammatory cells around the amoebae and the remnants of parenchymal cells and hepatocytes located close to the lesion had a positive stain for amoebic molecules. In the amoebae, at the ultrastructural level, molecules were attached to the membranes and inside the vesicles. These molecules were apparently released into the space formed between the parasite and the endothelial cells. The endothelial cells and the nearby and distant hepatocytes captured amoebic molecules, and later they became necrotic. Contrarily, when fixed amoebae were inoculated, amoebic molecules were captured by endothelial cells and polymorphonuclear (PMN) leukocytes, but neither suffered any damage. In this work, we are presenting evidence clearly showing that some molecules of the amoeba can diffuse away long distances causing cytotoxic effects and even necrosis on hepatic cells of hamster liver without the need of the trophozoite being in close contact with the target cells. They also may promote lytic or proinflammatory effects by inducing the secretion of enzymes or cytokines in other nonparenchymal cells, like PMN leukocytes and endothelial cells. Our results suggest that the accepted mechanisms of cytotoxicity by amoebae are not exclusively restricted to the following sequence: adhesion, phagocytosis, and necrosis.     

Virulence of entamoeba histolytica strains from Bombay (India) to rats

The results of intracaecal inoculation experiments on young albino rats with different strains of E. histolytica isolated in cultures from patients with different varieties of clinical amoebiasis, such as acute amoebic dysentery, amoebic liver abscess and chronic amoebic colitis, as well as from asymptomatic infections, have been presented and discussed. Caecal scoring was performed with regard to the condition of the wall of the caecum and the contents of its lumen. Successful infection with virulent amoebae was associated with the presence of considerable ulceration of the caecum, whereas, in infections with avirulent amoebae, the changes were apparently only slight or absent. The virulence and pathogenicity were considerably different for clinical case and symptomless carrier strains of E. histolytica. In the former group, high caecal scores and pathogenicity indices with ulceration of the caecum were noted, whilst in the latter, the amoebae were either virulent or avirulent. Results similar to those for the former group were obtained with two carrier strains only.     

Role of pure and biologically active amoebal RNA in assessment of lymphokines: a report of 55 ALA cases.

Amoebic liver abscess cases (55) were assessed for release of lymphokines (LMIF) using pure and biologically active amoebal RNA of axenic Entamoeba histolytica (NIH: 200) obtained with cesium chloride centrifugation. Lymphokines released by T lymphocytes in response to both amoebal RNA and whole amoebic lysate (WAL) were tested by leukocyte migration inhibition test (LMIT) on blood samples from amoebic liver abscess cases. A significant increase was observed in the release of lymphokine and 100% positivity was observed with amoebal RNA compared to whole amoebic extract with a positivity of only 78%. The difference between means leukocyte migration inhibition of the above two with regards to release of lymphokine was highly significant (P less than 0.001). This shows that patients had high degree of leukocyte sensitization to amoebal RNA of E. histolytica compared to whole amoebic lysate. These findings suggest that the amoebal RNA plays an important role as a potent antigen in the elicitation of cell mediated immune responses in amoebic liver abscess cases.     

Initiation of inflammation and cell death during liver abscess formation by Entamoeba histolytica depends on activity of the galactose/N-acetyl-D-galactosamine lectin

The parasite Entamoeba histolytica colonizes the human intestine causing amoebic colitis and disseminates through the vascular route to form liver abscesses. The Gal/GalNAc lectin is an adhesion protein complex which sustains tissue invasion by E. histolytica. Disruption of the Gal/GalNAc lectin function in engineered parasites (HGL-2 trophozoites) changed the pathophysiology of hamster liver abscess formation. HGL-2 trophozoites produced numerous small inflammatory foci located in the vicinity of blood vessels. The low penetration of HGL-2 trophozoites into hepatic tissue was shown to be associated with weak attraction of neutrophils and macrophages to the infiltrated areas and absence of pro-inflammatory tumour necrosis factor, in contrast to wild type or control vector infections. The low host inflammatory response in HGL-2 infections correlated with a delay in apoptosis of hepatic cells, whereas apoptosis of endothelial cells was not detected. Triggering of apoptosis in both host cell types most likely has a central role in modulating inflammation, a major landmark in hepatic amoebiasis. These data highlight the key role of the Gal/GalNAc lectin in initiation of E. histolytica hepatic infection.     

Elucidation of cellular population and nature of anti-amoebic antibodies in cytotoxicity to Entamoeba histolytica (NIH:200)

Interactions between trophozoites of Entamoeba histolytica and cells of the immune system were studied in vitro by mixing effector cells obtained from normal and immunized guinea pigs with trophozoites in a ratio of 10:1. Crude amoebic extract (CAE) and its highest molecular weight (MW 650,000) fraction (F-I) were used for priming the effector cells in vivo. The effector cells were collected from the spleens of normal and immunized animals. Lymphocytes were prepared by allowing splenic mononuclear cells to adhere to plastic, followed by passage through nylon wool column. There was no significant difference between the killing capacity of mononuclear cells, monocyte depleted mononuclear cells or nylon wool fractionated lymphocytes, indicating that probably monocytes and B-cells were not involved in the cytotoxicity against E. histolytica. The data suggest that effector cells probably belonged to the T-cytotoxic and K-cell category. Both CAE and F-I sensitization could induce cytotoxicity of lymphocytes against trophozoites. Similarly, anti-CAE and anti-F-I sera were found to enhance the killing capacity of effector cells in vitro. The ability of anti-amoebic serum to induce cytotoxicity was found to be independent of complement involvement and resided in IgG but not in IgM.     

Entamoeba histolytica: inflammatory process during amoebic liver abscess formation involves cyclooxygenase-2 expression in macrophages and trophozoites

It has been demonstrated that expression of cyclooxygenase-2 (COX-2) isoform is induced by Entamoeba histolytica in macrophages and polymorphonuclear cells during amoebic liver abscess (ALA) formation in hamsters. Trophozoites present in the lesion were also positive for COX-2 signal. However, no cross reactivity of the anti-COX-2 antibody with protein extract of cultivated trophozoites was found. To clarify if trophozoites are involved in PGE(2) production during ALA development, COX-2 expression was detected by in situ hybridization and RT-PCR in liver tissue from intrahepatically infected hamsters. COX-2 mRNA was in polymorphonuclear cells since 4h postinfection, and subsequently, local macrophages expressed COX-2 mRNA in a similar way. Additionally, a positive signal for COX-2 mRNA expression was detected in E. histolytica trophozoites, suggesting that, in vivo, parasite COX expression may be an important mechanism to promote inflammation.     

Entamoeba histolytica: immunohistochemical study of hepatic amoebiasis in mouse. Neutrophils and nitric oxide as possible factors of resistance

Studies in mice have not rendered conclusive data on cell and humoral factors to support the resistance of this rodent to Entamoeba histolytica infection. In Balb/c and C3H/HeJ mice inoculated with live or fixed trophozoites, we studied the evolution of the hepatic lesion, the kinetics of inflammatory cells, and the participation of some humoral factors in the development of the hepatic amoebic lesion. From the first hour, amoebae were surrounded by neutrophils containing inducible nitric oxide synthase (iNOS); macrophages also expressing iNOS appeared lately, whereas NK cells were not part of the inflammatory infiltrates. On the fourth day, neutrophils, macrophages, T and B lymphocytes, plasma cells, and some NK cells limited the lesions and anti-amoeba antibodies appeared when most parasites had been eliminated. Therefore, the resistance of the mice to E. histolytica probably lies in non-specific immune responses, among which the activation of neutrophils and the production of nitric oxide (NO) may be important amoebicide factors.     

Effect of phosphatidylcholine-cholesterol liposomes on Entamoeba histolytica virulence

Trophozoites of Entamoeba histolytica HM-1:IMSS become less virulent after long-term maintenance in axenic cultures. The factors responsible for the loss of virulence during in vitro cultivation remain unclear. However, it is known that in vitro cultivation of amoeba in culture medium supplemented with cholesterol restores their virulence. In this study, we analyzed the effect of adding phosphatidylcholine-cholesterol (PC-Chol) liposomes to the culture medium and evaluated the effect of this lipid on various biochemical and biological functions of E. histolytica HM-1:IMSS in terms of its virulence. The addition of PC-Chol liposomes to the culture medium maintained the virulence of these parasites against hamster liver at the same level as the original virulent E. histolytica strain, even though these amoebae were maintained without passage through hamster liver for 18 months. The trophozoites also showed increased endocytosis, erythrophagocytosis, and carbohydrate residue expression on the amoebic surface. Protease activities were also modified by the presence of cholesterol in the culture medium. These findings indicate the capacity of cholesterol to preserve amoeba virulence and provide an alternative method for the maintenance of virulent E. histolytica trophozoites without the need for in vivo procedures.     

Detergent dissection of membrane proteins of Entamoeba histolytica and its effect on lymphokine release in in vitro.

Fifty-two amoebic liver abscess cases were assessed for the release of lymphokines (LMIF) using detergent dissected membrane proteins (DDMP) of axenic Entamoeba histolytica (NIH:200) obtained with sodium deoxycholate treatment. Lymphokines release by T lymphocytes in response to both DDMP and whole amoebic lysate (WAL) was tested by leukocyte migration inhibition test on blood samples from amoebic liver abscess cases. A significant increase was noted in the release of LMIF and 100% positivity was observed with DDMP compared to whole amoebic extract with a positivity of 73%. The difference between means of the above two with regards to release of LMIF was found to be highly significant (P less than 0.005). This shows the patients had high degree of leukocyte sensitization to surface antigens of E. histolytica compared to the whole amoebic lysate. These findings suggest that the antigens shed might have important role as a potent antigen in elicitation of CMI response in amoebic liver abscess cases.     

Amoebiasis: diagnosis by aspiration and exfoliative cytology

The present study was undertaken to evaluate the use of fine needle aspiration and exfoliative cytology in the identification of amoebic cysts/trophozoites, and to characterize amoebiasis. The subjects consisted of 15 patients, 11 diagnosed by fine needle aspiration cytology (FNAC) as amoebic abscesses (14 liver and one pulmonary) and four women whose cervical smears contained Entamoeba histolytica cysts or trophozoites. Of 128 ultrasonographically guided FNAC of hepatic lesions over a four year period, 17 were abscesses of which 10 were diagnosed as amoebic. A single case of pulmonary amoebiasis was detected in an 18-year-old male. The case was initially diagnosed as tubercular due to deceptive symptomatology. Three cases of amoebic cysts and one trophozoite were reported on routine cervical smear screening. All four cases were unsuspected for amoebic infection. The disease may easily go undetected unless meticulous screening is exercised, and the search for cysts or trophozoites is made with clear concepts of the morphological characteristics of E. histolytica in mind.