Amebiasis their virulence differences in infected humans. The unclear pathogenic status of E. histolytica was noted

Transcription

1 CLINICAL MICROBIOLOGY REVIEWS, Oct. 1992, p /92/ $02.00/0 Copyright 1992, American Society for Microbiology Vol. 5, No. 4 Amebiasis DAVID A. BRUCKNER Department ofpathology and Laboratory Medicine, UCLA Medical Center, Los Angeles, California INTRODUCTION EPIDEMIOLOGY LIFE CYCLE...o357 Trophozoite.357 Precyst Cyst Metacyst Metacystic Trophozoite PATHOGENICITY AS DETERMINED BY ZYMODEMES PATHOGENESIS AND PATHOLOGY......o.oooo.oo360 Asymptomatic Infection o Symptomatic Infection..361 DIAGNOSIS THERAPY Asymptomatic Patients.364 Symptomatic Patients Extraintestinal Disease ooooooo..oooo. 364 Follow-Up o364 SUMMARY AND FUTURE DIRECTIONS.364 REFERENCES INTRODUCTION Entamoeba histolytica is one of six parasitic amebae of the genus Entamoeba that are known to infect humans (24, 38). Entamoeba coli, E. gingivalis, E. moshkovskii, and E. hartmanni are not associated with pathologic sequelae, but E. poleckh and E. histolytica are. L6sch in 1875 was the first to describe the trophozoite form of E. histolytica and the pathology associated with the infection in a patient from St. Petersburg, Russia. Although Losch was able to infect a dog with organisms obtained from the patient, he was not able to mimic the disease produced in humans and failed to recognize the relationship (59). Councilman and LaFleur (26), in a classic monograph, described the clinical and pathological evidence of the association of E. histolytica with dysentery and liver abscesses. Quincke and Roos in 1893 described the cyst form, and Schaudinn in 1903 named E. histolytica and differentiated it from E. coli (59). Early literature used the genus names Endamoeba and Entamoeba to describe E. histolytica. In 1954, the International Commission on Zoological Nomenclature ruled that E. coli was to be the type species and that Entamoeba was to be used in place of Endamoeba to describe E. histolytica. Walker and Sellards (154) used prisoners in the Philippines to carry out controlled experiments with E. histolytica. They noted that (i) transmission was most likely accomplished by cysts, not trophozoites; (ii) asymptomatic carriers were probably reservoirs and responsible for transmission; (iii) there were differences between individuals' disease risk; and (iv) there were differences in organism strain virulence. Boeck and Drbohlav (11) were the first to successfully cultivate E. histolytica by using Locke's egg serum medium. Other media frequently used to culture pathogenic strains include those described by Robinson (123), Balamuth (3), and Jones (54); however, the first axenic cultivation was 356 accomplished by Diamond (32, 33). Culture techniques have greatly aided our understanding of E. histolytica strains and their virulence differences in infected humans. The unclear pathogenic status of E. histolytica was noted early in the literature with the terminology "small and large race" E. histolytica. Burrows (19, 20) renamed the small race E. hartmanni on the basis of morphological criteria. Although clinical symptoms and pathological sequelae were known to vary considerably in areas where E. histolytica infection was endemic, it was not until reports of infections in homosexuals appeared that the status of nonvirulent and virulent strains took on renewed significance (44). EPIDEMIOLOGY E. histolytica infections occur worldwide but are more prevalent in the tropics. It has been estimated that approximately 480 million people, or 12% of the world's population, are infected and that annual mortality is 40,000 to 110,000 persons (45). The prevalence of the infection differs from one area to another, as does the severity of disease from patient to patient. Differences in prevalence figures often depend on the detection methods used and the number of fecal examinations done. About 10% of those infected every year have clinical symptoms (155). Of the approximately 48 million persons with clinical symptoms, 80 to 98% have symptoms related to the intestinal mucosa (diarrhea or dysentery); however, in only 2 to 20% of the population with clinical symptoms do amebae invade beyond the intestinal mucosa. Evidence of recurrence of invasive colitis or amebic abscess is unusual. DeLeon (27) monitored more than 1,000 patients with amebic liver abscess for 5 years and found a recurrence rate of 0.29%. Humans are the major reservoir of infection with E. histolytica, although natural infections in macaque monkeys

2 VOL. 5, 1992 AMEBIASIS 357 TABLE 1. Characteristics of E. histolytica in permanently stained smears Characteristic Trophozoite Cyst Size (diam or length)a 10-60,um; usual range, p.m Atm; usual range, p.m Shape In wet preparation, hyaline, fingerlike Usually spherical pseudopodia; motility may be rapid Nucleus (no. and visibility) 1; difficult to see in unstained preparations Mature cyst, 4; immature cyst, 1 or 2 Peripheral chromatin Fine granules, uniform in size and usually Peripheral chromatin present; fine, uniform evenly distributed; may appear beaded granules evenly distributed Karyosome Small, usually compact; centrally located but Small, compact; usually centrally located but may be eccentric occasionally eccentric Cytoplasm Finely granular, "ground glass" appearance Finely granular Inclusions Noninvasive organisms may contain bacteria; Chromatoidal bodies may be present; usually presence of erythrocytes diagnostic elongate, blunt, rounded, smooth edges, may be round or oval; the more mature the cyst, the less chance chromatoidal bodies will be present a The organisms will be to 1.5 pm larger in wet preparations than organisms on permanently stained smears. and pigs have been reported (50). Ingestion of food and drink contaminated with E. histolytica cysts from human feces and direct fecal-oral contact are the most common means of infection. Transmission of E. histolytica by water is common in Third World countries, where much of the water supply for drinking is untreated. The use of human feces for fertilizer is also an important source of infection (81). Only the cysts are infective, and most cases originate from asymptomatic human carriers who are cyst passers. The fact that both pathogenic and nonpathogenic isolates can produce cysts may help explain the propensity for family clustering in the transmission of this organism (41). Depending on environmental conditions, cysts may remain viable for as long as 3 months. Viable cysts in water can be destroyed by hyperchlorination or iodination (57, 81). Recognized high-risk groups include travelers, immigrants, migrant workers, immunocompromised individuals, individuals in mental institutions, and sexually active male homosexuals. E. histolytica has been found in as many as 32% of homosexual men in North America (1). Of special interest is the rarity of invasive disease in homosexual men and in patients with AIDS and AIDS-related complex. Almost all the strains of E. histolytica isolated from human immunodeficiency virus-positive and homosexual populations have, through isoenzyme analysis, been classified as nonpathogenic (1, 18, 44, 157). Human immunodeficiency virus type 1 antigens, possibly resulting from phagocytosis of infected host cells, have been detected within E. histolytica; however, transmission of human immunodeficiency virus type 1 from E. histolytica to human cells is not possible (17). The risk factors contributing to predisposition to infection and the factors contributing to increased severity of disease are primarily speculative at present. There are a few publications concerning whether individuals are more readily colonized by nonpathogenic or pathogenic isolates or whether there are virulence differences in pathogenic isolates (34, 58, 95, 97, 98, 124). Evidence supports the view that there are two morphologically identical forms of E. histolytica, one nonpathogenic and the other pathogenic. This information will be discussed in the sections on pathogenicity as determined by zymodemes and pathogenesis. LIFE CYCLE The life cycle of E. histolytica was extensively described by Dobell (35) and includes trophozoite, precyst, cyst, metacyst, and metacystic trophozoite stages. Lushbaugh and Miller (75) have reviewed the ultrastructure of E. histolytica obtained by scanning and transmission electron microscopy. Trophozoite Trophozoites vary in size from 12 to 60,um in diameter, with larger forms found in tissue and smaller forms found in asymptomatic carriers (45). The large tissue forms may result from increased cellular activity, as shown by increases in the DNA and RNA content of the amebae (51, 77). Depending on environmental conditions, motility and pseudopod formation are rapid and unidirectional. However, movement rarely occurs in a straight line. The ectoplasm is clear, and the endoplasm is granular and may contain bacteria and/or erythrocytes in various stages of digestion. On permanent stained slides, the trophozoite nucleus is spherical and one-fifth to one-sixth the size of the trophozoite. The inner surface of the nuclear membrane is evenly lined with a rim of delicate chromatin. A small, compact karyosome is generally located near the center of the nucleus. On permanent stained smears, the organism size may be reduced by 1 to 1.5,um because of shrinkage by the staining reagents (38) (Table 1). Precyst In the precyst stage, the trophozoite becomes approximately the same size as the cyst. The cytoplasm is cleared of all food inclusions but usually contains diffuse glycogen deposits and occasional chromatoid bodies. The chromatoid body is composed of ribosomes; however, its functional role is uncertain (68, 81, 111). The precystic form is uninucleate, and the enlarged nucleus contains a karyosome that is more or less eccentric. Cyst A cyst wall develops around the precystic form, and the single nucleus divides to form the mature quadrinucleate stage. Generally, glycogen and chromatoidal material disappear as the cyst matures. Once immature cysts are cooled to room temperature, they cease to develop, and all the singleand double-nucleated cysts degenerate and die (35). On permanent stained slides, cysts ranges from 8.5 to 19 p.m in diameter. Again because of shrinkage caused by the dehydration reagents, cysts may be 1 to 1.5 p.m smaller than organisms seen on wet preparations. They are usually spher-

3 358 BRUCKNIER CLIN. MICROBIOL. REV. HK Origin I - - m - - m m m m - - m - - m - m m m ME mm m m ---m OriginI 16 GPI s Origin PGM Origin 16 zi a a- l e - - m m m I- m m m m m minm - - m m - - = x x Zymodeme 2 > 5,>, X X X X X X R FIG. 1. Zymodemes of E. histolytica identified by using GPI (EC ), L-malate:NADP+ oxidoreductase (ME; oxaloacetate decarboxylating) (EC ), PGM (EC ), and HK (EC ). (Adapted from reference 131 with permission of the publisher.) Pathogenic zymodemes include II, II alpha, VI, VII, XI, XII, XIII, XIV, XIX, and XX. Nonpathogenic zymodemes include I, III, III alpha, IV, V, VIII, IX, X, XV, XVI, XVII, XVIII, and (not shown) XXI. ical and contain four nuclei. The nuclear membrane is uniformly lined with peripheral chromatin. The karyosome is small and usually centrally located within the nucleus. Metacyst During the process of excystation, the encysted ameba containing four nuclei becomes very active, separating from the cyst wall. The quadrinucleate ameba escapes from the cyst wall through a tiny pore, and the nuclei clump together. Metacystic Trophozoite Outside the cyst, the nuclei within the quadrinucleate ameba begin to separate from the surrounding cytoplasm and undergo division to form eight uninucleate metacystic trophozoites. The resulting trophozoites are always smaller (8,um) than the trophozoites seen in the bowel of an infected human. The metacystic trophozoites continue to feed and grow, finally achieving the size normally associated with the trophozoite (Table 1). PATHOGENICITY AS DETERMINED BY ZYMODEMES In 1925, the pathogenic nature of E. histolytica was questioned by Brumpt, who thought that there were two morphologically similar strains of E. histolytica, one pathogenic and the other nonpathogenic (cited in reference 43). According to epidemiological studies, only 10% of the world's population and a few homosexual males and AIDS patients infected with E. histolytica develop disease (45, 68a, 157). One explanation for this percentage is that most people are infected with nonpathogenic strains (43). Reeves and Bischoff (122) were the first investigators to use enzyme electrophoresis to differentiate "typical" from "atypical" E. histolytica. Typical E. histolytica were those organisms, including those associated with human disease, requiring m m m m a culture temperatures of 33 to 39 C, whereas atypical E. histolytica organisms could adapt to lower temperatures. This concept was further explored by Sargeaunt and associates in a series of studies designed to differentiate E. histolytica from other intestinal amebae and to distinguish pathogenic strains (36, ). A zymodeme is defined as a population of amebae in which the electrophoretic mobilities of specific enzymes differ from those in similar populations. The enzymes used to classify E. histolytica include glucosephosphate isomerase (GPI), hexokinase (HK), L-malate:NADP+ oxidoreductase, and phosphoglucomutase (PGM) (Fig. 1, Table 2). Zymodeme analysis is performed by observing the migration pattern of isoenzyme bands on a thin-layer starch gel or in polyacrylamide gel electrophoresis (63, 82). Only one band is found with L-malate:NADP+ oxidoreductase, whereas GPI and PGM show multiple bands that are labeled a through B. Blanc and Sargeaunt (10) have shown that the concentration of starch in the growth medium can influence the appearance of zymodeme patterns. Expression of y and 8 bands of GPI and PGM was observed more often when higher concentrations of starch were present in the culture media. HK is characterized by two bands, one migrating slowly and the other migrating faster. Pathogenicity is associated with the presence of a fast-migrating HK band, a PGM I band, and no PGM a band. The only known exceptions are zymodeme XIII, which, although considered a pathogenic zymodeme, lacks a fast-migrating HK band, and zymodeme XXI (112). Organisms belonging to zymodeme XXI were recently isolated from a male homosexual. On the basis of clinical history and serological tests, the isolate was considered nonpathogenic. The zymodeme pattern was unusual in that the isolate had a PGM a band and a fast-migrating HK band, a pattern which is often associated with pathogenic zymodemes. Sargeaunt (130, 131) has shown a positive correlation between zymodemes and the clinical histories and serologies

4 VOL. 5, 1992 AMEBIASIS 359 TABLE 2. Differential characteristics of zymodeme patterns Isoenzyme band Characteristics or group Nonpathogenic isolates Pathogenic isolates HK Fast migration Slow migration PGM Beta Alpha L-Malate:NADP+ None None oxidoreductase GPI None None Zymodeme groups II, II gamma, VI, VII, XI, XII, XIII,' I, III, III alpha, IV, V, VIII, IX, X, XV, XVI, XIV, XIX, and XX XVII, XVIII, and XXI6 a Zymodeme XIII is pathogenic but lacks the fast-migrating HK band. b Zymodeme XXI is nonpathogenic but has a fast-migrating HK band. of more than 6,000 pathogenic and nonpathogenic E. histolytica patient isolates. In order to perform zymodeme studies, nonaxenic culture techniques must be used to isolate amebae from either feces, exudates, liver aspirates, or other suspect material. The media most frequently used are those described by Robinson (123) and Jones (54). Generally, three to nine subpassages requiring at least 2 days per subpassage are needed before enough organisms are obtained for enzyme electrophoretic analysis. Although the cultures of amebae from patients are xenic, containing multiple bacteria, the bacterial enzyme bands migrate faster than do those of E. histolytica. However, it is still necessary to run bacterial and amebic controls when axenic cultures are not obtained. At present, zymodeme analysis is not easily incorporated into routine clinical laboratory work because of the expertise required, technical difficulties, and cost. Although microscopic diagnosis does not address the issue of pathogenicity, it is still the method of choice for detecting amebic infections. The ability to culture organisms from specimens determined by microscopic examination to be positive is limited to approximately 50% (112, 144). By enzyme electrophoresis, 23 zymodemes of E. histolytica have been identified (Fig. 1). Of these, nine have been associated with pathology in the host. These include isolates from stool in which the trophozoites contained ingested erythrocytes and isolates from intestinal ulcers and liver abscess material. Thirteen zymodemes have been obtained from asymptomatic patients passing cysts. The zymodeme analysis data include information on isolates obtained from various areas of the world (47, 96, 130, 132). Even 9 years after the organisms were first isolated, Sargeaunt and co-workers have not detected changes in isoenzyme patterns with continued and repeated testing of isolates. Interestingly, mixtures of nonpathogenic zymodemes but no mixtures of pathogenic zymodemes have been found in the same specimen from the same patient. However, two pathogenic zymodemes have been detected from two specimen sources (stool and liver aspirate) from the same patient (132). When specimens are cultured for E. histolytica, pathogenic strains always outgrow nonpathogenic strains and all other amebae (112). Clinically, patients who harbor pathogenic zymodemes generally have high antibody titers, whereas individuals with nonpathogenic zymodemes are either seronegative or have low antibody titers (40, 135). Asymptomatic carriers of pathogenic zymodemes and carriers of nonpathogenic zymodemes have been found in many instances to self-cure within 1 to 9 months (40, 44, 93). On the basis of this evidence, it has been proposed that patients who have E. histolytica in their stool specimens but no clinical signs of infection and no antibody titer need not be treated (40, 44, 66, 68a, 112). Jackson and Gathiram (52) conducted a seroepidemiological study in which they found asymptomatic individuals who were serologically positive but harbored nonpathogenic zymodemes. Whether these individuals were previously treated for invasive amebiasis was not addressed. It has not been established whether the zymodeme patterns reflect genotypic or phenotypic traits. Sargeaunt (131) proposed that pathogenic and nonpathogenic E. histolytica isolates represent distinct subspecies, whereas Mirelman and co-workers and others (12, 14, 86, 87, 90, 97) have presented evidence that zymodeme patterns may be a phenotypic trait and that these patterns could change under environmental influences. If the latter were true, treatment of only symptomatic patients or those who have a serological response could pose potential public health problems if asymptomatic cyst passers who harbor pathogenic zymodemes are not treated; they could serve as an important reservoir for transmission. The reliability of differentiating between pathogenic and nonpathogenic strains of E. histolytica on the basis of phenotypic expression of isoenzyme patterns and the stability of the isoenzymes has been questioned (14, 86, 87, 97). Many authors have published information concerning the influence of bacteria on the virulence of amebae (7, 14, 94, 97, 107, 159, 160). During the process of axenization, Andrews et al. (2) and Mirelman et al. (90) converted a nonpathogenic zymodeme to a pathogenic one in culture. This change was manifested not only by changes in the mobilities of the HK and PGM bands but also by changes in in vitro and in vivo pathogenicity. The isolate was capable of destroying tissue culture monolayers and inducing liver abscesses in hamsters. The ameba-bacterium association and the influence of bacteria on the virulence of E. histolytica have been reviewed by Mirelman (88). The ability to change from a nonpathogen to a pathogen or vice versa does not appear to be a universal trait among E. histolytica isolates (87). Some isolates do not undergo pathogenic conversion. Mirelman suggested that the pathogenic state of E. histolytica is interchangeable and that virulence is expressed only under unique growth conditions. However, the mechanisms or conditions that induce these changes are unknown. Through the use of polymerase chain reaction, Mirelman et al. (89) have shown that nonpathogenic isolates contain extrachromosomal circular DNA that hybridizes with DNA probe P-145, which has been used to characterize pathogenic isolates. They suggested that the extrachromosomal circular DNA characteristic of pathogenic isolates was induced to amplify when changes occurred in bacterial

5 360 BRUCKNER flora in the growth medium. If the pathogenic status of an isolate can readily change because of environmental influences, then all E. histolytica isolates would have to be considered potential pathogens. Sargeaunt and co-workers (129, 133) and Blanc et al. (9) have shown that genetic transfer between pathogenic strains can occur both in vivo and in vitro, with new zymodeme patterns different from those of the parental strains being detected. The possibility that mutation is responsible for the zymodeme pattern changes was believed to be remote because the frequency of mutation was postulated to be extremely low (9). Recent work strongly supports the conclusion that there are two species of E. histolytica, one nonpathogenic and one pathogenic (24, 25, 102, 120, 144, 145, 147, ). Genetic studies do not support the genotype interconversion theory but rather provide considerable evidence that two stable genotypes exist (7, 124). Nevertheless, issues of mutation, changes in the cell surface that affect virulence, and whether pathogenic forms have a nonpathogenic stage in the intestinal tract need to be addressed. Recently, Strachan et al. (144) have used immunofluorescence methods with monoclonal antibodies to the HK isoenzyme to differentiate pathogenic from nonpathogenic amebae. Others have shown that monoclonal antibodies directed against membrane protein or the galactose-specific adherence lectin can be used as a rapid method of distinguishing pathogenic from nonpathogenic isolates after the organisms have been cultured (101, 144, 150). DNA probes that selectively hybridized to DNA from either pathogenic or nonpathogenic E. histolytica have been developed (39, 149). Use of monoclonal antibodies and DNA probes was as effective as isoenzyme electrophoresis for determining the pathogenicity of the isolate. However, neither technique currently allows one to determine the zymodeme pattern, which may be useful for epidemiological purposes, and the question still unanswered is whether E. histolytica can change its pathogenic status as a result of environmental changes. All of these studies were performed with organisms from culture. PATHOGENESIS AND PATHOLOGY E. histolytica is unique among the intestinal amebae parasitizing humans because it is able to invade tissue. After the host ingests cysts, metacystic trophozoites escape from the metacyst in the gastrointestinal tract and colonize the cecum. At present, the mechanisms of colonization and pathogenicity are not well understood (43). Bacterial flora have always been thought to have a direct role in the virulence of E. histolytica and its ability to colonize the host (14, 86, 106). Clinical syndromes associated with amebiasis vary with the host and the organism. Syndromes may range from an asymptomatic infection to a disseminated fatal disease. When the infection disseminates to extraintestinal sites, it is found most frequently in the right lobe of the liver (113). Research on the mechanisms of pathogenicity has evolved along two pathways: one considers pathogenesis to be dependent on export of soluble toxins, and the other considers it to be dependent on cellular contact (31). It has been postulated that amebae can export pore-forming proteins (amebapores) that form aqueous pores in target cell surface membranes and also produce cytotoxic products that are released from the organism and cause the initial cytotoxic effect on the host cell (62, 74, 76, 115, 161). This poreforming peptide was recently purified by Leippe et al. (69). CLIN. MICROBIOL. REV. Recent studies have focused on contact-dependent killing by the trophozoite, including adherence, extracellular cytolysis, and phagocytosis. Host cell leukocytes also play a major role in pathogenesis (45, 152, 158). A number of E. histolytica adherence receptors have been identified, but the galactose-specific lectin receptor has been the most thoroughly studied and is thought to be responsible for mediating attachment to colonic mucus and colonic epithelial cells (71, 78, 79, 83, 101, 104, 116). Human colonic epithelial cells and mucus contain large numbers of galactose or N-acetylgalactosamine residues (23). Incubation of E. histolytica with colonic mucin prevents trophozoites from attaching to and killing target cells such as those in the intestinal epithelium (23). E. histolytica also enhances mucus secretion, which correlates with its pathogenicity (22). More virulent strains stimulated mucus secretion and depleted goblet cells of mucin, thereby making epithelial surfaces more vulnerable to invasion. The binding to mucin was reversed by the addition of galactose and was completely inhibited by monoclonal antibodies directed against the amebic lectin receptor (23, 103). Monoclonal antibodies directed against the galactose-specific lectin have also been found to enhance adherence (102, 105). Monoclonal antibodies to the galactose-specific lectin receptor have been used to differentiate pathogenic and nonpathogenic strains, and the primary structure of this lectin has been sequenced (80, 102, 148). It has been noted that a galactose-specific lectin receptor is present on nonpathogenic zymodemes; however, it is antigenically altered (102). The galactose-specific lectin receptor on pathogenic strains is a potential target for therapy or vaccine production. Petri and Ravdin (103) were able to immunize animals with galactose-specific adherence lectin and protect them from liver abscesses. Leitch et al. (70) postulated that amebae may invade the intestinal epithelium in areas where the mucous blanket is thin and there are fewer galactose or N-acetylgalactosamine residues to protect the host. Penetration of the mucous blanket was thought to be due primarily to mechanical ameboid movement and not to require enzymatic digestion. Talamas-Rohana and Meza (146) have shown that E. histolytica binds to a receptor on fibronectin, a glycoprotein, of the target cell. Whether this receptor contains galactose or N-acetylgalactosamine residues was not discussed. Once the organism was bound to the fibronectin, there appeared to be changes in its cytoskeleton, and the organism rapidly degraded and internalized the fibronectin receptor. Pore-forming proteins that help destabilize the host cell and bring about cytolysis have been studied by a number of investigators, but no conclusive role for them in cytolysis by E. histolytica has been provided (62, 69, 76, 125, 161). The pore-forming amebapore protein is transferred from the trophozoite to the target cell, causing a disruption of the transmembrane gradient and contributing to the cell's death by the colloid-osmosis lysis mechanism. Tachibana et al. (145) postulated that the monoclonal antibody they used was directed against the pore-forming protein found only in pathogenic isolates. Luaces and Barrett (73) and Muller et al. (92) have reported finding extracellular proteinases and hydrolases in E. histolytica. No direct role in host cell lysis has been found for these enzymes, but they might be involved in amebic invasiveness at intercellular junctions, thereby allowing amebae to invade host tissues. Cysteine proteinases are the major proteolytic enzymes detected in E. histolytica (30, 60, 61, 121). Cysteine proteinase can degrade cellular attach-

6 VOL. 5, 1992 ment and matrix proteins such as collagen, laminin, and fibronectin. There is a direct correlation between the amount of proteinase activity and the pathogenicity of E. histolytica (30, 60, 100, 121, 150, 153). Pathogenic isolates secrete larger amounts of proteinase than do nonpathogenic isolates, and the expression of mrna is 10- to 100-fold greater in pathogenic than in nonpathogenic zymodemes (60, 121, 150). The effects of calcium and phorbol esters on lipases and proteinases in the cytolytic event have been studied by Long-Krug et al. (72), Ravdin et al. (118, 119), and Weikel et al. (156). Release of intracellular calcium increased the activity of phospholipases, and phospholipase increased the hemolytic activity of E. histolytica through the release of free fatty acids (126). Phorbol esters were noted to increase the cytolytic activity of E. histolytica, whereas sphingosine inhibited cytolytic events. These two compounds are known to stimulate and inhibit the activity of protein kinase C, which is thought to have a major role in the cytolytic event (29, 156). The definitive chronology of cytolytic events awaits further studies. Asymptomatic Infection Most human infections are asymptomatic, with the parasite acting as a harmless commensal organism in as many as 90% of those infected (31, 45). Individuals harboring the organisms may have either a negative or a weak antibody titer and show no occult blood in stool samples, and cysts can usually be detected in the stool when examinations for ova and parasites are done. If trophozoites are detected, they do not contain any phagocytized erythrocytes. Isoenzyme analyses of organisms isolated from asymptomatic individuals have usually shown that the isolates belong to nonpathogenic zymodemes (130). Many asymptomatic individuals never become symptomatic and usually excrete cysts in the stool for only a short time, as shown by repeat examinations for ova and parasites or by culture. This pattern is found in persons infected with pathogenic and nonpathogenic strains (1, 4, 16, 40, 44, 93). Symptomatic Infection About 10% of individuals infected with E. histolytica have clinical symptoms, with intestinal and/or extraintestinal pathology. Although this proportion of infected individuals is small, the disease represents a major public health problem in terms of morbidity and mortality in Third World countries. Patients with noninvasive infection may exhibit nonspecific gastrointestinal symptoms, including abdominal pain and increased frequency of bowel movements, that may be intermittent or chronic. When amebae invade host tissues, only trophozoites are seen and the organism loses its ability to encyst. Amebic invasion of the intestinal mucosa occurs most frequently (in descending order) in the cecum, ascending colon, sigmoid colon, appendix, descending colon, and transverse colon (56). The symptoms associated with invasive intestinal amebiasis are usually nonspecific. The onset of symptoms is usually gradual, with abdominal pain, diarrhea, dysentery, or weight loss occurring, depending on the severity of disease. Almost all patients have occult blood in the stools, and nearly one-third exhibit fevers. Histological examination of tissue may be required to rule out malignancy and inflammatory bowel disease. When amebic colitis is detected, it is usually (70% of cases) associated with segmentary ulceration of the colon (21). If ulceration with perforation occurs, it occurs mainly AMEBIASIS 361 in the cecum (53). Dissemination of amebic infection to extraintestinal sites most frequently involves the liver, lungs, pericardium, brain, and skin. Colonic mucosal invasion begins in the interglandular epithelium, where the overlying mucous layer is noticeably thin (110). With initial microulceration, there is an outflow of tissue fluids, erythrocytes, neutrophils, lymphocytes, eosinophils, plasma cells, epithelial cells, and Charcot-Leyden crystals (108). The inflammatory response to amebic invasion is minimal and may be due to the lysis of tissue cells by E. histolytica. One or several ulcerations may be present, but if there are several ulceration sites, they are usually separated. At first, the ulceration is superficial, and cellular infiltration and tissue necrosis are limited to the invasion site. Proctoscopic examination reveals an edematous and friable mucosa. The ulcer has a pinhead center that is composed of eosinophilic granular debris, some inflammatory cells, and E. histolytica trophozoites. One can see normal mucosal surface between ulcers (109). As the ulcer deepens and progresses, it forms the classic flask-shaped ulcer of amebic colitis, which extends from the mucosa and muscularis mucosa into the submucosa. When ulcers coalesce, the mucosa undergoes necrosis and sloughs. Amebae are found on the advancing edges of the ulcer and usually not in the necrotic areas. Many times, the ulcers become secondarily infected with bacteria. "Toxic megacolon" is the result of total confluence of ulcerations and necrosis of the colon. This form of amebiasis has a very high mortality rate. Ameboma, the result of chronic ulceration, is predominantly found in the cecum, sigmoid colon, and rectum (15). These tumorlike lesions are often mistaken for malignancies and are sometimes present as palpable masses. Histologically, the ameboma is nonfibrotic and contains granulation tissue with lymphocytes and giant cells (15). Perforation of the colon by an amebic ulcer occurs in as many as 21% of patients with acute amebic colitis (15) and may result in peritonitis and liver abscess. Perforations occur most commonly in the cecum (58). Peritonitis is frequently the result of liver abscess rather than intestinal perforation, but the latter has a high mortality rate (91). Amebic infection of extraintestinal sites is the result of previous intestinal infection. Liver abscess can occur concurrently with colitis; however, there is frequently no clinical history of recent E. histolytica intestinal infection or evidence of parasites in stool specimens. Onset may be acute, with abdominal pain and fever, or subacute, with weight loss; less than 50% of patients have fever or abdominal pain. Patients may have leukocytosis without eosinophilia, elevated alkaline phosphatase and transaminase levels, and a high erythrocyte sedimentation rate. E. histolytica trophozoites infecting the muscularis mucosa frequently digest the walls of mesenteric venules, enter the circulation, and are carried to the liver via the portal venules (15). The trophozoites are detected more frequently in the right lobe rather than the left lobe of the liver, perhaps because of increased perfusion of the larger liver mass (15). Liver abscess is diagnosed more frequently in adults (ages 20 to 60 years) than in children and more frequently in males than in females (142). The liver abscess has a thin capsular wall with a necrotic center composed of a thick fluid, an intermediate zone of coarse stroma, and an outer zone of nearly normal tissue that is being invaded by amebae (15). The contents of the abscess have been described as resembling anchovy paste because of the light brown color. Generally, the abscess is bacteriologically sterile. Secondary bacterial invasion of an amebic

7 362 BRUCKNER CLIN. MICROBIOL. REV. TABLE 3. Laboratory diagnosis of amebiasis: specimen typea Sample Fixative Examination Stain Stool PVA, 10% Formalin, Schaudinn's Concentrate, permanently stained slide Gomori trichrome, iron hematoxylin fixative, SAF Sigmoid colon PVA, Schaudinn's fixative Permanently stained slide Gomori trichrome, iron hematoxylin Aspirate Direct None Wet mount with or without enzyme digest Fixed PVA, Schaudinn's fixative Permanently stained slide Gomori trichrome, iron hematoxylin Biopsy Formalin Routine histology PAS, H&E Blood None Serology" None a PVA, PVA with either HgCl2 or CuS04; SAF, sodium acetate-acetic acid-formalin; PAS, periodic acid-schiff (organism stains intensely pink and has a distinct outline, but cytoplasmic and nuclear details are obscured); H&E, hematoxylin and eosin stain (cytoplasm is pink and nucleus is blue; in sections, the nucleus may not always be present). b Standard serological tests include indirect hemagglutination, counterimmunoelectrophoresis, indirect fluorescent antibody, and enzyme immunoassay. abscess occurs infrequently. Microscopic examination of the abscess fluid reveals granular eosinophilic debris with no or few cells (67). In animal models, amebae are not directly responsible for the damage to the liver cells. The tissue destruction apparently results from lysis of leukocytes and macrophages by the amebae (152). Neutrophils are attracted to amebae and are rapidly killed on contact with E. histolytica. Although the neutrophils have little or no cytolytic effect on the virulent organisms (46), the release of toxic intracellular products from these destroyed host cells promotes lysis and necrosis of the hepatic cells. If the amebic lesions heal, they invariably heal without forming scar tissue (15). Because of the cytolytic action of the amebae, leukocytes may be prevented from accumulating in the abscess area, thereby inhibiting the initiation of a cellular immune response. E. histolytica trophozoites inhibit the mobility of monocytes but not leukocytes (42). Monocytes (macrophages) play a crucial role in any wound-healing process by stimulating collagen deposition (37, 64, 65). DIAGNOSIS E. histolytica infections can be diagnosed in the laboratory by the detection of organisms in stool preparations, sigmoidoscopy smears, tissue biopsy samples, or hepatic aspirate samples or by serological tests (Table 3). Organism detection depends on proper specimen collection, processing, and examination by trained personnel. Examination for ova and parasites in a minimum of three stool specimens, using concentration and permanent stain techniques, is the recommended standard for the detection and identification of the organism (38). Reliance on inadequately trained personnel for laboratory diagnosis may lead to missed infections or, more commonly, false-positive results because of misidentification of leukocytes as E. histolytica. Although wet mounts of fresh specimens can be screened for motile trophozoites containing erythrocytes, most patients do not harbor trophozoites with erythrocytes. If fresh stool samples are examined, they should be examined within 30 min after passage of the specimen, not 30 min after arrival in the laboratory. If the specimen cannot be examined within 30 min, it should be preserved in a suitable fixative such as polyvinyl alcohol (PVA). Although some textbooks advocate the use of direct wet mounts to identify E. histolytica, the identification should always be confirmed by using a permanent-stained slide. The importance of the stained slide cannot be overemphasized, particularly when organisms are few in number in the stool specimen. Even if sigmoidoscopy is performed, a minimum of three stool specimens should also be collected and examined. No fewer than six areas of the mucosa should be sampled, with examination slides made at the bedside or in the examination room. Material scraped or aspirated from the mucosal surface can be examined by three methods. If there is no delay, a wet mount in 0.85% NaCl can be made and examined for ameboid movement. It may be several minutes after the NaCl preparation has been made before the amebae exhibit any motility. Microscopic examination may be difficult because of the need for low light intensity and difficulties in differentiating the unstained amebae from cellular material. In wet preparations, other protozoa such as E. coli may resemble E. histolytica (Table 3). Only careful examination will distinguish amebae from other host cells, such as polymorphonuclear leukocytes, which mimic E. histolytica cysts, and macrophages, which may be mistaken for E. histolytica trophozoites (Table 4). The mucosal material may also be smeared on a slide and immediately immersed in Schaudinn's fixative. An alternative method of preparing a permanent fixed-slide specimen is to add 2 to 3 drops of PVA directly to the mucosal material directly on the slide, mix it, and allow the slide to air dry. These two methods provide a permanent mount that can be stained with trichrome stain (38). Histologic examination of tissue biopsy sections can be done with periodic acid-schiff and hematoxylin-eosin stains (Table 3). In histologic material, all the nuclear detail may not be seen in one organism, and sequential sections may have to be examined to identify the organism. The amebae must be differentiated from normal host cells such as histiocytes. Hepatic aspirates can be examined for the organism, but this examination is rarely performed and rarely done correctly. Aspirated material should be collected in a number of different containers as it is obtained from the abscess. The amebae are sparse in necrotic material from the center of the abscess, but they are more abundant on the marginal walls. Therefore, they are more commonly found in the last portions of aspirated material. Enzyme liquefaction has been used to facilitate the search for organisms by freeing the organisms from aspirate material (5). Demonstration of the organisms is often very difficult because they may be trapped in viscous pus or debris and will not exhibit typical motility. In extraintestinal amebiasis, organisms may or may not be found in the stool; therefore, one of the many types of available serological tests may be necessary for diagnosis (48, 99). Serology should not be used solely for the diagnosis of intestinal amebiasis, because antibody titers may be low

8 VOL. 5, 1992 AMEBIASIS 363 TABLE 4. Differential characteristics of host cells and E. coli, commonly mistaken for E. histolytica, in wet preparations Nucleus Diam or Cell Diam or length Motility Visibility (ratio of ~~~~~~~~~~~~Cytoplasmic appearance Inclusions (Lm) No. nuclear material to (stained) cytoplasm) E. histolytica Trophozoites 12-60; usual Progressive, with hyaline 1 Hard to see in un- Clear differentiation Presence of erythrorange, fingerlike pseudopodia; stained prepns (1: of ectoplasm and cytes diagnostic may be rapid 10-1:12) endoplasm; vacuoles usually small, if present; looks like ground glass Cysts 10-20; usual None 1-4 1:2-1:3 Clear Chromatoid bodies range, (stained) may be present; usually elongate with blunt, smooth, rounded edges; round or oval E. coli Trophozoites 15-50; usual Sluggish, nondirectional; 1 Often visible in un- Granular, little dif- Bacteria, yeast cells, range, blunt, granular stained prepn ferentiation into other debris pseudopodia ectoplasm and endoplasm; usually vacuolated Cysts 10-35; usual None 1-8 >16 nuclei seen oc- Clear Chromatoid bodies range, casionally (stained) may be present, less frequently than in E. histolytica; splinter shape with rough, pointed ends Host cells Polymorphonuclear Avg, 16 None segments; if Granular None leukocytes lobed nucleus fragments may mimic the 4 nuclei found in E. histolytica cyst (1:1) Macrophages 20-60; may be Sluggish 1 Large, may be irreg- Coarse; may be Usually contain in ular in shape (like highly vacuolated gested debris, monocyte); may polymorphonumimic E. histoljt- clear leukocytes, ica trophozoite; and erythrocytes can also ingest erythrocytes and/or impossible to interpret, particularly with patients from an area of endemicity where a high prevalence of seropositivity already exists. Asymptomatic cyst passers usually have negative serological test results, but there is good correlation between infection with pathogenic zymodemes and positive serological tests (130). Positive indirect hemagglutination, indirect fluorescent-antibody, and immunodiffusion tests give results that correlate well with the presence or absence of an amebic liver abscess. An important major problem common to most serological tests is the persistence of antibody for prolonged periods (up to 2 years) after diagnosis and therapy. These tests should therefore be interpreted with caution. However, indirect fluorescentantibody titers decrease within 6 to 12 months of successful therapy. Serological tests can also be used to rule out amebiasis in patients with inflammatory bowel disease (49). Although numerous methods for the serological detection of amebiasis have been reported in the literature, the use of these methods in clinical laboratories is severely limited by the unavailability of commercial reagents or test kits (38). Newer diagnostic methods being proposed include the use of purified antigens and assays for immunoglobulin M and A antibodies (6, 28, 55, 102, 117, 139). The Centers for Disease Control offer amebic serological tests. Submission regulations vary from state to state; therefore, laboratories should check their county or state public health department for guidelines. Recently, culture techniques with Robinson's medium, Diamond's medium, Jones's medium, and Locke's egg yolk serum medium in conjunction with isoenzyme electrophoresis have been used to differentiate infections caused by pathogenic and nonpathogenic E. histolytica strains (85-87, 136). At present, this technique is not practical for routine clinical laboratories. Results usually are not available for 4 days or more. Even though a patient may be diagnosed by clinical history, serological tests, or zymodeme analysis as harboring a nonpathogenic strain, the question of whether or not the strain will revert to pathogenicity in vivo has been raised (42, 85-87, 89, 130, 131). To provide clinically relevant information to physicians

9 364 BRUCKNER for treatment of patients infected with pathogenic strains of E. histolytica, methods using monoclonal antibodies, purified antigens, or DNA probes may be required (8, 13, 39, 102, 127, 143, 144, 147, 151). Although there have been numerous reports of using techniques such as enzyme immunoassay to detect antigens in specimens, none of these methods are routinely used in clinical laboratories (84). If zymodeme patterns are found to remain stable, these techniques could be incorporated into a routine diagnostic laboratory, provided that they can be used directly on clinical specimens rather than after the isolation of organisms by culture. Zymodeme patterns could be very useful for epidemiological purposes in tracking organism transmission and potential differences in organism invasiveness, virulence, and therapeutic susceptibility. The use of purified antigens and RNA or DNA probes for routine diagnostic testing would offer significant improvements over current serological or direct detection methods. The serological techniques commonly used for diagnostic purposes are not sensitive or specific enough to be solely relied on clinically. Reliable probes would help to eliminate false-positive test results due to misidentification of E. coli or host cells such as polymorphonuclear leukocytes or macrophages. Commercial kits that make use of purified antigens or DNA probes are not available at this time. For a more comprehensive review of laboratory diagnostic techniques, see Garcia and Bruckner (38). THERAPY Two classes of drugs are used in the treatment of amebic infections. Luminal amebicides, such as iodoquinol and diloxanide furoate, act on organisms in the intestinal lumen and are not effective against organisms in tissue. Tissue amebicides, such as metronidazole, chloroquine, and dehydroemetine, are effective in the treatment of invasive amebiasis but less effective or ineffective in the treatment of organisms in the bowel lumen. Asymptomatic Patients Asymptomatic patients who are cyst passers may be treated with diloxanide furoate, iodoquinol, or paromomycin (38, 81, 112, 128, 140, 141). Like iodoquinol, diloxanide furoate is a luminal amebicide. Diloxanide furoate is available only from the Centers for Disease Control on special request. Although known as a luminal amebicide, diloxanide furoate is readily hydrolyzed in the intestinal tract and absorbed into the bloodstream. For asymptomatic patients who are passing both cysts and trophozoites in the stool, the recommended treatment is metronidazole plus iodoquinol (38, 81, 128, 140, 141). From zymodeme analysis, it has been determined that many asymptomatic patients harbor nonpathogenic zymodemes, and the use of drug therapy in these individuals is controversial. As discussed above, methods to differentiate pathogenic from nonpathogenic E. histolytica strains are not easily incorporated into routine clinical laboratory use. Therefore, from a public health viewpoint, it is best to treat all infected patients. Symptomatic Patients The treatment of symptomatic patients varies with the clinical stage of the infection. Different therapeutic approaches are employed for intestinal and extraintestinal amebiasis (38, 81, 114, 128, 140, 141). CLIN. MICROBIOL. REV. Mild intestinal disease due to amebic dysentery should be treated with metronidazole plus iodoquinol or with metronidazole plus diloxanide furoate. Patients should be warned to refrain from ingesting alcohol when taking metronidazole, for they may have a reaction similar to that to disulfiram. Symptoms may be abrupt and severe and include abdominal pain, nausea, vomiting, and headache. The safety of iodoquinol taken during pregnancy has not been determined. In cases of dehydration, fluid and electrolyte replacement may be necessary. Extraintestinal Disease A number of combination therapies have been used for treating extraintestinal amebiasis (38, 81, 114, 128, 140, 141). These include metronidazole and dehydroemetine with chloroquine. Chloroquine is useful for treating amebic liver abscess because it is concentrated in the liver; however, chloroquine is not useful for intestinal amebiasis. Dehydroemetine can be toxic to the heart, so patients should be monitored throughout therapy. Although surgical drainage of larger abscesses may be necessary, aspiration of small and moderate-sized abscesses can usually be avoided. Smaller abscesses resorb, and their disappearance can be monitored by scanning procedures (computerized tomography, liver scan, or ultrasound). Extraintestinal therapy should be followed with a luminal amebicide such as iodoquinol or diloxanide furoate. Follow-Up All treated patients should be followed up clinically and with laboratory tests. Patients who had positive stool sample results should have follow-up stool examinations at 2 to 4 weeks posttherapy. The stool examination must include a permanent stained slide. Patients with extraintestinal amebiasis should show a clinical response after therapy and can be monitored with the scanning procedures mentioned above. Although antibody titers in serum may remain elevated for prolonged periods and may be of little use in follow-up, indirect fluorescent-antibody titers have been shown to decrease within 6 to 12 months of successful therapy. SUMMARY AND FUTURE DIRECTIONS Although approximately 12% of the world's population is infected with E. histolytica, only 10% of those infected will develop clinical symptoms, and of this group, only 2 to 20% will have invasive disease. It has been postulated that the large number of asymptomatic people are infected with a nonpathogenic strain that cannot be distinguished morphologically from pathogenic strains. It has been shown that pathogenic and nonpathogenic strains can be differentiated by the use of isoenzyme (zymodeme) analysis, monoclonal antibodies, and molecular probes, and the results correlate with clinical symptoms and serological tests. Whether the isoenzyme patterns are a phenotypic or genotypic trait will have to be answered by future research. It is possible to transfer genetic material between organisms, thus transforming one pathogenic zymodeme into a completely different pathogenic zymodeme, and the presence or absence of specific bacterial flora has been associated with the interconversion of nonpathogenic and pathogenic strains in culture. Mirelman (87) found that nonpathogenic E. histolytica isolates contain extrachromosomal DNA that hybridizes with

10 VOL. 5, 1992 pathogen-specific DNA probes. However, through the use of recombinant DNA technology, two genetically distinct forms of E. histolytica have been found to exist. Genetic analysis has shown substantial differences between the nonpathogenic and pathogenic forms. Until our understanding of the pathogens is elucidated, there will continue to be questions concerning the influence of bacterial flora and genetic exchange between isolates. Because of the uncertainty of isoenzyme stability, the expression or repression of pathogenic genes, or transfer of genetic material between organisms, the designation of an organism as nonpathogenic or pathogenic should not be solely relied on for therapeutic purposes. Although recent research has helped to elucidate the molecular mechanisms of organism attachment, cytolysis, and phagocytosis of host cells, pathogenesis is still not well understood, even though E. histolytica has been studied for over 100 years. Most recent work cited has dealt with the galactose-specific lectin receptor of E. histolytica. Prevention or alteration of attachment may eliminate the pathogenicity of this organism. The sequencing of this receptor and others involved in attachment may lead in the near future to a vaccine or a specific therapeutic agent that will prevent or reduce the pathological sequelae associated with pathogenic zymodemes. Diagnosis of E. histolytica infections depends primarily on examinations for ova and parasites and on serological tests. There are no commercially available or practical and reliable methods for differentiating pathogenic from nonpathogenic strains for routine use in clinical laboratories. Currently, there are no antigen tests for routine clinical laboratory use, and monoclonal antibodies and DNA probe tests have been used only with culture isolates to detect pathogenic zymodemes. With the current interest in developing genomic libraries (51) and in using amplification methods to selectively look at specific functions of the organism, our knowledge of E. histolytica will be considerably enhanced. Amplification methods and serological tests that are sensitive and specific enough to directly detect very small numbers of organisms are currently needed. In the near future, clinical laboratories will be evaluating a number of new test products for diagnosing infections caused by pathogenic E. histolytica. REFERENCES 1. Allason-Jones, E., A. Mindel, P. Sargeaunt, and P. Williams Entamoeba histolytica as a commensal intestinal parasite in homosexual men. N. Engl. J. Med. 315: Andrews, B. J., L. Mentzoni, and M. Bjorvatn Zymodeme conversion of isolates of Entamoeba histolytica. Trans. R. Soc. Trop. Med. Hyg. 84: Balamuth, W Improved egg yolk infusion for cultivation of Entamoeba histolytica and other intestinal protozoa. Am. J. Clin. Pathol. 16: Beaver, P. C., R. Jung, H. Sherman, T. Read, and T. Robinson Experimental Entamoeba histolytica infection in man. Am. J. Trop. Med. Hyg. 5: Beaver, P. C., R. C. Jung, and E. W. Cupp Clinical parasitology. Lea and Febiger, Philadelphia, Pa. 6. Bhattacharya, A., S. Bhattacharya, M. P. Sharma, and L. S. Diamond Metabolic labeling of Entamoeba histolytica antigens: characterization of a 28-kDa major intracellular antigen. Exp. Parasitol. 70: Bhattacharya, A., R. Ghildyal, J. Prasad, S. Bhattacharya, and L. S. Diamond Modulation of a surface antigen of Entamoeba histolytica in response to bacteria. Infect. Immun. 60: Blakely, P., P. G. Sargeaunt, and S. L. Reed An AMEBIASIS 365 immunogenic 30-kDa surface antigen of pathogenic clinical isolates of Entamoeba histolytica. J. Infect. Dis. 162: Blanc, D., R. Nicholls, and P. G. Sargeaunt Experimental production of new zymodemes of Entamoeba histolytica supports the hypothesis of genetic exchange. Trans. R. Soc. Trop. Med. Hyg. 83: Blanc, D., and P. G. Sargeaunt Entamoeba histolytica zymodemes: exhibition of -y and 8 bands only of glucose phosphate isomerase and phosphoglucomutase may be influenced by starch content in the medium. Exp. Parasitol. 72: Boeck, W. C., and J. Drbohlav The cultivation of Endamoeba histolytica. Am. J. Hyg. 5: Bracha, R., A. Chayen, I. Rosenberg, L. G. Warren, and D. Mirelman Isolation and partial characterization of the hexokinase isoenzymes from pathogenic and non-pathogenic strains of Entamoeba histolytica. Mol. Biochem. Parasitol. 25: Bracha, R., L. S. Diamond, J. P. Akers, G. D. Burchard, and D. Mirelman Differentiation of clinical isolates of Entamoeba histolytica by using specific DNA probes. J. Clin. Microbiol. 28: Bracha, R., and D. Mirelman Virulence of Entamoeba histolytica trophozoites. Effects of bacteria, microaerobic conditions and metronidazole. J. Exp. Med. 160: Brandt, H., and R. Perez-Tamayo Pathology of human amebiasis. Hum. Pathol. 1: Bray, R. S., and W. G. Harris The epidemiology of infection with Entamoeba histolytica in The Gambia, West Africa. Trans. R. Soc. Trop. Med. Hyg. 71: Brown, M., S. Reed, J. A. Levy, M. Busch, and J. H. McKerrow Detection of HIV-1 in Entamoeba histolytica without evidence of transmission to human cells. AIDS 5: Burchard, G. D., F. T. Hufert, and D. Mirelman Characterization of 20 Entamoeba histolytica strains isolated from patients with HIV infection. Infection 19: Burrows, R. B Endamoeba hartmanni. Am. J. Hyg. 65: Burrows, R. B Morphological differentiation ofentamoeba hartmanni and E. polecki from E. histolytica. Am. J. Trop. Med. Hyg. 8: Castro, H. F Anatomic and pathological findings in amebiasis report of 320 cases, p In C. A. Padilla y Padilla (ed.), Amebiasis in man. Charles C Thomas, Springfield, Ill. 22. Chadee, K., K. Keller, J. Forstner, D. J. Innes, and J. I. Ravdin Mucin and nonmucin secretagogue activity of Entamoeba histolytica and cholera toxin in rat colon. Gastroenterology 100: Chadee, K., W. A. Petri, Jr., D. J. Innes, and J. I. Ravdin Rat and human colonic mucins bind to and inhibit adherence lectin of Entamoeba histolytica. J. Clin. Invest. 80: Clark, C. G., and L. S. Diamond The Laredo strain and other "Entamoeba histolytica-like" amoebae are Entamoeba moshkovskdi. Mol. Biochem. Parasitol. 46: Clark, C. G., and L. S. Diamond Ribosomal RNA genes of "pathogenic" and "nonpathogenic" Entamoeba histolytica are distinct. Mol. Biochem. Parasitol. 49: Councilman, W. T., and H. A. LaFleur Amoebic dysentery. Johns Hopkins Hosp. Rep. 2: DeLeon, A Pronsotico tardio en el absceso hepatico ambiano. Arch. Invest. Med. 1: del Mundo, R. E., E. Acosta, E. Merino, W. Glender, and L. Ortiz-Ortiz Diagnosis of intestinal amebiasis using salivary IgA antibody detection. J. Infect. Dis. 162: DeMeester, F., D. Mirelman, T. Stolarsky, and D. S. Lester Identification of protein kinase C and its potential substrate in Entamoeba histolytica. Comp. Biochem. Physiol. 97b: DeMeester, F., E. Shaw, H. Scholze, T. Stolarsky, and D. Mirelman Specific labeling of cysteine proteinases in pathogenic and nonpathogenic Entamoeba histolytica. Infect.

11 366 BRUCKNER Immun. 58: Denis, M., and K. Chadee Immunopathology of Entamoeba histolytica infections. Parasitol. Today 4: Diamond, L. S Axenic cultivation of Entamoeba histolytica. Science 134: Diamond, L. S Entamoeba histolytica Schaudinn, 1903: from xenic to axenic cultivation. J. Protozool. 33: Diamond, L. S., B. P. Phillips, and I. L. Bartgis A comparison of the virulence of nine strains of axenically cultivated Entamoeba histolytica in hamster liver. Arch. Invest. Med. 5(Suppl. 2): Dobell, C Research on the intestinal protozoa of monkeys and man. Parasitology 20: Farri, T. A., P. G. Sargeaunt, 0. C. Warhurst, J. E. Williams, and R. Bhojnani Electrophoretic studies of the hexokinase of Entamoeba histolytica groups I to IV. Trans. R. Soc. Trop. Med. Hyg. 74: Freundlich, B., J. S. Bomalaski, E. Neilson, and S. A. Jimenez Regulation of fibroblast proliferation and collagen synthesis by cytokines. Immunol. Today 7: Garcia, L. S., and D. A. Bruckner Diagnostic medical parasitology. Elsevier Science Publishing Co. Inc., New York. 39. Garfinkel, L. I., M. Giladi, M. Huber, C. Gitler, D. Mirelman, M. Revel, and S. Rozenblatt DNA probes specific for Entamoeba histolytica possessing pathogenic and nonpathogenic zymodemes. Infect. Immun. 57: Gathiram, V., and T. F. H. G. Jackson A longitudinal study of asymptomatic carriers of pathogenic zymodemes of Entamoeba histolytica. S. Afr. Med. J. 72: Gathiram, V., and T. F. H. G. Jackson Pathogenic zymodemes of Entamoeba histolytica remain unchanged throughout their lifecycle. Trans. R. Soc. Trop. Med. Hyg. 84: Gimenez-Scherer, J. A., M. G. Pacheco-Cano, E. Cruz de Lavin, P. Hernandez-Jauregui, M. T. Merchant, and R. R. Kretschmer Ultrastructural changes associated with the inhibition of monocyte chemotaxis caused by products of axenically grown Entamoeba histolytica. Lab. Invest. 57: Gitler, C., and D. Mirelman Factors contributing to the pathogenic behavior of Entamoeba histolytica. Annu. Rev. Microbiol. 40: Goldmeier, D., A. B. Price, 0. Billington, P. Borriello, A. Shaw, P. G. Sargeaunt, P. E. Munday, I. Dixon, J. M. Carder, J. Hilton, and D. J. Jeffries Is Entamoeba histolytica in homosexual men a pathogen? Lancet i: Guerrant, R. L Amebiasis: introduction, current status and research questions. Rev. Infect. Dis. 8: Guerrant, R. L., J. Brush, J. I. Ravdin, J. A. Sullivan, and G. I. Mandell Interaction between Entamoeba histolytica and human polymorphonuclear neutrophils. J. Infect. Dis. 143: Haque, R., A. Hall, and S. Tzipori Zymodemes of Entamoeba histolytica in Dhaka, Bangladesh. Ann. Trop. Med. Parasitol. 84: Healy, G. R Immunologic tools in the diagnosis of amebiasis: epidemiology in the United States. Rev. Infect. Dis. 8: Healy, G. R, and S. C. Kraft The indirect hemagglutination test for amebiasis in patients with inflammatory bowel disease. Am. J. Dig. Dis. 17: Hoare, C. A Reservoir hosts and natural foci of human protozoal infection. Acta Trop. 19: Huber, M., B. Koller, C. Gitler, D. Mirelman, M. Revel, S. Rozenblatt, and L. Garfinkel Entamoeba histolytica ribosomal RNA genes are carried on palindromic circular DNA molecules. Mol. Biochem. Parasitol. 32: Jackson, T. F. H. G., and V. Gathiram Seroepidemiological study of antibody responses to the zymodemes of Entamoeba histolytica. Lancet i: Jimenez, F Pathology of amebiasis. Bull. N.Y. Acad. Med. 57: Jones, W. R The experimental infection of rats with CLIN. MICROBIOL. REV. Entamoeba histolytica with a method for evaluating the antiamoebic properties of new compounds. Ann. Trop. Med. Parasitol. 40: Joyce, M. P., and J. I. Ravdin Antigens of Entamoeba histolytica recognized by immune sera from liver abscess patients. Am. J. Trop. Med. Hyg. 38: Joyce, M. P., and J. I. Ravdin Pathology of human amebiasis, p In J. I. Ravdin (ed.), Amebiasis: human infection by Entamoeba histolytica. John Wiley & Sons, Inc., New York. 57. Kahn, F. H., and B. R. Visscher Water disinfection in the wilderness-a simple effective method of iodination. West. J. Med. 122: Kapoor, 0. P., B. N. Nathwani, and V. R. Joshi Amoebic peritonitis: a study of 73 cases. J. Trop. Med. Hyg. 75: Kean, B. H., K. E. Mott, and A. J. Russell Tropical medicine and parasitology, vol. 1, p Classics Investigations, Cornell University Press, Ithaca, N.Y. 60. Keene, W. E., M. E. Hidalgo, E. Orozco, and J. H. McKerrow Entamoeba histolytica: correlation of the cytopathic effect of virulent trophozoites with secretion of a cysteine proteinase. Exp. Parasitol. 71: Keene, W. E., M. G. Pettit, S. Allen, and J. H. McKerrow The major neutral proteinase of Entanoeba histolytica. J. Exp. Med. 163: Keller, F., C. Walter, U. Lohden, W. Hanke, T. Bakker- Grunwald, and D. Trissl Pathogenic and non-pathogenic Entamoeba pore formation and hemolytic activity. J. Protozool. 35: Kollaritsch, H., H. Stemberger, M. Binder, and G. Wiedermann Suitability of the Phast system for discrimination of Entamoeba isolates by isoenzyme isoelectrofocusing. Trans. R. Soc. Trop. Med. Hyg. 83: Kretschmer, R. R Immunology of amebiasis, p In A. Martinez-Palomo (ed.), Amebiasis: human parasitic diseases. Elsevier Science Publishing Co., New York. 65. Kretschmer, R. R., M. L. Collado, M. G. Pacheco, M. C. Salinas, M. Lopez-Osuna, M. Lecuona, E. M. Castro, and J. Arellano Inhibition of human monocyte locomotion by products of axenically grown E. histolytica. Parasite Immunol. 7: Krogstad, D. J Isoenzyme patterns and pathogenicity in amebic infections. N. Engl. J. Med. 315: LaFleur, H. A Dysentery-abscess of liver-amoeba coli in stools and sputum-death-necropsy. Johns Hopkins Hosp. Bull. 2: Lake, J. A., and H. S. Slayter Three dimensional structure of the chromatoid body of Entamoeba invadens. Nature (London) 227: a.Lancet Editorial. i: Leippe, M., S. Ebel, 0. L. Schoenberger, R. D. Horstmann, and H. J. Muller-Eberhard Pore-forming peptide of pathogenic Entamoeba histolytica. Proc. Natl. Acad. Sci. USA 88: Leitch, G. J., S. A. Harris-Hooker, and I. A. Udezula Movement of Entamoeba histolytica trophozoites in rat cecum and colon intact mucus blankets and harvested mucus gels. Am. J. Trop. Med. Hyg. 39: Li, E., A. Becker, and S. S. Stanley, Jr Chinese hamster ovary cells deficient in N-acetylglucosaminyltransferase I activity are resistant to Entamoeba histolytica-mediated cytotoxicity. Infect. Immun. 57: Long-Krug, S. A., K. J. Fischer, R. M. Hysmith, and J. I. Ravdin Phospholipase A enzymes of Entamoeba histolytica: description and subcellular localization. J. Infect. Dis. 152: Luaces, A. L., and A. J. Barrett Affinity purification and biochemical characterization of histolysin, the major cysteine proteinase of Entamoeba histolytica. Biochem. J. 250: Lushbaugh, W. B., A. B. Kairalla, J. R. Cantey, A. F. Hofbauer, and F. E. Pittman Isolation of a cytotoxinenterotoxin from Entamoeba histolytica. J. Infect. Dis. 139:9-17.

12 VOL. 5, Lushbaugh, W. B., and J. H. Miller The morphology of Entamoeba histolytica, p In J. I. Ravdin (ed.), Amebiasis: human infection byentamoeba histolytica. John Wiley & Sons, Inc., New York. 76. Lynch, E. C., I. M. Rosenberg, and C. Gitler An ion-channel forming protein produced by Entamoeba histolytica. EMBO J. 1: Mackenstedt, U., M. Schmidt, W. Raether, H. Mehlhorn, and M. Uphoff Increase in DNA content in tissue stages of Entamoeba histolytica strain SFL3. Parasitol. Res. 76: Mann, B. J., D. Mirelman, and W. A. Petri, Jr The D-galactose-inhibitable lectin of Entamoeba histolytica. Carbohydr. Res. 213: Mann, B. J., and W. A. Petri, Jr Cell surface proteins of Entamoeba histolytica. Parasitol. Today 7: Mann, B. J., B. E. Torian, T. S. Vedvick, and W. A. Petri, Jr Sequence of a cysteine-rich galactose-specific lectin of Entamoeba histolytica. Proc. Natl. Acad. Sci. USA 88: Markell, E. K., M. Voge, and D. T. John Medical parasitology. W. B. Saunders Co., Philadelphia, Pa. 82. Mathews, H. M., D. M. Moss, G. R. Healy, and G. S. Visvesvara Polyacrylamide gel electrophoresis of isoenzymes from Entamoeba species. J. Clin. Microbiol. 17: Mattern, C. T., D. B. Keister, and P. C. Natovitz Entamoeba histolytica "toxin": fetuin neutralizable and lectinlike. Am. J. Trop. Med. Hyg. 29: Merino, E., W. Glender, R. del Muro, and L. Ortiz-Ortiz Evaluation of the ELISA test for detection of Entamoeba histolytica in feces. J. Clin. Lab. Anal. 4: Meza, I., M. De La Garza, M. A. Meraz, B. Gallegos, M. De La Torre, M. Tanimoto, and A. Martinez-Palomo Isoenzyme patterns of Entamoeba histolytica isolates from asymptomatic carriers: use of gradient acrylamide gels. Am. J. Trop. Med. Hyg. 35: Mirelman, D Effect of culture conditions and bacterial associates on the zymodemes of Entamoeba histolytica. Parasitol. Today 3: Mirelman, D Ameba-bacterium relationship in amebiasis. Microbiol. Rev. 51: Mirelman, D Ameba-bacterial relationship in amebiasis, p In J. I. Ravdin (ed.), Amebiasis: human infection by Entamoeba histolytica. John Wiley & Sons, Inc., New York. 89. Mirelman, D., R. Bracha, S. Rozenblatt, and L. I. Garfinkel Repetitive DNA elements characteristic of pathogenic Entamoeba histolytica strains can also be detected after polymerase chain reaction in a cloned nonpathogenic strain. Infect. Immun. 58: Mirelman, D., R. Bracha, A. Wexler, and A. Chayen Changes in isoenzyme patterns of a cloned culture of nonpathogenic Entamoeba histolytica during axenization. Infect. Immun. 54: Monga, N. K., S. Sood, S. P. Kaushik, H. S. Sachdeva, K. C. Sood, and D. V. Datta Amebic peritonitis. Am. J. Gastroenterol. 66: Muller, F.-W., A. Franz, and E. Werries Secretory hydrolases of Entamoeba histolytica. J. Protozool. 35: Nanda, R., U. Baveja, and B. S. Anand Entamoeba histolytica cyst passers: clinical features and outcome in untreated subjects. Lancet ii: Nauss, R. W., and I. Rappaport Studies on amoebiasis: pathogenesis of mucosal penetration. Am. J. Trop. Med. 20: Neal, R. A., and P. Vincent Strain variation in Entamoeba histolytica. 1. Correlation of invasiveness in rats with the clinical history and treatment of the experimental infections. Parasitology 45: Nozaki, T., I. Aca, E. Okuzawa, M. Magalhaes, S. Tateno, and T. Takeuchi Zymodemes of Entamoeba histolytica isolated in the Amazon and the north-east of Brazil. Trans. R. AMEBIASIS 367 Soc. Trop. Med. Hyg. 84: Orozco, E Pathogenesis in amebiasis. Infect. Agents Dis. 1: Orozco, E., G. Guarneros, A. Martinez-Palomo, and T. Sanchez Entamoeba histolytica phagocytosis as a virulence factor. J. Exp. Med. 158: Perez-Montfort, R., and R. P. Kretschmer Humoral immune responses, p In R. P. Kretschmer (ed.), Amebiasis: infection and disease by Entamoeba histolytica. CRC Press, Boca Raton, Fla Perez-Tamayo, R., I. Becker, I. Montfort, P. Ostoa-Saloma, and R. Perez-Montfort Role of leukocytes and amebic proteinases in experimental rat testicular necrosis produced by Entamoeba histolytica. Parasitol. Res. 77: Petri, W. A., Jr Invasive amebiasis and the galactosespecific lectin of Entamoeba histolytica. ASM News 57: Petri, W. A., Jr., T. F. H. G. Jackson, V. Gathiram, K. Kress, L. D. Saffer, T. L. Snodgrass, M. D. Chapman, Z. Keren, and D. Mirelman Pathogenic and nonpathogenic strains of Entamoeba histolytica can be differentiated by monoclonal antibodies to the galactose-specific adherence lectin. Infect. Immun. 58: Petri, W. A., Jr., and J. I. Ravdin Protection of gerbils from amebic liver abscess by immunization with the galactosespecific adherence lectin of Entamoeba histolytica. Infect. Immun. 59: Petri, W. A., Jr., R. D. Smith, P. H. Schlesinger, C. F. Murphy, and J. I. Ravdin Isolation of the galactose-binding lectin that mediates the in vitro adherence of Entamoeba histolytica. J. Clin. Invest. 80: Petri, W. A., Jr., T. L. Snodgrass, T. F. H. G. Jackson, V. Gathiram, A. E. Simjee, K. Chadee, and M. D. Chapman Monoclonal antibodies directed against the galactose-binding lectin of Entamoeba histolytica enhance adherence. J. Immunol. 144: Phillips, B. P., L. S. Diamond, I. L. Bartgis, and S. A. Stuppler Results of intracecal inoculation of germ-free and conventional guinea pigs and germ-free rats with axenically cultivated Entamoeba histolytica. J. Protozool. 19: Phillips, B. P., and F. Gorstein Effects of different species of bacteria on the pathology of enteric amebiasis in monocontaminated guinea pigs. Am. J. Trop. Med. Hyg. 15: Pittman, F. E., W. K. El-Hashimi, and J. C. Pittman Studies of human amebiasis. II. Light and electron-microscope observations of colonic mucosa and exudate in acute amebic colitis. Gastroenterology 65: Pittman, F. E., and G. R. Henningar Sigmoidoscopic and colonic mucosal biopsy findings in amebic colitis. Arch. Pathol. 97: Prathap, K., and R. Gilman The histopathology of acute intestinal amebiasis: a rectal biopsy study. Am. J. Pathol. 60: Proctor, E. M., and M. A. Gregory Ultrastructure of cysts of E. histolytica. Int. J. Parasitol. 3: Proctor, E. M., Q. Wong, J. Yang, and J. S. Keystone The electrophoretic isoenzyme patterns of strains of Entamoeba histolytica isolated in two major cities in Canada. Am. J. Trop. Med. Hyg. 37: Radin, D. R., P. W. Ralls, P. M. Colletti, and J. M. Halls CT of amebic liver abscess. Am. J. Roentgenol. 150: Rakel, R. E. (ed.) Current therapy. W. B. Saunders Co., Philadelphia, Pa Ravdin, J. I Pathogenesis of disease caused by Entamoeba histolytica: studies of adherence, secreted toxins, and contact-dependent cytolysis. Rev. Infect. Dis. 8: Ravdin, J. I., and R. L. Guerrant Role of adherence in cytopathogenic mechanisms of Entamoeba histolytica: study with mammalian tissue culture cells and human erythrocytes. J. Clin. Invest. 68: Ravdin, J. I., T. F. H. G. Jackson, W. A. Petri, Jr., C. F.

13 368 BRUCKNER Murphy, B. L. P. Ungar, V. Gathiram, J. Skilogrannis, and A. E. Simjee Association of serum antibodies to adherence lectin with invasive amebiasis and asymptomatic infections with pathogenic Entamoeba histolytica. J. Infect. Dis. 162: Ravdin, J. I., F. Moreau, J. A. Sullivan, W. A. Petri, Jr., and G. L. Mandell Relationship of free intracellular calcium to the cytolytic activity of Entamoeba histolytica. Infect. Immun. 56: Ravdin, J. I., C. F. Murphy, R. L. Guerrant, and S. A. Long-Krug Effect of antagonists of calcium and phospholipase A on the cytopathogenicity of Entamoeba histolytica. J. Infect. Dis. 152: Reed, S. L., B. M. Flores, M. A. Batzer, M. A. Stein, V. L. Stroeher, J. E. Carlton, D. L. Diedrich, and B. E. Torian Molecular and cellular characterization of the 29-kilodalton peripheral membrane protein of Entamoeba histolytica: differentiation between pathogenic and nonpathogenic isolates. Infect. Immun. 60: Reed, S. L., W. E. Keene, and J. H. McKerrow Thiol proteinase expression and pathogenicity of Entamoeba histolytica. J. Clin. Microbiol. 27: Reeves, R. E., and J. M. Bischoff Classification of Entamoeba species by means of electrophoretic properties of amebal enzymes. J. Parasitol. 54: Robinson, G. L The laboratory diagnosis of human parasitic amoebas. Trans. R. Soc. Trop. Med. Hyg. 62: Rodriguez, M. A., and E. Orozco Isolation and characterization of phagocytosis- and virulence-deficient mutants of Entamoeba histolytica. J. Infect. Dis. 154: Rosenberg, I., D. Bach, L. M. Loew, and C. Gitler Isolation, characterization, and partial purification of a transferable membrane channel (amoebapore) produced byentamoeba histolytica. Mol. Biochem. Parasitol. 33: Said-Fernandez, S., and R. Lopez-Revilla Free fatty acids released from phospholipids are the major heat-stable hemolytic factor of Entamoeba histolytica trophozoites. Infect. Immun. 56: Samuelson, J., R. Acuna-Soto, S. Reed, F. Biagi, and D. Wirth DNA hybridization probe for clinical diagnosis of Entamoeba histolytica. J. Clin. Microbiol. 27: Sanford, J. P Guide to antimicrobial therapy. Antimicrobial Therapy, Inc., West Bethesda, Md Sargeaunt, P. G Zymodemes expressing possible genetic exchange in Entamoeba histolytica. Trans. R. Soc. Trop. Med. Hyg. 79: Sargeaunt, P. G The reliability of Entamoeba histolytica zymodemes in clinical diagnosis. Parasitol. Today 3: Sargeaunt, P. G Zymodemes of Entamoeba histolytica. Parasitol. Today 3: Sargeaunt, P. G Zymodemes of Entamoeba histolytica, p In J. I. Ravdin (ed.), Amebiasis: human infection by Entamoeba histolytica. John Wiley & Sons, Inc., New York Sargeaunt, P. G., T. F. H. G. Jackson, S. R. Wiffen, and R. Bhojnani Biological evidence of genetic exchange in Entamoeba histolytica. Trans. R. Soc. Trop. Med. Hyg. 82: Sargeaunt, P. G., and J. E. Williams Electrophoretic isoenzyme patterns of Entamoeba histolytica and Entamoeba coli. Trans. R. Soc. Trop. Med. Hyg. 72: Sargeaunt, P. G., and J. E. Williams Electrophoretic isoenzyme patterns of the pathogenic and non-pathogenic intestinal amoebae of man. Trans. R. Soc. Trop. Med. Hyg. 73: Sargeaunt, P. G., J. E. Williams, and J. D. Grene The differentiation of invasive and noninvasive Entamoeba histolytica by isoenzyme electrophoresis. Trans. R. Soc. Trop. Med. Hyg. 72: Sargeaunt, P. G., J. E. Williams, J. Kumate, and E. Jimenez The epidemiology of Entamoeba histolytica in Mexico City. A pilot survey. I. Trans. R. Soc. Trop. Med. Hyg. 74: Sargeaunt, P. G., J. E. Williams, and R. A. Neal A CLIN. MICROBIOL. REV. comparative study of Entamoeba histolytica (NIH:200, HK9, etc.), "E. histolytica-like" and other morphologically identical amoebae using isoenzyme electrophoresis. Trans. R. Soc. Trop. Med. Hyg. 74: Sathar, M. A., B. L. F. Bredenkamp, V. Gathiram, A. E. Simjee, and T. F. H. G. Jackson Detection ofentamoeba histolytica immunoglobulins G and M to plasma membrane antigen by enzyme-linked immunosorbent assay. J. Clin. Microbiol. 28: Schroeder, S. A., L. M. Tierney, Jr., S. J. McPhee, M. A. Papadakis, and M. A. Krupp Current medical diagnosis and treatment. Appelton & Lange, San Mateo, Calif Seidel, J. S Treatment of parasitic infections. Pediatr. Clin. N. Am. 32: Sepulveda, B., and A. Martinez-Palomo Amebiasis, p In K. S. Warren and A. A. F. Mahmoud (ed.), Tropical and geographical medicine. McGraw-Hill, New York Stanley, S. L., T. F. H. G. Jackson, S. L. Reed, J. Calderon, C. Kunz-Jenkins, V. Gathiram, and E. Li Serodiagnosis of invasive amebiasis using a recombinant Entamoeba histolytica protein. J. Am. Med. Assoc. 266: Strachan, W. D., P. L. Chiodini, W. M. Spice, A. H. Moody, and J. P. Ackers Immunological differentiation of pathogenic and nonpathogenic isolates of Entamoeba histolytica. Lancet i: Tachibana, H., S. Kobayashi, Y. Kato, K. Nagakura, Y. Kaneda, and T. Takeuchi Identification of a pathogenic isolate-specific 30,000-Mr antigen of Entamoeba histolytica by using a monoclonal antibody. Infect. Immun. 58: Talamas-Rohana, P., and I. Meza Interaction between pathogenic amebas and fibronectin: substrate degradation and changes in cytoskeleton organization. J. Cell Biol. 106: Tannich, E., and G. D. Burchard Differentiation of pathogenic from nonpathogenic Entamoeba histolytica by restriction fragment analysis of a single gene amplified in vitro. J. Clin. Microbiol. 29: Tannich, E., F. Ebert, and R. D. Horstmann Primary structure of the 170-kDa surface lectin of pathogenic Entamoeba histolytica. Proc. Natl. Acad. Sci. USA 88: Tannich, E., R. D. Horstmann, J. Knobloch, and H. H. Arnold Genomic DNA differences between pathogenic and nonpathogenic Entamoeba histolytica. Proc. Natl. Acad. Sci. USA 86: Tannich, E., H. Scholze, R. Nickel, and R. D. Horstmann Homologous cysteine proteinases of pathogenic and nonpathogenic Entamoeba histolytica. J. Biol. Chem. 266: Torian, B. E., S. L. Reed, B. M. Flores, C. M. Creely, J. E. Coward, K. Vial, and W. E. Stamm The 96-kilodalton antigen as an integral membrane protein in pathogenic Entamoeba histolytica: potential differences in pathogenic and nonpathogenic isolates. Infect. Immun. 58: Tsutsumi, V., R. Mena-Lopez, F. Anaya-Velazquez, and A. Martinez-Palomo Cellular bases of experimental amebic liver abscess formation. Am. J. Pathol. 117: Vargas, M. A., A. Isibasi, J. Kumate, and E. Orozco Nonpathogenic Entamoeba histolytica: functional and biochemical characterization of a monoxenic strain. Mol. Biochem. Parasitol. 40: Walker, E. L., and A. W. Sellards Experimental entamoebic dysentery. Philipp. J. Sci. 8: Walsh, J. A Problems in recognition and diagnosis of amebiasis: estimation of the global magnitude of morbidity and mortality. Rev. Infect. Dis. 8: Weikel, C. S., C. F. Murphy, E. Orozco, and J. I. Ravdin Phorbol esters specifically enhance the cytolytic activity of Entamoeba histolytica. Infect. Immun. 56: Weinke, T., B. Friedrich-Janicke, P. Hopp, and K. Janitschke Prevalence and clinical importance of Entamoeba histolytica in two high-risk groups: travelers returning from the tropics and male homosexuals. J. Infect. Dis. 161:

14 VOL. 5, Weiss, S. J Tissue destruction by neutrophils. N. Engl. J. Med. 320: Westphal, A Betrachtungen und experimentelle Untersuchungen zur Virulenz der Entamoeba histolytica beim Menschen. Arch. Schiffs. Trop. Hyg. 41: Wittner, M., and R. M. Rosenbaum Role of bacteria in AMEBIASIS 369 modifying virulence of Entamoeba histolytica. Am. J. Trop. Med. Hyg. 19: Young, J. D. E., T. M. Young, L. P. Lu, J. C. Unkeless, and Z. A. Cohn Characterization of a membrane poreforming protein from Entamoeba histolytica. J. Exp. Med. 156: