IVCVM | 2000

Severe Hepatic Necrosis Associated with Entamoebiasis in a Gopher Tortoise (Gopherus polyphemus)

Nancy L. Stedman, AnaPatricia García, and Kenneth S. Latimer

Department of Pathology, College of Veterinary Medicine, The University of Georgia, Athens, Georgia 30602-7388 (USA)

Abstract. A 5-year-old captive male Gopher tortoise died after a two day history of lethargy. The entire right hepatic lobe and most of the left lobe were diffusely necrotic with minimal inflammation and frequent Entamoeba trophozoites consistent morphologically with E. invadens. Gastrointestinal lesions were not detected grossly or histologically. Heavy growth of Aeromonas hydrophila was recovered from the necrotic liver and numerous pinworm ova were detected in a colonic fecal sample. The tortoise was part of a small captive colony with no known medical problems. The source of the E. invadens and factors predisposing this tortoise to severe amebiasis could not be determined.

Key Words. Entamoeba invadens; gopher tortoise; Gopherus polyphemus; amebiasis, hepatic necrosis

Case History

A 5-year-old male Gopher tortoise (Gopherus polyphemus) died in March after a two day history of lethargy. The tortoise was captive raised and part of a colony of seven captive Gopher tortoises with no known medical problems. The diet for the colony consisted of green, leafy vegetables and fruits. The tortoises were offered a pan of water for soaking twice weekly. An adult female Gopher tortoise of unknown age also had died two days prior to the male. This tortoise was a wild caught female introduced to the colony several months previously. She had been captured in the previous fall and quarantined for three weeks prior to being placed with the other tortoises. The adult female had never showed any clinical signs of illness during quarantine or while in captivity with the other tortoises. Necropsy examination of the female tortoise did not reveal any significant lesions to account for her death. The colony of tortoises also had shared an enclosure with several box turtles the previous winter, but had been separated from the box turtles in the spring.

Pathology

The carcass of the male tortoise was in poor body condition with severe fat atrophy and moderate autolysis. The right liver lobe was small, firm, and diffusely light yellow. Multifocal 1 to 5 mm diameter hemorrhages were present throughout the lobe. The left lobe had a central, 3 cm diameter focus of similar appearing tissue surrounded by a thin rim of light orange parenchyma. The intestines contained formed stool with occasional fragments of astroturf. The bladder was distended with clear, light orange urine and urates.

Histologically, the liver had severe, diffuse parenchymal necrosis sharply demarcated from the viable hepatic tissue (Figure 1).

Figure 1. Severe hepatic necrosis (left) consisting of eosinophilic debris and pyknotic nuclei sharply demarcated from viable tissue with blood filled sinusoids. Hematoxylin and eosin.

The necrosis consisted of eosinophilic outlines of cells with severe vacuolar degeneration and vesicular, pyknotic, or karyolytic nuclei. In some areas of the necrosis, only eosinophilic stroma and pyknotic nuclei remained. Occasional necrotic cells contained melanin granules, interpreted as necrotic melanomacrophages. Vessels within the affected liver had diffuse mural necrosis and contained eosinophilic cell debris. The viable parenchyma had numerous PAS positive protozoal trophozoites within sinusoids and less frequently hepatic venules (Figures 2 and 3).

Figure 2. Trophozoites within sinusoids. Hepatocytes also have diffuse lipidosis. Hematoxylin and eosin. Figure 3. Abundant PAS positive material consistent with glycogen within sinusoidal trophozoites. PAS with hematoxylin counterstain.

The trophozoites measured between 11 and 16 microns in length and contained a single, large vesicular nucleus and abundant, densely packed cytoplasmic vacuoles. Frequent trophozoites contained a large cytoplasmic vacuole of homogeneous, PAS positive eosinophilic material compressing the nucleus to the periphery of the cell. Cysts were not observed. Occasional trophozoites were also present among the necrotic tissue (Figure 4).

Figure 4. Trophozoites within remnants of sinusoids in necrotic parenchyma. Necrotic tissue consists of eosinophilic stroma and cell debris with pyknotic nuclei.

The viable parenchyma had prominent hepatic lipidosis with mild, multifocal perivascular and periportal infiltrates of lymphocytes, plasma cells, and heterophils. Heterophils were also prominent within vessels. Trophozoites and inflammatory infiltrates were most frequent within the viable tissue adjacent to the necrosis. Small clusters of heterophils were occasionally infiltrating the necrotic parenchyma. Rare bacterial rods were present among the necrotic cell debris but not observed in the viable parenchyma. The spleen contained numerous plasma cells, occasional Mott cells, and frequent heterophils in the white pulp. One small splenic arteriole contained a small colony of bacilli. Sections of lung, heart, kidney, and the gastrointestinal tract were histologically unremarkable.

Additional findings

A fecal flotation recovered numerous pinworm ova from a colonic fecal sample. Heavy growth of Aeromonas hydrophila was recovered from the necrotic liver.

Discussion

Pathogenic reptilian amebiasis involving an Entamoeba histolytica like organism was first described during an outbreak at the Philadelphia Zoological Gardens involving a monitor lizard, blue tongued skink, milksnake, pseudoboa, and several water snakes. 12,14 Affected animals had intestinal, and less frequently gastric, mucosal erosions and ulcerations with multifocal liver abscesses associated with the presence of the amoeba. Morphologically, the organism was determined to be Entamoeba invadens (Rodhain 1934), and experimental infection confirmed its pathogenicity in snakes and lizards but found the organism to be nonpathogenic in turtles. 4,5,13,15 E. invadens is considered one of the most important parasitic diseases of reptiles and has also caused outbreaks of intestinal and hepatic amebiasis with high mortality in captive snakes and red footed tortoises (Geochelone carbonaria). 3,6 Lesions in these outbreaks consisted of erosions, ulcerations, and necrosis in various gastrointestinal segments with periportal or multifocal hepatic necrosis or abscessation associated with the presence of trophozoites. Turtles and alligators in one outbreak were not affected despite the presence of throphoozoites in their feces, implicating these animals as potential carriers of E. invadens. 3

The pathogenic mechanisms of E. invadens are not well described. Numerous studies have investigated E. histolytica pathogenesis with some comparative studies also focusing on E. invadens. Trophozoites of E. histolytica induced contact dependent cytopathic effects in MDCK cells within minutes after addition to the cultures. 8 Trophozoites induced separation of cells from each other and the substrate, and plasma membrane blebbing. Detached cells were commonly phagocytosed by the trophozoites. Less frequently, trophozoites lysed MDCK cells by "pinching off" a portion of the plasma membrane. E. invadens induced epithelial erosion and edema within two hours after addition to guinea pig cecal explants. 11 Trophozoites adhered to epithelial cells, inducing mild superficial lamina propria edema and some epithelial cell desquamation. Trophozoites invaded the mucosa at sites of epithelial cell loss, inducing more desquamation until the explants exhibited severe mucosal erosion and edema of the lamina propria. In contrast to E. histolytica, trophozoites of E. invadens only occasionally phagocytosed cell debris.

Some specific mechanisms of cytotoxicity and invasion have been investigated. Both E. histolytica and E. invadens have thiol dependent protease activity that may be involved in cytotoxicity. 1 Trophozoites of both E. histolytica and E. invadens induced chromium release from cultured human hepatocytes, indicating plasma membrane damage. 9 A phospholipase was proposed as the mechanism of cytotoxicity, since phospholipase inhibitors could reduce this effect. E. invadens trophozoites induced pore formation in lipid bilayers and liposomes and some lysis of human erythrocytes, but were generally much less active than E. histolytica trophozoites. 7 However, pathogenic and nonpathogenic E. histolytica were similar in their pore forming and hemolyzing activities, indicating these factors may not be directly responsible for virulence. Trophozoites of E. invadens bound also more reptilian erythrocytes than human erythrocytes, whereas E. histolytica trophozoites bound more human than reptilian erythrocytes. 7 The increased ability of E. invadens to bind reptilian erythrocytes may be related to its pathogenicity in reptiles.

Host and environmental factors also may contribute to the pathogenicity of E. invadens. Experimentally infected water snakes, Dekayi and red belly snakes, garter snakes, ribbon snakes, milk snakes, green snakes, and ringneck snakes failed to develop lesions when maintained at 13C although E. invadens could be recovered from the intestinal tract. 2 In contrast, all species developed significant and often fatal intestinal and liver lesions with intralesional E. invadens when maintained at 25C. The severity and distribution of the liver and intestinal lesions varied with species of snake. Experimentally, encystation of the trophozoites (to allow completion of its life cycle) was dependent on the presence of adequate particulate, ingestible polysaccharides or mucopolysaccharides, and the presence of living bacteria. 10 The author speculated that herbivorous turtles provide adequate plant derived polysaccharides in their intestinal tracts for E. invadens to complete its life cycle without any pathologic effects. In carnivorous snakes, the trophozoites would have to consume mucus secreted in the gastrointestinal tract to obtain carbohydrate, thereby denuding the wall of a protective mucus coating and allowing invasion of the wall by both trophozoites and secondary bacteria. Possibly the differences in intestinal polysaccharide content account for some of the host differences in E. invadens pathogenicity. However, this speculation does not explain why this organism was apparently nonpathogenic in alligators in one outbreak.

The morphology of the amoeba and hepatic pathology in this case are typical for E. invadens. However, this case is unusual since intestinal pathology was not detected grossly or histologically. Since the intestinal lesions of E. invadens can be segmental and differ among reptilian species, perhaps lesions were not present in areas sampled for histopathology. The hepatic necrosis is this tortoise was more severe and widespread in distribution than in previously described cases. This case may represent a more chronic form of the infection in which the intestinal lesions may have had some time to resolve. The histopathology of the liver is consistent with infarction. Although thrombi were not observed in this case, thrombi have been described in hepatic necrosis associated with E. invadens. 6 This case is also unusual since only one animal in the colony was affected. The factors contributing to the development of pathogenic amoebiasis in this tortoise are unknown. It is interesting that this tortoise died in March, similar to spring outbreak previously reported in red footed tortoises. The authors speculated that increasing temperatures in the spring may have contributed to the pathogenicity of E. invadens, similar to the temperature effect described experimentally in snakes. 2,6 Also, the tortoise in this case had a heavy concurrent pinworm infestation, similar to the affected red footed tortoises in a published outbreak. 6 Although pinworms are extremely common in chelonians and of questionable clinical significance, possibly a heavy infestation may have induced sufficient mucosal damage to allow invasion by E. invadens trophozoites in these cases. The source of the E. invadens was never identified in this case. Since turtles are considered carriers, exposure to the box turtles during the winter is the most likely source of the E. invadens. The wild caught female gopher tortoise may also have been an asymptomatic carrier, contaminating the soaking water with infectious cysts. Histologically, Entamoeba sp. was not detected in any of her tissues, although culture would have been a more sensitive method to detect E. invadens. The presence of Aeromonas hydrophila in the necrotic liver in this case is also not surprising, since secondary colonization of the liver with intestinal organisms is frequent in E. invadens infection . 6

References

1. Avila EE, Sanchez-Garza M, Calderon J: Entamoeba histolytica and E. invadens: sulfhydryl dependent proteolytic activity. J Protozool 32: 163-166. 1985.

2. Barrow J Jr. , Stockton JJ: The influences of temperature on the host parasite relationships of several species of snakes infected with Entamoeba invadens. J Protozool 7: 377-383. 1960.

3. Donaldson M, Heyneman D, Dempster R, Garcia L: Epizootic of fatal amebiasis among exhibited snakes: epidemiologic, pathologic, and chemotherapeutic considerations. Am J Vet Res 36: 807-817. 1975.

4. Geiman QM: Cross infection experiments with three species of amoebae from reptiles. J Parasitol 23: 557. 1937.

5. Geiman QM, Ratcliffe HL: Morphology and life cycle of an amoeba producing amoebiasis in reptiles. Parasitol 28: 208-228. 1936.

6. Jacobson E, Clubb S, Greiner E: Amebiasis in red-footed tortoises. J Am Vet Med Assoc 183: 1192-1194. 1983.

7. Keller F, Walter C, Lohden U, Hanke W, Bakker-Grunwald T, Trissl D: Pathogenic and nonpathogenic Entamoeba: pore formation and hemolytic activity. J Protozool 35: 359-365. 1988.

8. Martinez-Palomo A, Gonzales-Robles A, Chavez B, Orozco E, Fernandez-Castelo S, Cervantes A: Structural bases of the cytolytic mechanisms of Entamoeba histolytica. J Protozool 32:166-175. 1985.

9. McCaul TF, Poston RN, Bird RG: Entamoeba histolytica and Entamoeba invadens: chromium release from labeled human liver cells in culture. Exp Parasitol 43: 342-352. 1977.

10. Meerovitch E: Some biological requirements and host parasite relations of Entamoeba invadens. Can J Zool 36: 513-523. 1958.

11. Mora-Galindo J, Ramirez-Romo S, Jauregui-Topete M: Pathogenic action of Entamoeba invadens: intestinal epithelium invasion by trophozoites in vitro. Int J Parasitol 26: 297-302. 1996.

12. Ratcliffe HL, Geiman QM: Eleven cases of amoebiasis in reptiles. J Parasitol 20: 139. 1934.

13. Ratcliffe HL, Geiman QM: Experimental amebic infection in reptiles. J Parasitol 20: 328-329. 1934.

14. Ratcliffe HL, Geiman QM: Amebiasis in reptiles. Science 79: 324-325. 1934.

15. Rodhain J: Entamoeba invadens, parasite de serpents. C R Soc Biol 117: 1195. 1934.

This Page Last Updated October 16, 2000

 

List of Papers | IVCVM Menu | Pathology Department | College of Veterinary Medicine | UGA

The content and opinions expressed on this Web page do not reflect the views of nor are they endorsed by the administration of the University of Georgia or the University System of Georgia.