Cochlosoma anatis is a protozoal organism found in various avian species. Kotlan first described C. anatis from the intestinal contents of European domestic ducks in 1923, which was later reviewed by Travis (13,25). The trophozoite stage of this parasite is ovoid to pyriform in shape and 6–12 μm long by 4–7 μm wide with six flagella, an axostyle, lateral groove, and undulating membrane (2,15). Trophozoites are uninucleate and have a prominent ventral adhesive disk (2,6). The disk serves to adhere the parasite to the intestinal mucosa of the host (6,21,29). Attachment to the microvillus border of the intestines appears to be the main host–parasite interaction; however, whether pathogenicity is due to destruction of the microvilli or mechanical blockage of nutrient absorption is unclear (2,6,21,29).

Cochlosoma species have been observed to infect turkeys, chickens, bobwhite quail, songbirds, waterfowl, bats, and shrews (2,4,5,6,7,14,15,20). The trophozoites may infect the whole intestinal track but can also be restricted to one or several locations (4,6,15,20,25,29). Oral transmission has been demonstrated for turkeys, chickens, bobwhite quail, and ducks. Trophozoites are shed in fecal material at approximately 5–7 days postinoculation (4,15). Infected turkeys are depressed, have ruffled feathers, and may have diarrhea resulting in fluid-filled, swollen small intestines and ceca, often observed during necropsy (2,6,19).

Cochlosoma anatis infections have been linked to runting of ducklings, enteritis in turkeys, and mortality, dehydration, and malabsorption of young finches (4,6,16,22). Cooper et al. reported that a 16% decrease in body weight was associated with C. anatis infection in turkeys (6). Reduced fat absorption has been observed in turkey poults infected with C. anatis, but final weight and performance parameters were not affected (3). Cochlosoma is often present with other intestinal pathogens, and coinfection of C. anatis and turkey coronavirus has been shown to be more pathogenic for infected turkeys than an infection with either pathogen alone (24). There is no known procedure for purifying or culturing this parasite, which makes studying C. anatis difficult (2,27). Advantageous conditions for C. anatis are current pathogenic intestinal infections or young age.

Morphologic similarities between Giardia, Trichomonas, and Cochlosoma have been well documented (6,20,21). The adhesive disk is the main characteristic that Giardia and Cochlosoma share (2,14); however, Pecka et al. demonstrated ultrastructural differences in the adhesive disk of these species and concluded that, based on detailed ultrastructural analysis, Cochlosoma has a closer relationship with trichomonads (21). Both Cochlosoma and Trichomonas are uninucleate and have a parabasal apparatus, a tubular axostyle, and a crescent-shaped pelta (14). Recently, Vrlik et al. confirmed, by sequence analysis of the16S rRNA gene, that Cochlosoma is genetically similar to Trichomonas (26).

Cochlosoma has been observed to divide by longitudinal binary fission of the trophozoite, which is consistent with most species from the class Zoomastigophorea (2,25). Kotlan observed longitudinal division and a cyst stage, but did not publish drawings or dimensions, and there are no other detailed descriptions of a C. anatis cyst or pseudocyst stage (2,13,25). This article presents evidence supporting longitudinal binary fission of the trophozoite and demonstrates pseudocyst formation of C. anatis.

MATERIALS AND METHODS

Several poultry farms with a history of C. anatis infections were identified for sampling. Turkey poults were euthanatizied and intestinal scrapings from the area around Meckel diverticulum were collected. These scrapings were mounted on slides and examined with light microscopy for the presence of C. anatis and the absence of other protozoa (11). Samples containing C. anatis were diluted in saline solution, then stored on ice for 24 hr and re-examined by light microscopy. No trophozoites were observed, and only pseudocyst-like stages were noted. Using this isolate, a 1-day-old poult was gavaged with 1 ml of the sample and then housed with five other poults of the same age. After 48 hr, all six of the turkeys were necropsied and sampled as previously described. Chilled, preservative-free saline, containing boric acid, was added to the sample at a 1 : 1 ratio. The sample was then left at room temperature for approximately 20 min and then fixed with 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer.

RESULTS

After storage on ice for 24 hr, no trophozoites were seen when examined by light microscopy; however, pseudocyst-like stages were observed. Electron microscopy observations suggest that the flagella and other external structures were internalized and the observed organisms were lacking a true cyst wall. Infected poults housed with noninfected hatch mates for a 48-hr incubation period resulted in all of the turkeys acquiring C. anatis infections.

DISCUSSION

After a 48-hr incubation period, a turkey poult gavaged with a pseudocyst-containing sample developed an active C. anatis infection. This demonstrates that the pseudocyst form of C. anatis may be capable of producing an infection when contracted orally by young poults. Five birds housed with an experimentally infected bird all became infected with C. anatis. This data suggests that organisms shed in the fecal material of poults infected with the pseudocyst stage are capable of infecting other poults.

SEM observations of trophozoites are consistent with published reports on the external morphology of C. anatis (2,6,15,21). The images show that there is a distinct intermediate stage between the trophozoite and pseudocyst stage. The rounding stage serves as this intermediate stage, in which external structures are internalized inside the cell. The adhesive disc and axostyle tip change shape, and the cell becomes more circular. The images indicate that the flagella were internalized starting from the base and working toward the tip. The undulating membrane appears to be internalized as well. The pseudocyst stage had no external structures, but a small structure protruded from the side of the otherwise round cell membrane.

TEM observations of trophozoites presented here confirm observations by Pecka et al. in 1996 (21). Trophozoites have been reported to divide by longitudinal binary fission, but little evidence was provided to support this theory (2,13,21,25). Images collected in this study support the theory and provide evidence that the trophozoites replicate and divide by longitudinal binary fission. Organelles and support structures are divided among the daughter cells evenly, and nuclear material contained in the nucleus appears to have replicated and dispersed to individual nuclei during this process.

TEM observations of the rounding stage support the theory that this is an intermediate stage in which external structures are internalized and compartmentalized inside the cell. The cytoplasm stains a lighter shade during this stage; the cause of this is still unknown. Organelles, such as vacuoles, the hydrogenosome-like organelles, nucleus, costa, and the adhesive disc, which are found in the trophozoite, can be found in the rounding stage. The undulating membrane and lateral groove are not seen during this stage. The adhesive disc and costa appear to be relocating from the exterior of the cell and condensing in one central location on the interior of the cell.

Examination of the pseudocyst stage showed that the costa, adhesive disc, and anterior flagella were internalized and compartmentalized in the cytoplasm. The nucleus, hydrogenosome-like organelles, and vacuoles were still visible in the cytoplasm during this stage. The structures all appeared to be intact; however, the undulating membrane and lateral groove were not seen. The absence of these structures on the surface of the cell reduces the surface area, effectively reducing exposure to harmful environmental factors (9). Evidence of the recurrent flagellum was often seen on the exterior of the cell. This may indicate that the recurrent flagellum is slow to be internalized. Granger et al. reported similar findings for Tritrichomonas fetus. When the trophozoite form of the cells were cooled to 0 C and then left at room temperature, the anterior flagella were internalized more rapidly than the recurrent flagella (9). This newly described stage of C. anatis is lacking a true cyst wall and is best classified as a pseudocyst.

The process of pseudocyst formation of C. anatis may be a reversible event, taking place with or without cell division. Granger et al. presented data showing that trophozoites of T. fetus underwent pseudocyst formation when the cells were cooled to 0 C for 60 min and then rewarmed to 37 C. T. fetus cells internalized their flagella within 3 min and then gradually externalized them, and TEM data showed the pseudocyst stage of T. fetus dividing after a similar treatment (9). Pseudocyst formation, which refers to compact, nonmotile forms without a true cyst wall, has also been described for many other gastrointestinal trichomonads (1,8,9,10,12,17,18,23,28). As noted, C. anatis has been reported to be closely related to trichomonads and may undergo similar pseudocyst formation (6,14,20,21,26). Pseudocyst formation is most likely a self-preservation mechanism that allows C. anatis to survive subphysiological temperatures and/or other adverse environmental conditions, until a suitable host ingests them. Vulnerable structures like the adhesive disc and the undulating membrane are internalized to decrease the surface area that is exposed to environmental stress. The flagella also appear to undergo internalization, but more data are required to confirm this process. Because these structures are attached or continuous with the plasma membrane, damage to one of these structures would be detrimental to the cell (1). Further studies are necessary to understand the mechanism by which C. anatis trophozoites undergo longitudinal binary fission and pseudocyst formation and what action can be taken to control the infectivity of this organism.