Swine multifocal ulcerative colitis and crypt abscesses associated with Entamoeba polecki subtype 3 and Salmonella Typhimurium.

Piglets aged approximately 50 days exhibited diarrhea and wasting. Multiple white foci were detected in the colon of a dead piglet; histopathological findings revealed multifocal ulcers and crypt abscesses with Entamoeba trophozoites and gram-negative bacilli in the piglet. These pathogens were identified as Entamoeba polecki subtype 3 and Salmonella enterica serovar Typhimurium, respectively. Numerous E. polecki subtype 3 trophozoites were located on the edge of the ulcerative and necrotic lesions in the lamina propria. Crypt abscesses were associated with S. Typhimurium. These results suggest that E. polecki subtype 3 caused multifocal ulcerative colitis accompanied by crypt abscesses with S. Typhimurium in the piglet. This study is the first report of colitis with E. polecki subtype 3 and S. Typhimurium coinfection.

The genus Entamoeba is an intestinal protist found in humans and other animals [19]. E. histolytica is a common protozoan source of human amebiasis [19] and causes colitis and liver abscesses [17]. E. dispar is considered to be a major protozoan owing to its high prevalence in humans, although it is non-pathogenic [15].

E. polecki and E. suis are associated with chronic diarrhea in pig amebiasis [9,10,11]. E. polecki has four subtypes (ST 1–4), of which ST 1 and 3 are identified in pigs [12]. E. polecki is thought to be non-pathogenic in cases of single infection [8]; however, it has been reported as a coinfection with some other pathogens [3, 8]. Although coinfections of E. polecki ST 3 with Lawsonia intracellularis and Salmonella spp. have been previously reported, the Salmonella serovars have not been identified [8] and coinfection of E. polecki ST 3 with Salmonella Typhimurium has not been reported to date. Furthermore, the pathogenic roles of this protozoan in coinfected diarrhea have yet to be elucidated.

E. polecki has a nucleus and vacuoles, but no mitochondria in the cytoplasm [11, 13]. Engulfed bacteria are also found in E. polecki [7, 11,12,13] and E. histolytica has bacteria only within its vacuoles; in contrast, E. dispar commonly has bacteria in the free cytoplasm [18]. Although the presence of bacteria in these protozoans (within vacuoles or free in the cytoplasm) could be related to pathogenesis [18], the effect of different bacterial locations in these protozoans on their pathogenicity remains unknown. Moreover, little information about E. polecki ST3 is available. An E. polecki transmission electron microscope (TEM) study identified engulfed bacteria in vacuoles in E. polecki ST 1; however, there was no mention of free bacteria in the cytoplasm [11, 13]. To date, there have been no reports of bacteria in the cytoplasm of E. polecki ST 3, although only one TEM study has been conducted [12]. In addition, the previous study did not address the morphological characteristics of E. polecki ST 3 [12].

S. Typhimurium is the most common serovar of pork-related salmonellosis in humans [16]. Multiple antimicrobial resistances associated with S. Typhimurium (related to pork consumption) could become a serious human health hazard [1]. In pigs, salmonellosis is commonly caused by S. Typhimurium and S. Choleraesuis [4].

This study was conducted to elucidate the effect of Entamoeba spp. trophozoites in lesions and describes an E. polecki ST 3 and S. Typhimurium coinfection. To the best of our knowledge, this is the first report of multifocal colitis caused by coinfection with E. polecki ST 3 and S. Typhimurium.

The study site consisted of a farrow to finish farm housing approximately 800 sows, located in Hiroshima Prefecture on Honshu Island, Japan. Twenty-five of 100 piglets aged approximately 50 days exhibited diarrhea and wasting in January 2016. The 25 diseased piglets were placed in an isolation pen on January 22, and penicillin was administered for 3 days. On January 25 and 26, ceftiofur was also administered to the diseased piglets. One piglet died and was subjected to necropsy on January 26.

For the isolation of Salmonella species, the intestinal homogenate was pre-enriched in Hajna tetrathionate broth (Eiken Chemical Co., Ltd., Tokyo, Japan) and incubated at 37°C for 24 hr. A loopful of each isolate was subcultured on deoxycholate hydrogen sulfide lactose agar at 37°C for 24 hr. The isolates were identified as Salmonella spp. based on the Rapid ID 32 E Microbial Identification Kit (bioMérieux Japan Ltd., Tokyo, Japan). Serovar identification was performed using antisera (Denka Seiken, Tokyo, Japan). Antimicrobial susceptibility tests were performed to determine susceptibility using BD BBL Sensi-Discs (Becton, Dickinson and Co., Franklin Lakes, NJ, USA) for 12 antibiotics: ampicillin, amoxicillin, penicillin, ceftiofur, kanamycin, gentamicin, streptomycin, oxytetracycline, sulfamethoxazole and trimethoprim mixture, chloramphenicol, colistin, and enrofloxacin.

For histopathological analysis, formalin-fixed (in 10% neutral-buffered formalin), paraffin-embedded (FFPE) tissue samples were sectioned at approximately 3 µm and stained using hematoxylin and eosin (HE), Gram stain (Brown Hopps method), Periodic Acid Schiff (PAS), Warthin-starry, and Immunohistochemistry staining using the Salmonella O4 (Denka Seiken Co.) and Lawsonia intracellularis antisera [11].

To confirm the presence of engulfed bacteria and the ultrastructural characteristics of E. polecki ST 3, the colon was subjected to TEM observation. The TEM procedure was conducted based on previous report [13].

FFPE samples were also used for molecular identification of protozoa. DNA was purified from three intestinal FFPE samples using the QIAamp DNA FFPE Tissue Kit (QIAGEN GmbH, Hilden, Germany). Polymerase chain reaction (PCR) was performed on the purified DNA to identify Entamoeba spp. (E. histolytica, E. dispar, E. suis, and E. polecki) and subtypes of E. polecki (ST 1–ST 4). These molecular identification procedures were also based on a previous report [8].

The intestinal isolates were identified as S. Typhimurium. Although the S. Typhimurium isolates exhibited resistance to penicillin, kanamycin, streptomycin, and oxytetracycline, they were susceptible to ceftiofur, which was administrated in the present cases.

Multiple white foci were observed in the pig colon (Fig. 1a). Crypt abscesses or ulcers were histopathologically confirmed in these foci. The ulcers were composed of inflammatory cells (primarily neutrophils) and cell debris and pseudomembranes were also detected in the upper section of the necrotic lesion (Fig. 1b). Many PAS-positive Entamoeba trophozoites were detected in the bottom section of the ulcers (Fig. 1c–f). Several gram-negative bacterial masses were also observed in the upper area. Most bacteria did not react with the Salmonella O4 antiserum. Diffuse necrosis of the lamina propria was also present with a high number of Entamoeba trophozoites. Few gram-negative bacteria were detected in the cytoplasm of Entamoeba trophozoites (Fig. 1g). Moreover, Entamoeba trophozoites were found in the lamina propria under the degenerated epithelium accompanied by gram-negative bacteria. Salmonella O4 antigen-positive bacteria and inflammatory cells were observed mainly in the crypt abscesses (Fig. 1h), and some of these were detected surrounding the degenerated epithelium and in the lamina propria by immunohistochemical staining (Fig. 1i). Numerous bacteria around the trophozoites did not react with the Salmonella O4 antiserum. L. intracellularis antigen was not detected in the intestinal specimens by immunohistochemistry.

Fig. 1.

(a) Gross findings in the colon. Many multifocal white foci were found throughout the whole colon; bar=1 cm. (b) Pseudomembrane and ulcer in the colon. The ulcer is composed of inflammatory cells and cell debris. The square inlay is enlarged in Fig. 1e. HE stain; bar=200 µm. (c) Many PAS-positive Entamoeba trophozoites (red) were detected in the bottom region of the ulcers PAS stain; bar=100 µm. (d) Serial section of Fig. 1c. Gram-negative bacterial masses (arrowheads) were observed in the upper right section. Most of the gram-negative bacteria did not react with the Salmonella O4 antiserum.Numerous Entamoeba trophozoites (arrows) were detected in the bottom region of the ulcers. Gram stain (Brown Hopps method); bar=100 µm. (e) Bottom field shown in Fig. 1b. Numerous Entamoeba trophozoites (arrows) were present in the lower section of the ulcer. Crypt structure was maintained below Entamoeba trophozoites. HE stain; bar=20 µm. (f) Serial section of Fig. 1c. Ulcerative colitis. Entamoeba trophozoites (arrows) were present in the bottom section of the necrotic region, while the bacterial masses (arrowheads) were observed in the upper right section. Gram stain (Brown Hopps method); bar=50 µm. (g) Mucosal epithelium of the colon. The epithelium was degenerated and Entamoeba trophozoites (arrows) were observed in the lamina propria. Many bacterial cells (arrowheads) were observed surrounding the degenerated epithelium. Most gram-negative bacteria did not react with the Salmonella O4 antiserum. Warthin-Starry staining; bar=20 µm. (h) A crypt abscess in the colon. Salmonella O4 antigen-positive bacteria (red) were observed in the abscess. Immunohistochemistry; bar=200 µm. (i) Mucosal epithelium of the colon. Salmonella O4 antigen-positive bacteria (arrows) and leukocytes were observed surrounding the sloughed epithelium. Immunohistochemistry; bar=20 µm. (j) Entamoeba trophozoites observed by TEM. Nucleus (N), debris (D) in vacuoles, and bacteria were found in the cytoplasm. Entamoeba trophozoites engulfed debris rather than bacteria. Bacteria were evenly dispersed in vacuoles (arrowheads) and free in the cytoplasm (arrows); bar=5 µm.

Nuclei, bacteria, and vacuoles were observed in the cytoplasm of the trophozoites in the TEM study (Fig. 1j). A few bacteria were found in the Entamoeba trophozoites, evenly distributed in and out of the vacuoles. In addition, a large amount of debris was observed in the vacuoles.

A PCR product was successfully amplified from a colon sample and confirmed (100% identity) as E. polecki ST 3 (accession number: LC067574) by sequencing.

Our investigation indicated that the multifocal ulcerative colitis and crypt abscesses were mainly associated with E. polecki ST 3 and S. Typhimurium, respectively. To our knowledge, an E. polecki ST 3 and S. Typhimurium coinfection has never been reported.

Our report demonstrates that the pathological amebiasis characteristics in the pig were due to E. polecki ST 3. In human amebiasis, E. histolytica is known as a highly pathogenic protozoan [6, 19]. E. histolytica damages the mucosa epithelium and lamina propria owing to secreted molecules such as cysteine proteinase [5]. Infiltration of neutrophils and other cells also necrotize the lamina propria [5]. It is unclear whether or not E. polecki and E. suis are implicated in these pathogenic mechanisms. A high number of E. polecki ST 3 cells were observed at the bottom section of the necrotic layer, indicating that E. polecki ST 3 induced the enlarged necrotic layer. E. histolytica enhances pathogenicity owing to the presence of non-pathogenic Escherichia coli [2, 14]. A high number of bacteria were observed at the upper section of the necrosis area, although E. polecki ST 3 was located below them. These results suggest the presence of several bacteria related to the pathogenicity of E. polecki ST 3, indicating that E. polecki ST 3-induced necrosis could be related to the presence of several bacteria.

E. polecki ST 3 cells engulfed a few bacteria in this study. The amount of engulfed bacteria can correlate with the pathogenicity of E. histolytica and E. dispar [18]. Although little information is available regarding the pathogenicity of E. polecki ST 3 [12], the present findings suggest that E. polecki ST 3 acquires high pathogenicity with a small number of bacteria. E. histolytica is highly pathogenic and is the main cause of human amebiasis [19]; in contrast, E. dispar, commonly found in humans, is thought to have low pathogenicity [15]. E. histolytica has been shown to have several bacteria within vacuoles and few in the cytoplasm [18]. In contrast, in E. dispar, some bacteria were mainly found free in the cytoplasm [18]. In this study, we observed a few engulfed bacteria within vacuoles, but also in the cytoplasm. Some debris were commonly found in the vacuoles, suggesting that E. polecki ST 3 prefers debris to bacteria. Our results indicate that E. polecki ST 3 has different characteristics compared with other Entamoeba spp.; in particular, the amount of engulfed debris in the vacuoles appears to be related to E. polecki ST 3 pathogenicity. Coincubation of E. histolytica with E. coli or Shigella dysenteriae enhances adhesion and cytotoxicity [5]. Sulfur and iron derived from Desulfovibrio desulfuricans and E. coli accelerated the culturing of E. polecki ST 1 [20]. Although the detailed mechanisms of pathogenicity in E. polecki ST 3 remain unknown, it is possible that several bacteria observed at the upper section of the necrotized lesion are related to sulfur and iron generation. Further studies are needed to elucidate the association between pathogenicity and these fast-acting factors.

Bacterial examination showed that the detected S. Typhimurium isolate, which necrotized the epithelium, was sensitive to ceftiofur, which was administered to the piglets in the present case. Most of the past case reports also mentioned that antibiotic therapy had failed [7, 10]. Ceftiofur is a second-choice antibiotic in Japan; thus, early administration was hindered in the present case. Had the antibiotic susceptibility to first choice antibiotics been known earlier, amebiasis could have been prevented, as epithelium degeneration would not have become severe. Adequate hygiene management (especially focused on early detection and treatment) should be defined for farms where E. polecki ST 3 is present.

Our results indicate that E. polecki ST 3 has the ability to cause severe necrotizing ulcerative colitis accompanied by crypt abscesses with S. Typhimurium. Histopathological and molecular examination is needed to detect the underlying cause of amebiasis in pig diarrheal cases. As the number of cases of swine amebiasis is increasing in Japan, additional critical prevention methods and in vitro studies for detecting pathogenic mechanisms are needed.