Molecular and histopathological characterization of Cryptosporidium and Eimeria species in bats in Japan

Abstract

Bats are potential reservoirs of Cryptosporidium and Eimeria. The genus Cryptosporidium infects various vertebrates and causes a diarrheal disease known as cryptosporidiosis. Many epidemiological studies in wild animals have been performed; however, most of them relied on only PCR-based detection because of the difficulty of performing pathological analyses. Accordingly, the natural host and pathogenicity of Cryptosporidium bat genotypes remain unclear. In this study, we captured Eptesicus nilssonii (Northern bats) in Hokkaido, Japan. Of the three intestinal samples obtained, two were positive for Cryptosporidium spp. and one was positive for Eimeria spp. The corresponding microorganisms were also confirmed histopathologically. We detected the novel Cryptosporidium bat genotype XII and Eimeria rioarribaensis in bat intestine.

Wild animals, especially bats, are found throughout the world [5], although some species of bat are now endangered. Bats have been implicated as potential reservoirs of many pathogens [1, 12, 18]. Bat Cryptosporidium was first reported in the big brown bat (Eptesicus fuscus) in the United States [2]. Subsequently, a Cryptosporidium sp. closely related to the Cryptosporidium mouse genotype was identified in a fecal sample from a large-footed mouse-eared bat (Myotus adversus) in Australia [11]. In China, two Cryptosporidium genotypes were identified (bat genotypes I and II) from Rhinolophus sinicus, Hipposideros fulvus, Rousettus leschenaultia, and Aselliscus stoliczkamus [19]. Bat genotypes III and IV were identified from Eptesicus fuscus and Pipistrellus pipistrellus in the US and the Czech Republic, respectively [9]. In the Philippines, three Cryptosporidium bat genotypes (V–VII) were detected from Rhinolophus inops, Cynopterus brachyotis, and Eonycteris spelaea [13]. Four further Cryptosporidium bat genotypes (VIII–XI) and C. hominis were described in Pteropus poliocephalus in Australia [15]. Despite the potential risk to public health imposed by these pathogens, there have been few pathological or molecular epidemiological studies. Therefore, the natural host and pathogenicity of the Cryptosporidium bat genotypes remain unclear.

Eimeria is also a protozoan parasite identified in bats. Traditionally, the identification of Eimeria species has relied primarily on oocyst morphology and host specificity. There are no pathological or molecular epidemiological studies of the Eimeria infected Japanese bats.

In this study, we detected Cryptosporidium and Eimeria from the Northern bat (Eptesicus nilssonii). We confirmed that a novel Cryptosporidium bat genotype (i.e., genotype XII) and E. rioarribaensis infected bat intestine.

During 2015, three bats (BT1-3), Eptesicus nilssonii (Northern bat), were captured from the same colony located in Tokachi district, Hokkaido, Japan. Species of the bats was confirmed based on morphology. The capture and handling of bats were performed under the license from the Japanese Ministry of Environment (license No. 21-27-0213). The protocol for the experiments was approved by the Committee on the Animal Experiments of the Obihiro University of Agriculture and Veterinary Medicine (Approval number: 260086). The bats did not show any clinical symptom. All animal work has been conducted according to the national guidelines of Japan. The captured bats were euthanized and their intestinal tissues were collected. Tissue samples were immediately immersed in 4% formaldehyde for histopathological analysis or RNAlater (Thermo Fisher Scientific, St. Louis, MO, U.S.A.) and then frozen at −80°C until DNA extraction. The captured bats were identified as Eptesicus nilssonii (Northern bat) based on morphology.

DNA was extracted from the intestine by using the Biomasher IV (Funakoshi, Tokyo, Japan) and NucleoSpin^® Tissue (Macherey-Nagel, Diiren, Germany). For Cryptosporidium detection, ~830-bp and ~850-bp fragments of 18S rRNA and actin genes, respectively, were amplified by using KOD FX Neo (TOYOBO, Osaka, Japan) with the primers described previously [4, 17]. For Eimeria detection, 18S rRNA and plastid 23S rRNA gene products of ~1,700 bp and ~1,100 bp were amplified, respectively. 18S rRNA gene was amplified with primers as described previously [7] and the plastid 23S rRNA gene was amplified with forward and reverse primers: Eim23SF (5′-AGGAGCGTTCTATATTTARGAAG-3′) and Eim23SR (5′-GGATCATTAAGACCGACTTCG-3′). All PCR products were cloned (Zero Blunt TOPO PCR Cloning Kit, Thermo Fisher Scientific, St. Louis, MO, U.S.A.) and sequenced in both directions by using an ABI 3130 Genetic Analyzer (Applied Biosystems Japan, Tokyo, Japan). Sequences were aligned by using Clustal X2 [10]. All gaps were eliminated. Maximum likelihood (ML) analyses were performed by using MEGA 7.0.25 [8]. Substitution models and optional parameter sets were also evaluated by using MEGA 7.0.25. We constructed a phylogenetic tree (Figs. 1 and 2), in which the substitution model and optional parameters used were the General Time Reversible model and Hasegawa-Kishino-Yano model [6, 14].

Fig. 1.

Phylogenetic trees based on partial sequences of the 18S rRNA and actin genes for Cryptosporidium spp. The black marks indicate that sequences detected in this study. The white marks indicate bat genotype clades including sequences detected in this study. Phylogenetic trees based on partial sequences of the 18S rRNA (A, B) and actin genes (C) constructed by ML for Cryptosporidium spp. using 712 (18S rRNA gene) and 724 (actin gene) nucleotides without gaps. B is a higher magnification of A. Substitution model and optional parameters=GTR+Γ+I. Only bootstrap values >50% from 500 pseudo-replicates are shown.

Fig. 2.

Phylogenetic trees based on partial sequences of the 18S rRNA and plastid 23S rRNA genes for Eimeria spp. Phylogenetic trees based on partial sequences of the 18S rRNA (A) and plastid 23S rRNA genes (B) constructed by ML for Eimeria spp. using 1495 (18S rRNA gene) and 1012 (plastid 23S rRNA gene) nucleotides without gaps. Substitution model and optional parameters: (A)=GTR+Γ (B)=HKY+Γ. Only bootstrap values >50% from 1,000 pseudo-replicates are shown.

The fixed small intestines were divided into three parts (upper, middle, and lower), trimmed, embedded in paraffin, and cut into 3-µm-thick sections. Paraffin sections were stained with haematoxylin and eosin (HE). For Cryptosporidium detection, selected sections were subjected to immunofluorescence staining using the Sporo-Glo Cy3 Kit, which contains polyclonal IgG against Cryptosporidium parvum (A600Cy3-R-1X, Waterborne, LA, U.S.A.). Microwave antigen retrieval (15 min in 0.01 M citrate buffer, pH 6.0) was also performed. After incubation with Sporo-Glo at 4°C overnight, the sections were washed with phosphate-bufferd saline, mounted with Vectashield (H-1200, Vector Laboratories, Burlingame, CA, U.S.A.), and examined under a BZ-X700 fluorescence microscope (Keyence, Osaka, Japan).

Of the three intestinal samples, two were positive for Cryptosporidium spp. and one was positive for Eimeria spp. Fragments of the 18S rRNA and actin nucleotide sequences of the Cryptosporidium, 18S rRNA and plastid 23S rRNA nucleotide sequences of the Eimeria acquired in this study were deposited in GenBank (LC276360–LC276363 and LC371915–LC371916), respectively. Based on the blast search, all of the sequences detected in this study were identifiable as either Cryptosporidium or Eimeria spp. Of the Cryptosporidium 18S rRNA and actin gene sequences obtained, the 18S rRNA sequences from BT1 and BT3 were identical. Actin sequence from BT1 was not amplified by PCR. The sequence similarity between BT1 and the other bat genotypes belonging to the same clade (Fig. 1: white marks) ranged from 95 to 98% (Table 1). For the Eimeria 18S rRNA gene sequences, BT2 had 99% (822/823) identity with E. rioarribaensis (AF307877) (bat Eimeria).

Table 1. Nucleotide identity among 18S rRNA sequences of Cryptosporidium genotypes in a bat clade.

Genotype	Ref	Bat species in whichCryptosporidium was detected	I (Rh)	I (As)	III	IV	VII	VIII	IX	X	XI
Bat genotype I (KC445650)	[19]	Rhinolophus sinicus
Bat genotype I (KC445654)	[19]	Aselliscus stoliczkanus	99
Bat genotype III (KR819167)	[8]	Eptesicus fuscus	96	96
Bat genotype IV (KR819168)	[8]	Pipistrellus pipistrellus	97	97	96
Bat genotypeVII (LC089979)	[13]	Rhinolophus inops	95	95	95	96
Bat genotype VIII (KX118594)	[15]	Pteropus poliocephalus	95	95	96	96	95
Bat genotype IX (KX118595)	[15]	Pteropus poliocephalus	96	95	97	97	95	98
Bat genotype X (KX118596)	[15]	Pteropus poliocephalus	96	96	97	97	96	98	98
Bat genotype XII (BT1) (LC276360)	this study	Eptesicus fuscus	96	96	97	98	96	96	97	97

Rh: Rhinolophus sinicus, As: Aselliscus stoliczkanus.

Figure 1 shows the phylogenetic relationships based on the partial sequences of 18S rRNA and actin among the various Cryptosporidium species and genotypes, including the sequences from the Northern bats captured and analyzed in this study. Figure 1B is a higher magnification of Fig. 1A. In Fig. 1B, the Cryptosporidium sequences obtained from BT1 and BT3 formed a single clade with the known Cryptosporidium bat genotypes (I, III, IV, VII, VIII, IX, and X; white mark). Monophyly among these branches was moderately supported (51%). In Fig. 1C, the actin nucleotide sequence from BT3 made a clade with Cryptosporidium bat genotype IV. A phylogeny inferred from the actin sequences was consistent with the phylogenetic relationships based on the partial sequences of the 18S rRNA.

Figure 2 shows the phylogenetic relationships based on partial sequences of 18S rRNA (Fig. 2A) and plastid 23S rRNA (Fig. 2B) genes among various Eimeria spp., including the sequences from the Northern bats in this study. In Fig. 2A and 2B, the Eimeria sequences detected from BT2 formed a single clade with E. rioarribaensis.

In BT1 and BT3, Cryptosporidium-like micro-protozoan organisms were observed sparsely as minute basophilic round organisms on and within the apical surface of the enterocytes throughout the small intestine (Fig. 3A and 3B). These organisms were positively stained on immunostaining with an anti-Cryptosporidium sporozoite antibody (Fig. 3B, inset). The antibody response was not observed in BT2. There were no other findings, such as villus atrophy or inflammation. In BT2, Eimeria-like protozoan organisms at various life cycle stages, including macrogametocytes, microgametocytes, oocysts, and meronts, were scattered in the mucosal epithelium and in the lamina propria mucosae throughout the small intestine (Fig. 3C and 3D). As with BT1 and BT2, no significant lesions indicative of an inflammatory response were observed in the small intestine.

Fig. 3.

Cryptosporidium and Eimeria organisms in the small intestine of bats. (A) Small intestine of BT3. Cryptosporidium (arrows) can be seen as minute basophilic round organisms on and within the apical surface of enterocytes. Inflammation was not noted. HE. Bar=25 µm. (B) Higher magnification of (A). HE. Bar=10 µm. Inset. Organisms showing a positive reaction on immunostaining with an anti-Cryptosporidium parvum antibody. (C) Small intestine of BT2. A macrogametocyte (black arrow) and a microgametocyte (white arrow) of Eimeria are shown. HE. Bar=10 µm. (D) Small intestine of BT2. An oocyst of Eimeria is shown. HE. Bar=10 µm.

In this study, we analyzed Cryptosporidium and Eimeria infection in Eptesicus nilssonii (Northern bat). Phylogenetic analyses revealed that the BT1 and BT3 sequences were closely related to a bat genotype clade (Fig. 1A and 1B: white marks), but the branches were not well supported (51%) because of the short reference sequence length. Actin sequences were not available for bat genotypes I, III, VII, VIII, IX, and X. The bat genotypes of this clade (Fig. 1A and 1B: white marks) were detected from Rhinolophus sinicus, Aselliscus stoliczkanus, Eptesicus fuscus, Pipistrellus pipistrellus, and Rhinolophus inops in previous studies in China, the United States and the Czech Republic, the Philippines, and Australia [9, 13, 15, 19]. Minimal intra-clonal variation between bat genotypes I has been reported with nucleotide similarity ranging from 99.4 to 99.8% [15]. In the present study, the sequence similarity between BT1 and the other bat genotypes was between 95% and 98%; therefore, our data suggest that the BT1 and BT3 sequences represent a novel genotype, namely Cryptosporidium sp. bat genotype XII. This study represents the first detection of Cryptosporidium from bats in Japan. Histopathologically, micro-protozoan organisms like ordinal Cryptosporidium species were detected in the absence of indicators of inflammatory response or tissue damage. Taking into account that the captured bats did not show any clinical symptoms, this Cryptosporidium species appears to be non-pathogenic for healthy bats. We thus confirmed that bat genotype XII (this study) could infect bat intestine and showed no pathogenicity. The results of the PCR and the histopathological finding stained with haematoxylin and eosin (HE) were corresponded with the immunofluorescence staining results, therefore, we also confirmed that the Cryptosporidium antigen cross-reacted with a commercially available anti-Cryptosporidium parvum sporozoite polyclonal IgG on immunostaining.

The Eimeria sequences detected from BT2 was that of E. rioarribaensis. E. rioarribaensis was first detected from Myotis ciliolabrum in North America [3, 16]. We also observed that the macrogametocyte, microgametocyte, oocyst, meront in the intestine. Therefore, it was confirmed that Eptesicus nilssonii was one of the natural host of E. rioarribaensis.