Molecular typing of Giardia duodenalis isolates from nonhuman primates housed IN a Brazilian zoo.

Giardia infections in captive nonhuman primates (NHP) housed at a Brazilian zoo were investigated in order to address their zoonotic potential. Fresh fecal samples were collected from the floors of 22 enclosures where 47 primates of 18 different species were housed. The diagnosis of intestinal parasites after concentration by sedimentation and flotation methods revealed the following parasites and their frequencies: Giardia (18%); Entamoeba spp. (18%); Endolimax nana (4.5%); Iodamoeba spp. (4.5%); Oxyurid (4.5%) and Strongylid (4.5%). Genomic DNA extracted from all samples was processed by PCR methods in order to amplify fragments of gdh and tpi genes of Giardia. Amplicons were obtained from samples of Ateles belzebuth, Alouatta caraya, Alouatta fusca and Alouatta seniculus. Clear sequences were only obtained for the isolates from Ateles belzebuth (BA1), Alouatta fusca (BA2) and Alouatta caraya (BA3). According to the phenetic analyses of these sequences, all were classified as assemblage A. For the tpi gene, all three isolates were grouped into sub-assemblage AII (BA1, BA2 and BA3) whereas for the gdh gene, only BA3 was sub-assemblage AII, and the BA1 and BA2 were sub-assemblage AI. Considering the zoonotic potential of the assemblage A, and that the animals of the present study show no clinical signs of infection, the data obtained here stresses that regular coproparasitological surveys are necessary to implement preventive measures and safeguard the health of the captive animals, of their caretakers and of people visiting the zoological gardens.

INTRODUCTION

Today, in view of the increasing interest for biodiversity conservation, wildlife species have been subject to many investigations; however the search for knowledge goes beyond the behavioral and ecological aspects of these animals and includes research on their infectious diseases and their role in public health. For this purpose, studies have been carried out on the occurrence of enteric parasites in a range of wild mammals' species, many of them kept in zoological gardens, conservation parks or sanctuaries, where captivity may raise their exposure to conditions that facilitate the spread of these pathogens[4,5,7,9,19,21]. In this context, it is important to stress that a wide range of captive species may be important reservoirs for zoonotic parasites acting as potential sources of human and animal infections and environmental contamination.

The enteric protozoan Giardia duodenalis (syn. G. intestinalis, G. lamblia) is one of the most common pathogens of zoonotic importance that infect humans and domesticated animals, in addition to many other mammalian wildlife species, including free-living and captive animals[1]. Currently, genetic analyses have shown that human and animal isolates can be clustered into at least eight groups, referred to as genotypes A to H[17,23], but only assemblages A and B are known to infect humans, as well as a variety of other mammals. Based on the data on the prevalence of giardiasis in different hosts and on existing knowledge of the genotypes of G. duodenalis, THOMPSON proposed the existence of transmission cycles that keep the infection among the mammals[31]. Even so, it has not been fully elucidated how these cycles interact and how often the transmission of zoonotic assemblages occurs.

Despite the interest of researchers, as yet, genotyping data from wildlife mammals' species are scarce. Lately, APPELBEE et al. highlighted that the infection of free-living and captive wildlife mammals by both host-adapted and zoonotic isolates of Giardia can provide insights into the host range and zoonotic potential of this parasite[1]. An overview of recent studies enables us to ascertain that almost all mammalian orders include free-living and captive species that can harbor Giardia isolates[2,18,26]. Among these groups, Giardia isolates have been obtained from nonhuman primates' species, however, until now, few studies have addressed the molecular characterization of these isolates and discussed their role in public health[3,5,15,20,22,32].

Thus, we focused our attention here to verify the occurrence of the Giardia infection in nonhuman primates' fauna of the Bauru Municipal Zoological Garden (São Paulo State, Brazil).

MATERIALS AND METHODS

Study site and sampling

The Bauru Municipal Zoological Garden, located in the city of Bauru in São Paulo State, covers an area of 43.5ha and houses 880 animals of 210 species, comprising mammals, birds and reptiles. Before the start of the study, approval was obtained from the Research Ethics Committee at Sacred Heart University (105/10 CEP-USC) as well as by the zoo's administration. From September to November 2009, fecal samples from forty-seven primates of 18 species, belonging to three families of New World monkeys (Atelidae, Callitrichidae and Pitheciidae) and one of Old World (Cercopithecidae) were analyzed. The animals were either kept individually (five animals) or in monospecific groups of 2-7 individuals (42 animals) sharing the same enclosures (Table 1).

Table 1

Intestinal parasites identified in nonhuman primates at Bauru zoo and G. duodenalis PCR analysis

Primate specie	Common name	No. of animals (no. enclosures)	Coprological analysis	Giardia PCR
Alouatta caraya	Black howler monkey	4 (1)	Giardia, Entamoeba spp	+
Alouatta fusca	Brown howler monkey	3 (1)	Giardia	+
Alouatta seniculus	Red howler monkey	4 (2)	Giardia, Entamoeba spp, Endolimax nana	+
Ateles belzebuth	White-fronted spider monkey	2 (1)	Giardia, Iodamoeba bütschilii	+
Ateles fusciceps	Brown-headed spider monkey	2 (1)	-	-
Ateles paniscus	Red-faced spider monkey	5 (1)	Oxyurid eggs	-
Callicebus nigrifrons	Black-fronted titi monkey	2 (1)	-	-
Callithrix argentata	Silvery marmoset	2 (1)	-	-
Callithrix geoffroyi	Geoffroy's tufted-ear marmoset	1	-	-
Lagothrix lagotricha	Woolly monkey	5 (3)	-	-
Leontopithecus chrysomelas	Golden-headed lion tamarin	7 (2)	-	-
Leontopithecus rosalia	Golden lion tamarin	2 (1)	Strongyle-type eggs	-
Mandrillus sphinx	Mandrill	2 (1)	Entamoeba spp	-
Papio hamadryas	Hamadryas baboon	1	-	-
Papio papio	Guinea baboon	2 (1)	-	-
Saguinus bicolor	Brazilian bare-faced tamarim	1	-	-
Saguinus fuscicollis	Brown-headed tamarin	1	-	-
Saguinus melanoleucus	White saddleback tamarin	1	Entamoeba spp	-

- negative samples

Due to the group housing, all the feces discharged by the animals in each pen were recovered from the floor and pooled in labeled plastic vials containing 2.5% potassium dichromate. For those animals kept alone in the enclosure, individual sampling was possible. Collection of stool was carried out in the morning, after the feeding of animals, on three alternate days.

The samples were assessed for helminthes and protozoans by conventional coprological techniques of spontaneous sedimentation and centrifuge-flotation with zinc sulfate, according to SLOSS et al. [27]. For Cryptosporidium diagnosis, a part of each stool sample was previously concentrated by centrifugation (500 xg, 10 min) and the sediment obtained was used to prepare smears stained by a modified Ziehl-Neelsen technique[13]. An aliquot of this sediment was also stored at -20 °C for subsequent DNA extraction.

DNA extraction

The genomic DNA was extracted using the QIAamp® Stool Mini Kit (Qiagen, Germany) following the manufacturer's instructions. Prior to extraction, the samples were subjected to three freeze/thaw cycles as follows: two cycles in liquid nitrogen for five min and at 70 °C for five min and concluding in liquid nitrogen for five min and thawed at 95 °C for five min. For molecular characterization of Giardia, DNA was extracted from all the samples regardless the coprological diagnosis.

PCR assays and sequencing

Molecular analysis was carried out by amplifying gdh and tpi fragment genes. A nested PCR procedure was used to amplify a 530 bp region of the tpi fragment using the primers AL3543 and AL3546 for the first step, and AL3544 and AL3545 for the second one[29]. For gdh, a semi-nested PCR was used to amplify a 432 bp fragment with the primers GDHeF and GDHiR in the primary reaction and GDHiF and GDHiR in the secondary[25] one. In all PCR reactions, positive (DNA from axenic Giardia culture) and negative controls were included. The amplified products were submitted to 1.5% agarose gel electrophoresis, stained with ethidium bromide.

Bands of the predicted sizes were excised from agarose gel, then, DNA was purified using the Ultrafree® DA kit (Millipore Corp., USA) and sequenced in both directions in Macrogen Inc. (Seoul, Korea). The nucleotide sequences obtained were manually corrected and multiple alignments for each locus were performed using CLUSTAL X version 1.8[30] and queried against known sequences of GenBank database using BLAST. The phenetic analyses were performed with the MEGA software (http://www.mega-software.net) and the sequences obtained in the present study were deposited in the GenBank data base under accession numbers JN172997-JN172999 for gdh sequences, and JN410841-JN410843 for tpi sequences.

The DNA sequences were translated into amino acids using the online software Emboss gui - European Molecular Biology Open Software Suite (http://bips.u-strasbg.fr/EMBOSS/).

RESULTS AND DISCUSSION

The microscopic examination of stool samples revealed that enteric parasites pathogenic and/or non pathogenic were detected in samples from 36% (8/22) of the enclosures. Among the 18 species of primates, intestinal parasites were found in feces from animals belonging to the genera Ateles, Alouatta, Leontopithecus, Mandrillus and Saguinus (Table 1). Giardia cysts were found in the feces collected in four enclosures (18%) where animals of the following species were kept: Ateles belzebuth, Alouatta seniculus, Alouatta caraya and Alouatta fusca (Table 1). Other intestinal protozoa were identified and their frequencies were as following: Entamoeba spp (18%), Iodamoeba butschlii (4.5%) and Endolimax nana (4.5%). No Cryptosporidium oocysts were found on fecal smears of all assessed samples. Parasitism by helminthes was detected in animals from two enclosures, whose fecal samples were positive for strongylid and oxyurid nematodes eggs (Table 1).

The occurrence of both protozoa and helminthes have been reported in captive nonhuman primates of different regions[5,7,9,16,19,24], with the predominance of the former[7,19,20]. According to some authors, the most frequent occurrence of protozoa infections can be explained by factors such as the short prepatent period, being immediately infective when excreted and the low infective dose required[19] for infection. Besides, the restricted and collective environments conditions where the animals are kept facilitate protozoan transmission, because these organisms can be acquired through direct contact, which may be favored by primates' highly social nature. In the present study, besides the occurrence of the protozoa Giardia and species of amoebae, the detection of oxyurid eggs reinforces these aspects, since unlike most other nematodes, this helminth is already able to infect the host when passed in feces.

Likewise as previously reported in the literature[5,19], herein, Giardia and Entamoeba spp. were the most frequent parasites diagnosed in NHP. Nevertheless, the clinical relevance of these infections is difficult to ascertain, especially regarding to Entamoeba, because several species are considered non-pathogenic[5]. So, the known species considered pathogenic for NHP include E. histolytica and its virulent variant E. nuttalli [5]. However, through direct microscopic identification of the parasite, it is impossible to distinguish between E. histolytica and the non-pathogenic species E. dispar and E. moshkovskii, once these amoebae are morphologically indistinguishable. It is important to emphasize that, although Giardia and Entamoeba spp. are commonly found in stools of wild and captive NHP and can be a significant cause of diarrhea, in the present survey, all fecal samples had a normal consistency and animals showed no symptoms of infection, even when both protozoan were detected in animals kept in the same pen (Table 1).

Given that Cryptosporidium is an important pathogenic protozoa that has been detected in a range of captive mammals' species [7,11,15,19,21,24], it was expected that this parasite could be detected in the animals of the present study. It is noteworthy that the occurrence of Cryptosporidium infection is still underestimated, since its diagnosis depends on the identification of oocysts in fecal smears stained, being a more laborious method that is generally not included in many surveys. This was not the point in our study since smears were prepared and examined, but a likely reason for negative results of Cryptosporidium could be the analysis of pooled stool samples, despite efforts to concentrate the material.

Although the increasing interest to gain insight into the role of nonhuman primates as hosts of Giardia, molecular characterization of isolates from New World species is still limited, especially in Brazil, which is the world's richest country in primate species diversity.

Herein, the four Giardia-positive samples produced PCR products of the expected size in the gdh and tpi genes, but no amplicon was obtained from the eighteen Giardia cyst-negative samples assessed. Clear sequences of amplicons were only obtained for the isolates from Ateles belzebuth (BA1), Alouatta fusca (BA2) and Alouatta caraya (BA3) (Tables 1 and 2). A comparison of these sequences with the homologous gdh and tpi deposited in GenBank showed that all three isolates were assigned to assemblage A (Table 2; Fig. 1). Taking as reference the comparative evaluation of G. duodenalis sequence data carried out by WIELINGA & THOMPSON, the alignment between these three sequences and other ones from humans and nonhuman primates revealed that in tpi locus, the isolates BA1 (JN410841), BA2 (JN410842), and BA3 (JN410843) were subtype AII, whereas in the gdh locus, only the isolate BA3 (JN172999) was classified as AII and the isolates BA1 (JN172997) and BA2 (JN172998) corresponded to sub-assemblage AI (Fig. 1; Table 2).

Table 2

Giardia duodenalis assemblage A: isolates, positions and breakdowns of intra-sub-assemblage substitutions in the gdh and tpi genes*

Assemblage	Isolate	GenBank™	Host/Country	Nucleotide position
	gdh			237	246	249	258	270	309	372	390	396	510	522	525	531	534	570	603	621	633	636	666	672	< b>687	693	699	720	723	753	756	771	786	807	831	861	867	870	894	902	1080	1266
AI	Portl 1	M84604	Human/Australia	C	C	T	C	T	C	C	T	A	C	G	C	C	A	C	T	C	C	T	C	C	T	C	T	T	C	C	C	C	C	C	C	T	T	T	C	C	T	G
	SQ678,681	FJ890951	B.-c.squirrel monkey/The Netherlands	-	-	-	-	-	-	-	-	-	T	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	T	.	.	.	.	.	.	.	.	.	-	-
	NLH45	AY826195	Human/The Netherlands	T	.	.	.	.	.	.	.	.	.	.	.	.	.	.	C	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
AII	BA1	JN172997	monkey/Brazil	-	-	-	-	-	-	-	-	-	.	.	.	.	.	.	T	.	.	.	T	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
	BA2	JN172998	monkey/Brazil	-	-	-	-	-	-	-	-	-	.	.	.	.	.	.	T	.	.	.	C	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
	BA3	JN172999	monkey/Brazil	-	-	-	-	-	-	-	-	-	.	.	.	.	.	.	C	T	.	.	T	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
AIII		NLR118	Roe deer/The Netherlands	.	.	C	G	C	T	A	C	G	.	C	T	T	G	T	C	.	T	C	.	T	C	T	C	C	T	.	.	A	T	.	.	.	C	C	-	-	-	-

The isolates of the present study are highlighted in bold as well as the substitution patterns proposed by WIELINGA & THOMPSON (2007). Novel substitution positions are highlighted in bold italics. Nucleotide sequences identical to reference isolates are indicated by (.) And nucleotide sequences not available for comparison are indicated by (-).

Fig. 1

Dendrograms of Giardia duodenalis based on nucleotide sequences of gdh and tpi genes. Trees were constructed using the neighbor-joining method implemented by the computer program Mega 4.1. Bootstrap values were calculated by the analysis of 1000 simulations (only bootstrap values >50% are shown). The reference sequences for both genes are underlined.

Interestingly, in regards to the analysis of polymorphisms in the nucleotide sequences (SNPs) obtained for the gdh gene, the BA2 (JN172998) isolate did show 100% homology with reference strains for sub-assemblage AI while the isolates BA1 (JN172997) and BA2 (JN172998) revealed at least one SNP (Table 2). Thus, the isolate BA1 showed one SNP at position 666, and the BA3 showed the same SNP at position 666 and a novel one at position 621(Table 2). Among the tpi sequences, the isolates BA2 (JN410842) and BA3 (JN410843) did show 100% homology with reference strains for sub-assemblage AII whereas the isolate BA1 (JN410841) reveled a novel SNP at position 107 (Table 2). In all sequences retrieved, nucleotide substitutions were characterized as transitions and the SPNs were synonymous, resulting in no change in the amino acids encoded. So, only in the tpi sequence of isolate BA1 (JN410841), a non-synonymous SNP in position 634 determined a change in the deduced amino acid sequence leading to a replacement of a serine by a phenylalanine.

In the present study, the finding of infection within assemblage A, confirms previous reports that have detected the same genotype in either the gdh or the tpi locus[20,22] as well as in β-giardin[20,32] and SSUrRNA[12] genes. Additionally, in these studies, assemblage A subtypes were identified as either AI or AII.

Recently, both assemblages A and B were found in New World monkeys[14,20,28,32], including species that live in South American forests[28,32]. In Brazil, the occurrence of genotype A in NHP was first reported by VOLOTÃO and colleagues, who detected only isolates belonging to the AI assemblage in samples from southern brown howler monkeys (Alouatta clamitans)[32]. In contrast, SOARES et al. recently identified assemblage B in all the samples from 20 howler monkeys Alouatta fusca housed in a wild animal center[27].

With respect to the separation of isolates into sub-assemblages, herein, the inconsistency observed among the loci assessed is not an unusual situation. According to authors BONHOMME et al., inconsistent results are sometimes generated when different loci are targeted, especially regarding the subtypes. Thus, in spite of the possibility of assemblage swapping, explaining different assemblages or sub-assemblages in different loci in the same isolate has not been an easy task.

Despite the finding of two different sub-assemblages in the same isolate, as we can see, the infection of captive nonhuman primates within assemblage A has an epidemiological relevance, since this genetic group is zoonotic and the most common non-host specific assemblage in animals[10]. Concerning subtypes of this assemblage, AI and AII are the two most frequently involved in the infections, even though they appear to differ in host preference, since there is evidence that humans are more commonly infected with AII whereas animals are more commonly infected with AI[10]. However, both subgroups have been detected in humans and animals.

Finally, despite the low number of tested isolates, the results presented herein outline some pertinent aspects about the occurrence of G. duodenalis assemblages in captive nonhuman primates' species and also provide insights for future elucidation of Giardia epidemiology in endemic areas. All things considered, this kind of investigation highlights the importance of coproparasitological surveys as part of practice in zoos in order to implement preventive measures and safeguard the health of captive animals and people near them (animal keepers and visitors). In this context, it is appropriate to take into account that the transmission of pathogens between humans and nonhuman species has to be considered in either direction, so that recognizing the links between human, animal and ecosystem health can provide an effective approach to understanding the transmission of pathogens among these hosts[8].