INTRODUCTION
Cryptosporidium species are protozoan parasites that infect humans and a wide variety of animals. Human cryptosporidiosis has been recognized as the most common cause of protozoal diarrhoea worldwide leading to significant morbidity and mortality both in developing and industrialized nations (Marshall et al. Reference Marshall, Naumovitz, Ortega and Sterling1997; Clark, Reference Clark1999). Furthermore, Cryptosporidium has caused multiple waterborne outbreaks in developed and developing countries (Insulander et al. Reference Insulander, Lebbad, Stenstrom and Svenungsson2005) with the largest Cryptosporidium waterborne outbreak reported in Milwaukee, USA, affecting 403 000 persons with more than 100 fatal cases (MacKenzie et al. Reference MacKenzie, Hoxie, Proctor, Gradus, Blair, Peterson, Kazmierczak, Addiss, Fox, Rose and David1994).
Cryptosporidium has recently been classified into 23 valid species (Fayer, Reference Fayer2010; Fayer et al. Reference Fayer, Santín and Macarisin2010; Traversa, Reference Traversa2010) and nearly 61 genotypes have been reported (Plutzer and Karanis, Reference Plutzer and Karanis2009). Of these, 1 species was reported from amphibians, 2 from reptiles, 3 from birds and 12 were reported from mammals (Fayer, Reference Fayer2010). Of the Cryptosporidium species infecting mammals, C. parvum and C. hominis are the most common species responsible for human infections. Although C. felis (Matos et al. Reference Matos, Alves, Xiao, Cama and Antunes2004), C. muris (Katsumata et al. Reference Katsumata, Hosea, Ranuh, Uga, Yanagi and Kohno2000), and C. canis (Cama et al. Reference Cama, Bern, Sulaiman, Gilman, Ticona, Vivar, Kawai, Vargas, Zhou and Xiao2003) have also been reported to infect humans, they are less common.
Avian cryptosporidiosis has been reported from more than 30 avian species and is usually caused by 3 Cryptosporidium species and 10 Cryptosporidium genotypes (Fayer, Reference Fayer2010). The recognized avian species are C. meleagridis (Slavin, Reference Slavin1955), C. baileyi (Current et al. Reference Current, Upton and Haynes1986) and C. galli (Pavlásek, Reference Pavlásek1999). Avian genotypes include avian genotypes I – IV that were identified from various avian hosts (Ryan, Reference Ryan2010) such as Eurasian woodcock genotype (Ryan et al. Reference Ryan, Xiao, Read, Zhou, Lal and Pavlasek2003), duck genotype from black duck (Morgan et al. Reference Morgan, Monis, Xiao, Limor, Sulaiman, Raidal, O'donoghue, Gasser, Murray, Fayer, Blagburn, Lal and Thompson2001) and goose genotypes 1 – IV from Canada geese (Jellison et al. Reference Jellison, Distel, Hemond and Schauer2004).
The risk of zoonotic transmission arises from C. meleagridis which is an emerging human pathogen and has been considered the third most common Cryptosporidium species infecting humans (Xiao and Fayer, Reference Xiao and Fayer2008). The role of C. parvum infection among birds is not fully understood and even the reported cases from wild birds were attributed to mechanical transmission rather than an established infection (Dieter et al. Reference Dieter, Dieter, Dieter and Gulliver2001; Zhou et al. Reference Zhou, Kassa, Tischler and Xiao2004; Majewska et al. Reference Majewska, Graczyk, Slodkowicz-Kowalska, Tamang, Jedrzejewski, Zduniak, Solarczyk, Nowosad and Nowosad2009). In Malaysia, no information on Cryptosporidium species and genotypes infecting avian hosts is currently available. Thus, the present study was carried out to determine the species and genotypes of Cryptosporidium found in birds at the National Zoo, Kuala Lumpur and whether these could be potential reservoirs for human infections.
MATERIALS AND METHODS
Source of samples
Ninety faecal samples were collected from 37 species of birds from different locations (i.e. the aviary, bird house, breeding area and the lake) in the National Zoo for a period of 3 months (i.e. April to June 2008). Samples were collected early in the morning in clean plastic containers and transported to the laboratory. The faecal samples were fixed with 2·5% potassium dichromate before being stored in a cold room at 4°C for further analysis.
Microscopic examination
The faecal samples were concentrated to obtain a better yield of Cryptosporidium oocysts with the formalin-ether concentration technique. Using a pipette, a drop of the sediment was placed on a glass slide and evenly spread and stained with Ziehl-Neelsen stain. Briefly the slide was dried and stained with carbol fuchsin for 10 min, destained using acid alcohol for 10–15 s and stained with malachite green for 3 min. Stained smears were examined under 1000×magnification using a light microscope.
Polymerase chain reaction
DNA was extracted from approximately 1 g of faecal samples using QIAmp DNA stool mini kit according to the manufacturer's instructions (QIAgen) and stored at −20°C until use. Nested-PCR was used to amplify a partial, polymorphic region of 18S rRNA, according to the method described by Nichols et al. (Reference Nichols, Campbell and Smith2003). Primary PCR amplified a 655 to 667 bp fragment, depending on the species of Cryptosporidium or C. parvum genotype, using the forward primer N-DIAGF2 (5′-CAA TTG GAG GGC AAG TCT GGT GCC AGC-3′) and the reverse primer N-DIAGR2 (5′-CCT TCC TAT GTC TGG ACC TGG TGA GT-3′), while the secondary PCR amplified a 435 bp fragment using the forward primer CPB-DIAGF (5′-AAG CTC GTA GTT GGA TTT CTG-3′) and the reverse primer CPB-DIAGR (5′-TAA GGT GCT GAA GGA GTA AGG-3′) that were previously developed by Johnson et al. (Reference Johnson, Pieniazek, Griffin, Misener and Rose1995).
Both primary and secondary PCR were conducted in a 50 μl reaction mixture containing 200 μ m of each dNTP (Fermentas, Ontario, Canada), 0·2 μ m of each of the primers CPB-DIAGF2/R2 (Research Biolab, Singapore), 400 μg/ml BSA (New England Biolabs, Ipswich, MA, USA), 3·5 mm MgCl2 (Fermentas, Ontario, Canada), 2·5 U Taq polymerase (New England Biolabs, Ipswich, MA, USA), and 1X ThermoPol PCR buffer (New England Biolabs, Ipswich, MA, USA). Two μl of DNA template was used in the primary PCR whereas 5 μl of the first PCR product was used as template in the secondary PCR. DNA template extracted from oocysts of C. parvum H3 isolate, was used as positive control. The cycling condition was as follows; hot start at 95°C for 5 min, followed by 35 cycles of denaturing for 30 s at 94°C, annealing for 1 min at 68°C and extension for 30 s at 72°C, followed by a final extension at 72°C for 10 min. The secondary PCR had a similar cycling condition except that the annealing temperature was 60°C instead of 68°C. The PCR products were electrophoresed on 2% agarose gels at 100 V.
DNA sequencing and phylogenetic analysis
The DNA was purified using the QIAquick Gel Extraction Kit (QIAgen, Germany) according to manufacturer's instructions. Cycle sequencing was carried out using the ABI Prism® BigDye™ terminator v3.0 Ready Reaction Cycle Sequencing Kit (Applied Biosystems, USA) according to the manufacturer's instructions, and amplified products were analysed using the ABI Prism® 3700 DNA Analyzer (Applied Biosystems, USA). DNA sequencing was carried out in both directions. Forward and reverse sequences were edited and assembled, and the consensus sequence was derived for each sample using BioEdit (www.mbio.ncsu.edu). Multiple sequence alignment was performed using MEGA4 (www.megasoftware.net). To root the trees, Eimeria tenella (GenBank Accession no. AF026388) sequence was retrieved as it appears to be the closest sister taxon to the ingroup of the present study. Neighbour-joining (NJ) and maximum parsimony (MP) analyses were performed in MEGA4. NJ analyses were performed with distances calculated with the Kimura 2-parameter (Kimura, Reference Kimura1980). Unweighted parsimony analyses were performed using the Close-Neighbour-Interchange algorithm (Nei and Kumar, Reference Nei and Kumar2000). To evaluate the support for inferred topologies, bootstrapping (Felsenstein, Reference Felsenstein1985) was done with 1000 replicates. Homology search was carried out using the Basic Local Alignment Search Tool (BLAST) (Altschul et al. Reference Altschul, Madden, Schaffer, Zhang, Zhang, Miller and Lipman1997).
RESULTS
A total of 90 bird faecal samples were collected and 9 (10%) out of these were positive for Cryptosporidium spp. microscopically. Cryptosporidium oocysts were identified as pink to red spheres with a size of approximately 4–6 μm. Nine samples from 8 avian species that were found to be positive with Cryptosporidium are shown in Table 1. However, the numbers of oocysts in each positive faecal sample were not counted.
List of bird species positive for Cryptosporidium oocysts using Ziehl-Neelson stain at the National Zoo
* Cryptosporidium species was not identified due to failure in DNA sequencing.
# Two samples were microscopically positive but only one was confirmed using PCR.
Microscopic results were in concordance with nested PCR results except in 1 case where, microscopically, the sample was positive but negative by nested PCR. Amplicons from all PCR positive samples were successfully sequenced except 1 sample. Seven sequences from this study (Table 1) and 13 reference sequences from GenBank, representing avian Cryptosporidium species; C. meleagridis, C. baileyi, C. galli and avian genotypes I – IV were multiple aligned and used for constructing the phylogenetic trees. The Neighbour-Joining method (Fig. 1) grouped all the isolates from this study together with C. parvum in 1 cluster (83% bootstrap support). Maximum Parsimony was concordant in topology with NJ tree placing the 7 isolates in 1 clade with C. parvum (72% bootstrap) (Fig. 1). Similarity search using BLAST showed that isolates 20 and 31 were identical while isolates 13, 15, 17, 37 and 71 were 99% similar to C. parvum (GenBank Accession number FJ984565) which was isolated from avian hosts in Hungary (Plutzer and Tomor, Reference Plutzer and Tomor2009).
Phylogenetic tree for Cryptosporidium spp.
DISCUSSION
The current study carried out at the National Zoo, Kuala Lumpur showed evidence that birds on display in the zoo have the potential to be sources of dissemination of Cryptosporidium in the environment. Two previous studies have been carried out to determine the prevalence of cryptosporidiosis in the birds in the same zoo. The two studies indicated that the prevalence of Cryptosporidium was rated at 6% (Rohela et al. Reference Rohela, Lim, Jamaiah, Khadijah, Laang, Nazri and Nurulhuda2005) and 3·4% (Mahat, Reference Mahat2006) compared to 10% in this study. However, the two previous studies did not conduct molecular characterization of the Cryptosporidium species recovered from infected birds.
In this present study, phylogenetic analysis using Maximum Parsimony and Neighbour-Joining methods identified all the 7 isolates as C. parvum. This finding was confirmed by BLAST search in which the 7 isolates were almost identical to C. parvum which was isolated from the faeces of wild aquatic birds (mallard, Anas platyrhynchos and coot, Fulica atra), in Hungary (Plutzer and Tomor, Reference Plutzer and Tomor2009). Previous studies have detected C. parvum in the faeces of Canada geese in the United States (Zhou et al. Reference Zhou, Kassa, Tischler and Xiao2004; Graczyk et al. Reference Graczyk, Fayer, Trout, Lewis, Farley, Sulaiman and Lal1998; Kassa et al. Reference Kassa, Harrington and Bisesi2004). In Western Poland, of 499 faecal specimens collected from birds, C. parvum was identified in the faeces of 1 mandarin duck (Aix galericulata), 2 common mergansers (Mergus merganser), 4 mute swans (Cygnus olor), 3 white storks (Ciconia ciconia), 6 carrion crow (Corvus corone) and 3 rooks (Corvus frugilegus) (Majewska et al. Reference Majewska, Graczyk, Slodkowicz-Kowalska, Tamang, Jedrzejewski, Zduniak, Solarczyk, Nowosad and Nowosad2009).
Since studies that attempted to infect birds with C. parvum from mammals had inconsistent results with mild to moderate and unsuccessful establishment of the infection, detection of C. parvum in the faeces of birds has been thought to be the result of mechanical transmission rather than established infection (Lindsay et al. Reference Lindsay, Blagburn and Ernest1987; Palkovic and Marousek, Reference Palkovic and Marousek1989; Sreter and Varga, Reference Sreter and Varga2000; Darabus and Olariu, Reference Darabus and Olariu2003). However, it was reported that C. parvum oocysts retain viability and infectivity following passage through the avian hosts (Graczyk et al. Reference Graczyk, Cranfield, Fayer, Trout and Goodale1997). In addition, previous studies showed that the concentration of C. parvum oocysts was 3·7×106/g in the faeces of Canada geese (Graczyk et al. Reference Graczyk, Fayer, Trout, Lewis, Farley, Sulaiman and Lal1998) and 4·8×102/g in the faeces of ducks (Kuhn et al. Reference Kuhn, Rock and Oshima2002).
A prediction model was developed to correlate between bird abundance, faeces input and faecal indicator bacteria (FIB) in aquatic environments (Kirschner et al. Reference Kirschner, Zechmeister, Kavka, Beiwl, Herzig, Mach and Farnleitner2004). Application of this model showed that approximately 9·3×106 infectious C. parvum oocysts could be introduced into the water due to a single visitation of an average-sized waterfowl flock (Graczyk et al. Reference Graczyk, Majewska and Schwab2008). Thus, birds may still play a significant role in environmental contamination with the human infectious C. parvum. It must be noted that C. parvum oocysts are highly infectious, with an ID50 ranging from 9 to 2980 oocysts based on the type of isolates (Okhuysen et al. Reference Okhuysen, Chappell, Crabb, Sterling and Dupont1999, Reference Okhuysen, Rich, Chappell, Grimes, Widmer, Feng and Tzipori2002; Messner et al. Reference Messner, Chappell and Okhuysen2001). In the study area of the present investigation it is important to note that the birds at the lake area are free to fly out of the National Zoo to the Putrajaya Lake which is about 35 km away. Therefore there is the possibility that if these birds are infected they may contaminate another water body. However, this postulation needs further study to confirm its significance since the present investigation was limited by the fact that oocysts were not counted.
In conclusion, the results of the present study showed a high prevalence of C. parvum among Zoo birds. Although C. parvum has not been determined to cause infection in birds, and the role of birds was postulated to be mainly as mechanical transporters, the present findings highlight the significant public health risk posed by birds that harbour the zoonotic species of Cryptosporidium. In addition, even though this study found no avian species or genotypes in zoo birds, it cannot be concluded that birds in Malaysia are free of avian Cryptosporidium species or genotypes. Therefore, further studies are warranted in order to have a better understanding of the epidemiology of cryptosporidiosis of birds in Malaysia.
ACKNOWLEDGEMENTS
The authors are grateful to the Director of Zoo Negara, Malaysia, for giving permission to collect bird faecal samples from the zoo. Special thanks are due to the zoo staff for providing guidance and technical assistance during sample collection. This project was funded by an International Medical University Research Grant, no. BMS I-01/2008(10).
