Cryptosporidium spp. in Freshwater Bivalves in Portugal PEDRO C. MELO, a JOAQUIM TEODOSIO, b JOAQUIM REIS, c ANA DUARTE, a JOSE ´ C. COSTA d and ISABEL P. FONSECA a a CIISA, Faculdade de Medicina Veterina´ria, Lisboa, and b CCMAR, Universidade do Algarve, Faculdade de Cieˆncias do Mar e Ambiente, Campus de Gambelas, Faro, and c Centro de Biologia Ambiental/Dep. de Biologia Animal, Faculdade de Cieˆncias da UL, Campo Grande, Lisboa, and d Centro de Imunologia e Biologia Parasita´ria, INSA, Porto, Portugal C RYPTOSPORIDIUM spp. in water has been implicated in several human cryptosporidiosis outbreaks (Fayer 2004). Studies on Cryptosporidium oocysts in surface waters are made difficult by the low concentrations of organisms present due to dilution. Since bivalves are filter-feeders oocysts in suspended in water can be removed by this filtration and the mollusc digestive enzymes reduce oocyst viability (Freire-Santos et al. 2002). If oocysts survive the bivalve digestive tract and are excreted, the dense fecal material containing viable oocysts are expected to sediment to the bottom of the riverbed. However, bivalve filtration mechanisms can also concentrate Cryptosporidium oocysts pres- ent in the water. Hence it is possible to recover higher numbers of oocysts from the bivalves compared with that recoverable from surface waters (Graczyk et al. 2004). Human consumption of freshwater clams and mussels is com- mon in northeast Portugal and in the Tejo and Guadiana river basins. These bivalves might be a source of human Cryptospo- ridium infection therefore, it is important to assess the public health risks of eating contaminated bivalves. As bivalves can be useful bioindicators of superficial water contamination (Graczyk et al. 2004), we examined clams in Portugal for the presence of Cryptosporidum oocysts examined evaluated considering that clams can be. MATERIALS AND METHODS Three freshwater bivalves, Anodonta anatina, Unio pictorum, and Corbicula fluminea, were analyzed. Sixteen sampling areas in the Guadiana river hydrographic basin in southeast Portugal were selected and the clams were randomly collected by searching and by kick sampling. A total of 55 samples were obtained between May 2003 and June 2005: 69% (24/55) were of U. pictorium, 50% (19/55) were C. fluminea, and 31% (12/55) were A. anatine. Vari- ation in the water levels, turbidity, and unfavourable conditions precluded, being able to collect all three species at each sampling area each time. Gills and the gastrointestinal tracts of each species were dis- sected from three to 15 bivalves (depending on their size) and the tissues were pooled to provide uniform sizes of preparations from different bivalves. The tissues were homogenized and oocysts were concentrated using a modification of the methods described by Ritchie (Pereira da Fonseca 2000) and purified by discontinu- ous sucrose density centrifugation. The material collected at the interface of the two layers with densities of 1,103 and 1,064 was washed and concentrated into a pellet by centrifugation and then resuspended in distilled water to a final volume of 1.5 ml. Detection of oocysts was performed by direct immunofluores- cent assay (IFA) using the monoclonal antibody (Mab) kit Crypto/ Giardia Cel (Cellabs, Brookvale, Australia). The isolated oocyst preparation was incubated with the Mab and observed by fluor- escence microscopy at 400 Â magnification. The size and shape of Cryptosporidium oocysts enabled us to distinguish these from Giardia cysts. Genotyping using 18S suRNA gene probes was performed on 47 samples by nested PCR analysis (Xiao et al. 1999). Genom- ic DNA was extracted by the proteinase K method and the Qiagen DNA extraction kit. Qiagen, Hilden, Germany. The primers used were 5 0 -TTCTAGAGCTAATACATGCG-3 0 and 5 0 -CCCTAAT CCTTCGAAACAGGA-3 0 in the primary PCR, and 5 0 -GGAAGG GTTGTATTTATTAGATAAAG-3 0 and 5 0 -AAGGAGTAAGGA ACAACCTCCA-3 0 in the secondary reaction. PCR was performed using 94 1C for 5min followed by 40 cycles of 94 1C for 50 min, 50 1C for 30 min, 72 1C for 45min, and then 72 1C for 10 min. RESULTS AND DISCUSSION Positive samples as determined by IFA identified 1–8 Crypto- sporidium oocysts/25 ml of the resuspended pellet. Oocysts were found in 18 (33.3%) of the 55 samples analyzed in this study; 29% (7/24) of the U. pictorum samples, 33% (4/12) of the A. anatine samples, and 37% (7/19) of the C. fluminea samples. Cryptospo- ridium oocysts were found in bivalves collected at six of the 16 sampling areas. There was no agreement in oocyst detection when more than one species was collected from the same site at the same time. This might be due to the physiology of different spe- cies. For example, A. anatina has the highest rate of filtration capacity and C. fluminea has the highest tolerance to pollution. Also, there might be differences in response to stress levels, dis- tribution at different sites, and total numbers and sizes of the pooled tissues analyzed in this study. In these initial attempts we were not successful in amplifying C. parvum DNA in all 47 specimens analyzed. This might be due to the small number of Cryptosporidium spp. oocysts in the sam- ples and potential inhibition of the PCR protocol by chemicals in the soil and water in these collection sites (Johnson et al. 1995). The possibility that the oocysts identified by IFA were not those of C. parvum but a different Cryptosporidium species cannot be ruled out. The primers used were specific for the genus Crypto- sporidium. Analysis using RFLP after PCR for identification at the species level is in progress. As we plan on future studies in this project, we must consider methods to increase the recovery of oocysts for these kinds of analyses. These include analyzing more samples, and each should be larger. Experiments in which bivalves (shown to be Crypto- sporidium free) are placed within cages at the sampling sites and then recovered and analyzed after a specified time can be conducted. For the IFA analyses, a larger aliquot of the isolated oocyst pellet should be examined, and immunomagnetic separation tech- niques tested for obtaining higher concentrations of purified oo- cysts. These methods and controls should be solidly established before expanding the scope of this study to include other fresh- water bivalve species. As we detected Cryptosporidium oocysts in clam tissues by IFA in these preliminary studies, this is consistent with the suggestion that the freshwater clams A. anatina, U. pictorum, and C. fluminea are useful indicators of surface water contamination. These clams Corresponding Author: I. Fonseca, CIISA, Parasitologia, Faculdade de Medicina Veterina ´ria, Lisboa, Portugal—Telephone number: 1351- 213-602038; FAX number: 1351-213-652895; e-mail: ifonseca@fmv. utl.pt S28 J. Eukaryot. Microbiol., 53(S1), 2006 pp. S28–S29 r 2006 The Author(s) Journal compilation r 2006 by the International Society of Protistologists DOI: 10.1111/j.1550-7408.2006.00164.x