Research & Cryptosporidium: future challenges

Challenge: Do all species within the Crytosporidium genus pose a threat to food safety?

Cryptosporidium parvum, an oocyst-forming apicomplexan protozoan, is an obligate intracellular parasite that infects the microvillus border of the epithelium in the gastrointestinal tract of humans and various animal hosts (Clark, 1999). To date, the genus, Cryptosporidium, consists of at least 10 recognised species (Fayer et al., 2000). Human infection, however, is predominately caused by C. parvum (Kosek et al., 2001) and human illness caused by Cryptosporidium has now been reported in more than forty countries in six continents (Kosek et al., 2001). Not all Cryptosporidium parasites have the same potential as gastrointestinal pathogens. Five Cryptosporidium parasites, including the C. parvum human and bovine genotypes, C. meleagridis, C. felis, and C. canis, are the most common causes of human cryptosporidiosis (Xiao et al., 2001). Others such as C. muris, C. andersoni, a cervine genotype and a pig genotype, have been found in a few human cases. Thus far, only the human and bovine genotypes of C. parvum have been identified as the cause of foodborne and waterborne outbreaks, indicating that they are probably more infectious to humans than other Cryptosporidium parasites. Within human and bovine genotypes of C. parvum, there is different virulence potential for causing human disease. Among nearly 50 subgenotypes of the C. parvum human genotype identified so far, only several subgenotypes have wide geographic distributions, and one such subgenotype has been found to be responsible for seven foodborne and waterborne outbreaks in North America and Europe, indicating that certain subgenotypes of the C. parvum human genotype are more infectious than other subgenotypes. Likewise, among the 30 subgenotypes of the C. parvum bovine genotype identified so far, only one or two have wide geographic distributions and one of these subgenotypes was responsible for two waterborne outbreaks in the U.S. (Xiao et al., unpublished observations). The wide geographic distribution of these Cryptosporidium parasites is probably indicative of their biological fitness.

Challenge: What role can molecular biological techniques play in aiding the epidemiology of the disease?

Recently, molecular tools have been developed to detect and differentiate Cryptosporidium parasites at the species, genotype and subgenotype levels. These tools now make it possible to determine the identity of Cryptosporidium parasites infecting humans, track the source of contamination in waterborne, foodborne and daycare outbreaks, compare the pathogenicity, infection patterns and disease spectrum among Cryptosporidium species/genotypes, characterize the transmission dynamics of Cryptosporidium infection in endemic areas, and assess the public health importance and contamination sources of Cryptosporidium oocysts in water. Using these molecular tools, several workers have characterized Cryptosporidium parasites from different human populations in several geographic areas. Thus far, only the human and bovine genotypes of C. parvum have been identified in cryptosporidiosis outbreaks in North America and Europe. In contrast, these two Cryptosporidium parasites as well as C. meleagridis, C. felis and the Cryptosporidium dog genotype have been identified in sporadic cases of cryptosporidiosis in both immunocompetent and immunocompromised persons living in the U.S., UK, Portugal, France, Japan, Switzerland, Peru and Kenya. Several cases of Cryptosporidium pig and cervine genotypes and C. muris/C. andersoni infection in humans have also been identified, suggesting that many Cryptosporidium species and genotypes have the potential to infect humans, and that zoonotic infections can play a significant role under certain circumstances. Geographic differences have been observed in the proportion of infections due to zoonotic or anthroponotic parasites, probably due to differences in exposure. Intensity and duration of oocyst shedding tends to be longer for infections with the C. parvum human genotype than for those with zoonotic genotypes. After an initial Cryptosporidium infection, some children experienced subsequent infections with homologous and heterologous Cryptosporidium parasites, often within a year of the first infection. Although many lineages of the C. parvum human or bovine genotype are detected in sporadic cases from the same geographic area, only one or two subgenotypes have been found in each foodborne or waterborne outbreak examined. One subgenotype of the C. parvum human genotype has been involved in multiple waterborne and foodborne outbreaks in the U.S. and UK, indicating that certain Cryptosporiodium parasites may have higher transmission potential than others. In contrast, many Cryptosporidium species and genotyps have been found in water, most of which are probably not human pathogenic. Direct genetic linkage of Cryptosporidium oocysts found in water with parasites in affected humans has been made in several waterborne outbreaks. These findings reveal the utility of molecular tools in the differentiation of Cryptosporidium parasites and in epidemiologic studies of cryptosporidiosis.

Challenge: What is the role of foodstuffs in the aetiology of human cryptosporidiosis?

Cryptosporidium oocysts have been isolated from several foodstuffs (Table 1) and these have mainly been associated with fruit, vegetables and shellfish. The association of oocyst contamination of these produce is particularly important from a public health viewpoint, as these products are frequently consumed raw without any thermal processing to inactivate contaminating oocyts. Mollusc filter feeders such as oysters, mussels and clams pose a risk because they can concentrate pathogens which are removed from large volumes of potentially contaminated water. Such waters may be polluted with sewage, industrial and agricultural run-off, and storm run-off water, on a regular basis (Fayer et al., 1998). In addition, Cryptosporidium has been implicated in several cases and outbreaks of human gastrointestinal disease, either by direct isolation of oocyts from the suspected foodstuff or by epidemiological association (Millar et al., 2002).

Challenge: Is there a role for novel techniques for testing the viability of oocysts?

Assessment of viability of this organism is important, as this may be related to the infectivity potential to humans of any positive water source or food item being consumed. Previously, there have been problems in the phenotypic identification of viable from non-viable oocysts. Historically such determinations were performed by animal challenge

studies or by excystation. However more recently, inclusion of the DAPI test [4'6-diamidino-2-phenylindole], which demonstrates viability through the presence of a fluorescent sky-blue coloration due to permeability of this molecule in viable oocysts, has been a valuable marker of oocysts viability. However Korich et al (1990) found vital dye exclusion to be unreliable as a viability indicator during study of the affects of disinfectants on oocyst survival. Emerging molecular technologies, including NASBA, (nucleic acid sequence-based amplification) may allow for reliable determination of the viability of oocyts. NASBA methodology is a novel technique in diagnostic microbiology, which as yet has not been applied to the molecular diagnosis of Cryptosporidium parvum. NASBA methodology offers the potential of a highly sensitive and specific method for the detection of Cryptosporidium parvum, without the need for highly complex reference laboratory facilities. This method effectively "deskills" complex molecular techniques, yet concurrently maintains the advantages of both specificity and sensitivity of molecular assays. This method would allow for differentiation of viable from non-viable oocysts and potentially offers a more reliable assay to the DAPI technique to assess viability, as well as allowing for quantitation of numbers of viable oocysts in a water or food source. Such an approach in these circumstances would allow for the immediate introduction of an intervention or several control strategies, thereby minimising risks to public health and maintaining corporate due diligence.

Challenge: What role can HACCP play in improving food safety with respect to Cryptosporidium?

Cryptosporidium present several potential hazards within the food processing sector. These hazards may be subdivided into (i) those where the parasite is introduced to the foodstuff through contaminated raw ingredients, e.g. unwashed lettuce destined for "ready-to-eat" (RTE) salads, (ii) where the parasite is introduced during food processing due to addition of contaminated water, as an important ingredient of the foodstuff, e.g. in soft drinks production, (iii) where the parasite is introduced during food processing, as a contaminant of cleaning of equipment with non-potable water or contaminated potable water, (iv) introduction of the parasite through pest infestations, e.g. cockroaches, house flies, mice and rats, and (v) introduction of the parasite to processed foodstuffs from positive food handlers. The associated risk from each of these potential routes of entry of oocyst into the foodstuff should be controlled through an integrated HACCP approach for the reduction/ elimination of viable oocyts in the final food product. Where manufacturers are producing RTE foodstuffs requiring no further processing, e.g. domestic cooking, then the critical control points in such circumstances are required to be absolute, i.e complete elimination of the hazard from the RTE foodstuff. Manufacturers should also be aware that the globalization of food production, including the sourcing of raw materials from several different countries. This may open new mechanisms for the transmission of this parasite, therefore food processors must be diligent in sourcing ingredients with stringent HACCP-controlled specifications and a commensurate degree of product sampling/testing, to verify the efficacy of such controls. Although industry should strive to obtain this objective even when processing raw foodstuffs , e.g. raw meats, the critical control points in such circumstances are in practice less stringent, as these foods will receive sufficient cooking to render viable oocysts non-infective. However, contaminated raw produce may pose an important cross-infection hazard with the potential indirect transmission through contaminated utensils and work surfaces. However, the effectiveness of any such control is reliant on a satisfactory method of isolation and detection from the foodstuff.

Challenge: Where do we go from here?

At a strategic level, it is important that each nation has the ability to reliably genotype and subgenotype human and food/water/animal/environmental Cryptosporidium, using a standardized molecular methodology. It is therefore important that as we develop such capabilities, we do so in unison, so that there is added-value to the epidemiological data, whereby comparisons may be made locally, nationally and internationally. Presently, there is no consensus on detection, genotyping and subgenotyping methodology. Further work is urgently required on all these aspects.

With the development of improved laboratory detection systems for both isolation, identification and viability testing, coupled with food-related outbreaks, more attention is being placed on the potential transmission of this agent through foodstuffs. Thus, food testing laboratories will experience an increased demand for having such assays in place to routinely monitor for this organism and more importantly, its viability. As the majority of modern detection systems are based on a variety of molecular platforms, including PCR, RT-PCR, NASBA, LightCycler, this may prove a diagnostic challenge for a number of food industry laboratories, which predominantly rely on conventional detection systems based on culture.