Viability Assessment of Cryptosporidium parvum Oocysts by Vital Dyes: Dry Mounts Overestimate the Number of “Ghost” Oocysts

Introduction

Worldwide, Cryptosporidium spp. are the most frequent causes of waterborne outbreaks of protozoal disease (Baldursson and Karanis, 2011). Moreover, in a recent global multicriteria-based risk ranking of foodborne parasites, Cryptosporidium spp. were ranked five (FAO/WHO, 2014). The infective stage is the small, thick-walled, robust oocyst, which can be found in the environment, including water bodies. As both live and dead oocysts are visible by direct screening of water samples, simple enumeration of visible oocysts gives no information about the viability. Obviously, the public health significance of dead oocysts is minimal, whereas live oocysts can have a significant public health impact (Campbell et al., 1992).

Live oocysts are described as either infective or viable; infectivity specifies that excysted sporozoites can invade and multiply within gut epithelial cells, whereas viability specifies that oocysts can develop under favorable conditions and are capable of excystation (Robertson and Gierde, 2007). Oocyst viability can be evaluated based on oocyst wall permeability, that is, inclusion and/or exclusion of the vital dyes 4′,6-diamidino-2-phenylindole (DAPI) and propidium iodide (PI) (Campbell et al., 1992). This method is relatively fast and easy, and is still widely used despite its age. The method is typically performed after acid pretreatment of oocysts, staining in suspension and mounting of a subsample to a microscopic slide (Campbell et al., 1992). However, some studies have used modifications by which oocysts were dried onto microscope slides before staining or the viability was evaluated in dry mounts after staining in suspension (Castro-Hermida et al., 2010; Forslund et al., 2011; Petersen et al., 2012). Robertson et al. (2014) demonstrated that drying oocysts to microscope slides before staining result in a considerable underestimation of viable oocysts compared with oocyst staining and evaluation in wet mounts (Campbell et al., 1992). However, drying oocysts onto slides after staining in suspension (Forslund et al., 2011; Petersen et al., 2012), relative to the method published by Campbell et al. (1992), has not previously been described. By drying oocysts onto microscope slides after staining a larger sample volume can be examined and the oocysts can more easily be located because they are fixed to the slide. Thus, the aims of this study were (1) to compare DAPI and PI inclusion/exclusion after staining in suspension and evaluation in dry and wet mounts, respectively, and (2) to evaluate the effect of acid pretreatment on the outcome of DAPI/PI staining.

Materials and Methods

Cryptosporidium parvum oocysts from a naturally infected diary calf were purified by immunomagnetic separation according to the manufactures instruction with slight modifications as described by Petersen et al. (2016), and the oocysts were identified to species level (Langkjaer et al., 2007). The purified oocysts were divided into three groups (protocol A, B and C), each consisting of nine 100 μl replicate samples. Each of the nine replicate samples was treated according to protocol A, B, or C as outlined in Table 1. All replicate samples were added 10 μL DAPI and 10 μL PI, incubated at 37°C for 2 h, washed twice in Milli-Q water, and incubated 1 h with 25 μL fluorescein isothiocyanate labeled monoclonal antibody (FITC) (Crypto-CEL IF test; Cellabs, Australia) to aid the identification either in suspension or after being dried to slides (Table 1). Oocysts in all replicate samples were scored microscopically at 400× magnification according to DAPI and PI inclusion (+)/exclusion (−) as described by Campbell et al. (1992). In brief, oocysts were categorized as DAPI+/PI− (viable), DAPI+/PI+ (nonviable), DAPI−/PI− (potentially viable), or “ghosts” (dead) (Fig. 1). The microscope was equipped with the following filter blocks: 350 nm excitation, 450 nm emissions for DAPI; 500 nm excitation, 630 nm emissions for PI; and 495 nm excitation, 519 nm emissions for FITC. At least 100 oocysts were scored per replicate sample. The final oocyst score was the proportion of viable oocysts (sum of DAPI+/PI− and DAPI−/PI−), nonviable oocysts (DAPI+/PI+), and “ghosts” oocysts. All data were transformed by natural logarithm of x (ln X) and differences in means of viable, nonviable, and ghost oocysts between protocols were analyzed by one-way analysis of variance (ANOVA).

OPEN IN VIEWER

FIG. 1.

Illustration of Cryptosporidium parvum oocyst categories based on inclusion and exclusion of DAPI and PI, and direct light microscopy of DAPI− oocysts. DAPI, 4′,6-diamidino-2-phenylindole; PI, propidium iodide.

Results and Discussion

Significantly more oocysts were scored as viable when evaluated in wet mounts compared with dry mounts (Fig. 2). This is in accordance with Robertson et al. (2014) who found that drying oocysts to slides before staining resulted in underestimation of the viability, since drying is known to break the oocyst membrane and inactivate oocysts. However, as all oocysts in our study were stained with vital dyes in suspension, we expect the oocysts to have been correctly stained according to viability and unaffected by subsequent drying. Nevertheless, viability evaluation based on dry mounts resulted in ∼25% underestimation of viable oocyst, whereas the proportions of nonviable oocysts (DAPI+/PI+) were comparable for dry and wet mounts (Fig. 2). Remarkably, the proportion of “ghosts” was significantly higher in dry mounts than in wet mounts (Fig. 2). Apparently, the drying after staining in wet mounts caused the oocyst walls of some otherwise viable oocysts to rupture and the sporozoites to escape, which made the oocysts appear as “ghosts,” although they might have been viable before drying. In general, the oocysts were more easily localized on the slides in dry mounts. However, viable oocysts (DAPI+/PI−) were more effortlessly identified in the wet mounts since the nuclei of the sporozoites were more prominent than dry mounts.

OPEN IN VIEWER

FIG. 2.

Proportion of Cryptosporidium parvum oocysts scored as viable (sum of semipermeable [DAPI+/PI−] and potentially viable [DAPI−/PI−]), non-viable (DAPI+/PI+), and “ghosts” after three different staining protocols. Protocol (A) acid pretreatment, oocysts scored in wet mounts, protocol (B) no acid pretreatment, oocysts scored in dry mounts, and protocol (C) no acid pretreatment, oocysts scored in wet mounts. Significant differences (p < 0.05) between the protocols are indicated for each of the DAPI/PI categories.

The proportions of viable oocysts were similar (p = 0.110) when scored in wet mounts irrespective of the acid pretreatment. This is in agreement with Campbell et al. (1992) who demonstrated that the proportion of viable oocysts (cervine–ovine strain) was independent of acid pretreatment, but they also showed that acid pretreatment of a different (bovine) Cryptosporidium strain increased the proportion of viable oocysts. The acid pretreatment step is known to permeabilize the oocyst wall of some otherwise impermeable (DAPI−/PI−) oocysts, whereby they become semipermeable (DAPI+/PI−) when assessed by the vital dye assay (Campbell et al., 1992; Robertson et al., 1993). Therefore, by omitting the acid pretreatment step, the proportion of DAPI+/PI− oocysts may be underestimated in some cases depending on Cryptosporidium species/strain.

In conclusion, we have demonstrated that evaluation of C. parvum in dry mounts can be used for assessment of oocyst inactivation rates provided that the oocysts are stained in suspension and assessment is based on the proportion of nonviable oocysts. However, viability data obtained by scoring oocysts in dry mounts give an erroneous impression of low viability even if the oocysts are stained in suspension.