Molecular detection and identification of Cryptosporidium species isolated from human and animal sources in Limpopo and Gauteng Provinces

Abstract

Background: Diarrheal diseases constitute an important problem among children but also among HIV positive patients particularly in developing countries such as South Africa. Cryptosporidium infect humans and has been shown to be an important cause of infection among different types of animals. Because of its small size, Cryptosporidium can easily go through the water purification system and can easily become a cause of an epidemic. Previous studies have shown that Cryptosporidium is an important cause of diarrhea in Limpopo Province. However, very few studies have been conducted on the genetic diversity of these organisms in the region. Therefore, the aim of this study was to detect and identify the genetic diversity of Cryptosporidium species from humans and animals in Giyani situated in the northern part of South Africa and Pretoria situated in the central part of the country. Methodology: A total of 560 samples were collected from human and animals and were all screened by microscopy using modified Ziehl-Neelsen staining technique. All the samples were tested by Enzyme-Linked Immunosorbent Assay (ELISA) using the Cryptosporidium II kits from Techlab, Virginia, USA. Positive samples from microscopy and ELISA were examined by different PCR protocols including conventional PCR for amplification of Cryptosporidium oocyst wall protein (COWP) region; Real-time PCR employing SYBR Green detection format for amplification of 18S rRNA region; Real-time PCR employing Hydrolysis probes detection format for amplification of SSU rRNA region; Real-time PCR specific for amplification of C. hominis region and C. parvum region. Positive samples from real-time PCR that gave clear bands on gel electrophoresis were sent for sequencing. The sequences were analysed using Staden package software to edit the nucleotides, Bioedit and MEGA6 software were used to align sequences and draw phylogenetic trees. The SPSS software was used for statistical analysis. xiii Results: The overall prevalence of Cryptosporidium as detected by ELISA method from the samples collected from humans was 41.2% (239/580). The prevalence was higher from the rural area 73.0% (159/218) compared to the urban area 22.1% (80/362) and the difference was statistically significant (χ2 = 145.1; p = 0.0001). Due to the limited amount of samples, only 134 ELISA-positive samples were tested using real-time PCR. Of these samples, 35.8% (48/134) tested positive. Of 48 real-time positive samples 25 were successfully sequenced and two different species (C. hominis and C. muris) were identified. Of all the sequences obtained, one (4.0%) was C. muris and 20 (80%) were C. hominis isolated from rural area, whereas 16.0% (4/25) were also C. hominis isolated from samples obtained from urban area. Cryptosporidium was not associated with diarrhea in the present study. A total of 85 samples were collected from animals (52 from cattle and 33 from goats) and of these 4 (4.7%) were positive by microscopy and ELISA. All these samples were non diarrheal. Conventional PCR also detected a similar number. Of these 4 positive samples, 1 was from a male goat, while the 3 others were obtained from female adult goats. Real-time PCR detected 56.5% (48/85) positive samples. Only 12 of the 85 animal samples were diarrheal and of these 4 were positive for Cryptosporidium. The prevalence of Cryptosporidium infection was higher 68.4% (13/19) in male animals compared to female animals 53.0% (35/66). The prevalence rates in cattle and goats were 55.8% (29/52) and 60.6% (20/33) respectively. Of 48 real-time positive samples from animals, 12 (25.0%) were successfully sequenced and two species (C. parvum and C. andersoni) were identified. Of these 6 were from cattle and the other 6 were from goats. Out of the 12 samples 10 (83%) were C. parvum while 2 (17%) were C. andersoni. Of the two C. andersoni, one was from a goat and one was from a cow. Of the 10 C. parvum, 5 were from goats and 5 were from cattle. xiv In conclusion, microscopy remains the low sensitive tool for the detection of Cryptosporidium while real time PCR appeared to be far much more sensitive by detecting more samples than all the three other methods combined. Closer to the real time PCR was ELISA that detected also more samples compared to conventional PCR and microscopy. The present study identified C. muris from humans’ samples in our area for the first time. However, C. hominis remains the dominant species that infects humans in our area. Cryptosporidium species was mostly found in samples from asymptomatic individuals. In animals, C. parvum was the most commonly isolated organism while C. andersoni was identified in our region for the first time as well and occurred in both goats and cattle. Populations in the affected areas need to be made aware of the infections so that care should be taken to avoid the spread of infection in water sources or in immunocompromised individuals.