Comparison of three diagnostic methods in the diagnosis of cryptosporidiosis and gp60 subtyping of Cryptosporidium parvum in diarrheic calves in Central Anatolia Region of Turkey
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Introduction
Cryptosporidium species, the causative agents of cryptosporidiosis, are the important opportunistic intestinal pathogens that can infect various hosts including cattle and humans (1,2,3). The ingestion of contaminated water and foods is the main risk factor for calves and other hosts as well as humans for Cryptosporidium spp. infections (4,5). Cattle, especially pre-weaned calves are known as the most common hosts of Cryptosporidium spp. with high infection and oocyst excretion rates in their feces (6,7). Cryptosporidium parvum is recognized as the primary zoonotic species causing cryptosporidiosis in pre-weaned calves resulting in diarrhoea, weight loss, dehydration, delayed growth, and important economic losses related to calf morbidity and mortality (6,8). Furthermore, young calves are regarded as an important potential source of human cryptosporidiosis in various outbreaks worldwide (9).
Several techniques are used for the laboratory detection of Cryptosporidium in fecal samples. The oocysts of Cryptosporidium cannot concentrate well using standard concentration techniques but can be identified by microscopy combined with various staining methods, i.e. modified Ziehl-Neelsen of fecal smears (10). On the other hand, the diagnostic efficiency of conventional microscopic examination might be low due to sporadic oocyst shedding and the presence of few oocysts in the fecal samples. Examination of the stained fecal smears also has a disadvantage for the detection of Cryptosporidium because the oocysts can easily be confused with other materials (e.g fecal debris, yeast cells, and bacterial spores) present in the smears (1). In spite of higher costs and needs for infrastructure needs and high technical expertise, the use of molecular tools for the detection of Cryptosporidium has the advantages of improved sensitivity and specificity. These methods also allowed us to better understand the genetic diversity of Cryptosporidium species that are highly host-specific in genotype and subtype levels, transmission routes, and related impacts on public health (8,12). Currently, small-subunit rRNA (SSU rRNA) gene-based nested PCR-RFLP analysis have been developed for the detection and identification of Cryptosporidium species in fecal and environmental samples (13,14,15). A quantitative real-time PCR (qPCR) assay has also been developed to identify and quantify Cryptosporidium DNA in fecal and various environmental samples (16,17,18).
Host-specific and zoonotic subtypes have been described using sequence analysis of the 60 kDa glycoprotein (gp60) gene that is widely used in subtyping of C. parvum because of its high polymorphism (19,20). To date, at least 20 subtype families have been described within C. parvum (21). Subtypes IIa and IId are considered zoonotic subtypes found in both humans and ruminants. The subtype family IIa of C. parvum is common in calves and the IIaA15G2R1 is the most prevalent subtype in many countries (9,21,22).
Cryptosporidium parvum is known as one of the main causative agents of neonatal calf diarrhoea in herds in Turkey. However, there have been little data on the molecular characterization and subtyping of C. parvum in pre-weaned calves in Turkey (23,24,25). This study was conducted to evaluate diagnostic efficiency and usefulness of the nested PCR, Real-Time PCR assay and conventional modified Ziehl-Neelsen staining of the faecal smears comparatively, for the detection of cryptosporidiosis in calves with neonatal diarrhoea in the traditional farms of private smallholders in three districts of Konya province located in Central Anatolia Region of Turkey. Cryptosporidium parvum subtypes were also investigated by sequence analysis of the gp60 gene to reveal zoonotic transmission dynamics of cryptosporidiosis in the research area.
Materials and Methods
Sampling strategy, data collection and identification of oocysts
Farms from three districts (Merkez, Karatay, and Selcuklu) were selected between 2011 and 2012 according to local veterinarians’ opinion. The inclusion criterion for farms was a size of 5 to 50 calves traditionally owned by private smallholders with a history of neonatal calf diarrhoea. Based on sample size calculations (EPIDAT 3.1), a total of 20 farms were selected from a total of 208 officially recorded farms belonged to private smallholders in the research area. Environmental conditions of the farms included in the study shared common characteristics such as similar climatic conditions, traditional barns and pasture based raising of calves. A total of 194 fecal samples from early pre-weaned (n=158, ≤15 days old) and pre-weaned calves (n=36, 15 to 40 days old) with neonatal calf diarrhoea were collected from the predefined farms (5 to 15 calves per farm). Approximately 30-50 g fecal samples were taken from fresh droppings on the ground after defecation, then immediately placed into a sterile plastic container and held at 4°C until analysis. The fecal smears were prepared and stained using the modified Ziehl-Neelsen staining method for the presence of oocysts (26).
Genomic DNA extraction
200 mg of each fecal sample within a microcentrifuge tube were frozen in liquid nitrogen and subsequently heated before DNA extraction. Genomic DNA (gDNA) was extracted from disrupted samples by the QIAamp DNA Stool Mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. DNA concentrations of the extracts were measured with the Qubit Fluorometric Quantitation (Thermo Fisher Scientific, USA) to optimizing the amount used in the PCR master mix. The extracted DNAs were then preserved at -20°C until the time of molecular analysis.
Quantitative Real time PCR (qPCR)
TaqMan qPCR assay with primers and probes targeting Cryptosporidium at genus level and C. parvum was carried out according to Jothikumar et al. (27). A CFX Connect Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) was used for all qPCR analysis. The reaction conditions for the qPCR were adjusted according to the protocol described by Jothikumar et al. (27) using SsoAdvanced Universal Probes Supermix (Bio-Rad, Hercules, CA, USA) and 10-30 ng template DNA in each reaction
Nested PCR-RFLP of SSU rRNA and amplifiication of gp60
gDNA isolates were analyzed with nested PCR to amplify approximately 840 bp and 850 bp fragments of the SSU rRNA and gp60 gene regions, respectively with specific primers following the protocols previously described (28,29,30). PCR products were electrophoretically resolved in 1.5 % agarose gels containing SYBR™ Safe stain (Thermo Fisher Scientific, USA) and visualized using Fusion FX Gel Documentation System (Vilber Lourmat, France). For RFLP analysis, secondary PCR products of the SSU rRNA gene were digested using FastDigest SspI and FastDigest MboII (Thermo Fisher Scientific, USA). Identification of Cryptosporidium species was performed according to restriction patterns described previously (31).
Sequence analysis of SSU rRNA and gp60 subtyping
All gel-purified secondary PCR products of the two target genes (SSU rRNA and gp60) were sequenced (Macrogen, Netherlands) in both directions to confirm RFLP results and detect subtypes of C. parvum. The nucleotide sequences were assembled with Geneious Prime 11.0.2 software (http://www.geneious.com) The consensus sequences were aligned with reference sequences downloaded from GenBank by MUSCLE (32) using Geneious prime. Subtypes of C. parvum were named with established for Cryptosporidium spp. gp60 nomenclature system (3).
Results
Identification and distribution of C. parvum according to the diagnostic methods
A total of 92 and 104 samples were found positive by utilizing microscopic examination (Figure 1a) and molecular assays, respectively (Table 1) and there was full agreement between the results of these two tests. Microscopic analysis revealed a lower number of Cryptosporidium positivity and its diagnostic sensitivity and specificity were determined as 88.5% and 100.0% compared to molecular tools, respectively (Table 1). 15 of samples from diarrheic pre-weaned calves were found positive in the C. parvum species-specific real-time PCR assay (Figure 1c) and there was no positivity at the genus level without C. parvum.
Figure 1
a: Cryptosporidium oocysts in fecal smears stained with the Ziehl-Neelsen by microscopic examination (×1000); b: Electrophoretic profiles of the SSU rRNA PCR products of C. parvum isolates digested with the endonucleases SspI (lanes 1 to 4) and MboII (lanes 5 to 8). M: Molecular size marker (100bp); c: Amplification curves with TaqMan real-time PCR analyses. x1, y1 and z1: C. parvum species specific qPCR curves; x2, y2 and z2: Cryptosporidium genus specific qPCR curves belonging to the same isolates; n: No DNA control and negative samples
Comparison of the results of microscopic examination and molecular assays in the detection of Cryptosporidium in calves and evaluation of diagnostic sensitivity and specificity.
Method
Molecular tools
Sensitivity
Specificity
Positive
Negative
Microscopy
Positive
92
0
85.5%
100%
Negative
12
90
(80.7-93.9%)
(96.0-100.0%)
Total
104
90
The distribution of C. parvum over the age groups of pre-weaned calves with diarrhoea based on the molecular assays is presented in Table 2. The prevalence in early pre-weaned calves (≤15 days old) (56.3%) was higher than the prevalence in pre-weaned calves (15 to 40 days old) (41.7%). However, this difference was not statistically significant (p > 0.05) (Table 2).
Distribution and subtype of C. parvum over age groups of the examined calves based on molecular diagnostic tests used in the study.
Age group (days)
Total no. of samples
Cryptosporidium parvum
No. positive (%)
χ2
p
gp60 subtype
Early pre-weaned calves (≤ 15)
158
89 (56.3)
2.535
0.139
IIaA13G2R1
Pre-weaned calves (15-40)
36
15 (41.7)
Total
194
104 (53.6)
RFLP and sequence analysis of SSU rRNA
RFLP analysis of all SSU rRNA nested PCR products from the isolates that were identified as C. parvum in TaqMan qPCR revealed three (449, 267, and 108 bp) and two (771 and 76 bp) band profiles specific to C. parvum with the restriction enzymes SspI and MboII, respectively (Figure 1b) confirming the qPCR results. No mixed infections were found in the examined specimens with the RFLP analysis of the SSU rRNA gene. DNA sequence analysis and BLAST search of the SSU rRNA nucleotide sequences of the isolates showed species-based homology with the corresponding sequences of C. parvum in GenBank further confirming the results of qPCR and RFLP analysis. SSU rRNA sequences of C. parvum characterized in the study were deposited to the NCBI GenBank database under accession numbers KF008173-75.
C. parvum subtypes with gp60 sequencing
Gp60 gene was successfully amplified and sequenced from isolates of all C. parvum positive samples. Sequence analysis identified only one subtype belonging to family IIa (Table 2).
IIaA13G2R1 subtype has 13 copies of TCA (A) and, 2 copies of TCG (G) trinucleotide repeat encoding serine, and one copy of the sequence ACATCA (R1) following the trinucleotide repeat. Gp60 sequences of the isolates were submitted to the NCBI GenBank database under accession numbers KF008176-85.
Discussion
Cryptosporidium parvum is an important causative pathogen of neonatal calf diarrhoea and definitive and differentiative diagnosis is crucial for effective treatment and taking control measurements against the disease (9). Therefore, the diagnosis of cryptosporidiosis needs to be highly specific and sensitive. Even though microscopy is currently the common and practicable method in the diagnosis of cryptosporidiosis it has the disadvantages of poor sensitivity and specificity, especially in the case of sporadic shedding of oocyst and also low oocysts in fecal samples. Besides, the oocysts can easily be confused with other particles present in the smears, especially in the absence of considerable training and expertise. Furthermore, it is difficult to distinguish Cryptosporidium species (34). In our study lower positivity of Cryptosporidium was detected using conventional microscopic examination of the fecal smears compared with molecular assays resulting in a decreased sensitivity of 88.5%. However, the specificity was determined as 100.0%. This result is by the previous studies that used various molecular techniques to detect Cryptosporidium species in human and animal fecal and environmental samples (9,35). Nested PCR-RFLP and qPCR assays determined the same samples as positive for Cryptosporidium revealing the high sensitivity and specificity of both techniques in our study. Despite their high accuracy in the diagnoses, these molecular assays have the dis-advantage of being time-consuming and having many steps in the diagnostic procedure. Furthermore, they are also expensive due to the complex equipment needed to run assays.
SSU rRNA-based PCR is the most commonly used assay in many epidemiological studies to detect and identify Cryptosporidium species. However, this technique requires further analysis involving RFLP or DNA sequencing to distinguish species (14,29,31,36,37). Recently, a quantitative real-time PCR assay has been developed for the identification of Cryptosporidium species and C. parvum strains in clinical samples and water matrices (14,16). qPCR has several advantages including the elimination of post-amplification handling, easier automation, reducing contamination risk and assay times, and processing of high numbers of samples (27,38,39). Jothikumar et al. (27) reported that the duplex TaqMan qPCR approach can specifically detect all Cryptosporidium species at the genus level and C. parvum in a single reaction. Our results agree with the findings of Jothikumar et al. (27) and also sequence analysis of the SSU rRNA gene region confirmed the results of the qPCR assay. Development of accurate qPCR assays to determine and discriminate the bovine Cryptosporidium species might be useful for epidemiological investigations.
We determined high C. parvum infection in pre-weaned calves with diarrhoea with an overall prevalence of 53.6%. Neonatal calf diarrhoea has a multifactorial etiology and C. parvum is recognized as frequently associated with neonatal diarrhoea (8,40). The dominancy of C. parvum as the causative pathogen of diarrhoea in pre-waned calves has been also reported in the herds of several countries (29,41,42,43) as well as Turkey (24,25,44). On the other hand, the other bovine Cryptosporidium species such as C. bovis and C. ryanae are responsible for the majority of cryptosporidiosis in post-weaned calves and heifers (25,45,46,47,48,49). The other bovine Cryptosporidium species, C. andersoni is commonly found in adult cattle (9,45,50). The absence of all the above species in diarrheic calves in our study further confirmed the age-related future of the bovine Cryptosporidium species. The predominance of C. parvum in our study is somewhat expected and consistent with the age of the studied calves, as this is the most frequent species in neonate animals. This approach allowed us to identify a unique subtype IIaA13G2R1 of C. parvum, within the family IIa, which is widely recognized for its zoonotic potential. This subtype has been previously characterized as a common subtype in calves. On the other hand, some researchers also reported the presence of C. bovis and C. ryanae in pre-weaned calves in the absence of C. parvum (49,51). Conversely, C. bovis was determined to be the most prevalent species in pre-weaned dairy calves in some countries such as China (52,53), Sweden (54), Ethiopia (49) and Western Australia and New South Wales (5), and some authors highlighted that this status may be associated with different farming practices and seasonal differences (56). We also determined that C. parvum was more prevalent in the early pre-weaned calves (≤15 days old) than the pre-weaned calves of 15 to 40 days old. However, this difference was not statistically significant and C. parvum could affect pre-weaned calves during the growing period as indicated by several researchers (50,57,58,59).
Sequence analysis of the gp60 gene revealed the presence and wideness of unique subtype IIaA13G2R1 of C. parvum in the zoonotic IIa family in the study area. In addition to C. parvum speciation, we decided to pursue subtyping using a nested PCR assay targeting the gp60 gene followed by sequencing. IIaA13G2R1 has been previously characterized as a common subtype in calves from the different regions of Central Anatolia and Mediterranean regions in Turkey (25). This subtype has also been reported in calves in Canada (60), Belgium (61), Algeria (62), the Netherlands (63), in people with HIV/AIDS in Malaysia (64), in goat kids and lambs in Algeria (65), in ponies in the United States (66). The occurrence and wideness of IIaA13G2R1 in several ruminants, equids, and also humans indicate the risk of zoonotic transmission of cryptosporidiosis in the research area.
Conclusion
In conclusion, our study confirms the usefulness of nested PCR-RFLP and qPCR assays in the accurate diagnosis of Cryptosporidium in calves. Regarding the high prevalence of zoonotic subtype IIaA13G2R1, our results highlight the potential risk of pre-weaned calves for human infections with C. parvum in the research area. Thus, control measures should be considered for reducing the risk of the zoonotic transmission of C. parvum. Comparative molecular surveys on humans from different geographic regions, especially in farmworkers that are in close contact with C. parvum-infected animals are needed to improve our understanding of cryptosporidiosis epidemiology and C. parvum subtype diversity in Turkey.
