*2.1. Cryptosporidium Infection on Rabbit Farms*

*Cryptosporidium* spp. were detected by PCR analysis of the SSU rRNA gene in 28 (11.9%) of the 235 fecal samples analyzed in the study. Eight of the nine farms examined were positive for *Cryptosporidium* spp. Among the eight positive farms, the infection rates ranged from 4% to 24% (Table 1). The farms El-Dakahlia 3 and El-Gharbia 2 had high infection rates of 21% and 24%, respectively, while no infection was detected on farm El-Gharbia 1. By age, *Cryptosporidium* spp. were detected in rabbits of all ages, with a higher infection rate found in rabbits of <3 months (20%; Fisher's exact test = 11.237, *p* = 0.003 in the overall comparison) (Table 2). By animal breed, *Cryptosporidium* spp. were identified in all breeds, with slightly higher detection rates in Hi-Plus rabbits (15%) than in New Zealand (11%) and Rex (7%), although these rates were not statistically different (Fisher's exact test = 2.283, *p* = 0.333 in the overall comparison). Farms in El-Dakahlia province recorded higher *Cryptosporidium* occurrence (17%) than El-Gharbia (11%) and Damietta (7%) provinces (Fisher's exact test = 3.685, *p* = 0.157 in the overall comparison).

provinces, Egypt. **Farm Age (Month) No. of Samples** *Cryptosporidium* **spp. \* No. Positive (%) 95% Confidence Interval Lower Limit Upper Limit**

**Table 1.** Distribution of *Cryptosporidium* spp. in rabbits by farm and age group in El Dakahlia, El-Gharbia, and Damietta



**Table 1.** *Cont.*

\* Fisher's exact test: *t* = 12.258, *p* = 0.106.

**Table 2.** Factors associated with *Cryptosporidium* infection in rabbits in Egypt.


#### *2.2. Cryptosporidium Genotypes and Subtypes*

All 28 samples amplified by PCR analysis of the *SSU* rRNA gene had *C. cuniculus* by RFLP analysis (Figure 1). They produced two types of nucleotide sequences. Among them, ten sequences were identical to those in GenBank (AY120901, FJ262724, etc.), while two sequences had an A to T substitution near the 5 end of the partial gene. In the phylogenetic analysis of the *SSU* rRNA sequences, *C. cuniculus* sequences obtained from the 12 samples clustered with reference sequences from GenBank (Figure 2). Of the 28 *C. cuniculus* samples, 16 were successfully subtyped by sequence analysis of the *gp60* gene, with two subtypes being identified: VbA19 (*n* = 1) and VbA33 (*n* = 15). These sequences were identical to each other in the non-repeat regions but had one A to T substitution compared to sequences (KU852732, GU097641, GU097647, GU971639, etc.) in GenBank (Figure 3). The VbA19 subtype was found only in a 6-month-old rabbit from farm El-Gharbia 3, while the VbA33 subtype was found on other *Cryptosporidium*-positive farms. Among the samples from four cages of animals with diarrhea, one sample from a cage of 4-month-old rabbits on El-Dakahlia 1 was positive for *C. cuniculus* VbA33 (Table 3).

**Figure 1.** RFLP analysis of PCR products of *SSU rRNA* gene from *Cryptosporidium cuniculus* from rabbits using *Ssp*I (**A**) and *Vsp*I (**B**) restriction enzymes. Lane M: 100 bp molecular markers; Lane 1–9: *C. cuniculus*; Lane 10: positive control (*C. baileyi*).

**Figure 2.** Phylogenetic relationships among *Cryptosporidium* spp. based on the nucleotide sequences of the *SSU* rRNA gene through a maximum likelihood analysis based on substitution rates calculated with the general time reversible model. Numbers at the internodes represent bootstrap values (>50%) from 1000 replicates. The *Cryptosporidium cuniculus* samples identified in this study are labeled with red rhombus.

**Figure 3.** Phylogenetic tree of *Cryptosporidium* spp. based on *gp60* sequences through a maximum likelihood analysis based on substitution rates calculated with the general time reversible model. Numbers at the internodes represent bootstrap values (>70%) from 1000 replicates. The *Cryptosporidium cuniculus* samples identified in this study are labeled with red rhombus.


**Table 3.** Distribution of 16 *Cryptosporidium cuniculus* subtypes at the *gp60* locus in rabbits in Egypt.

\* From rabbit with diarrhea.

#### **3. Discussion**

The results of the present study suggest a common occurrence of *Cryptosporidium* spp. in rabbits in the study areas. In this study, the overall occurrence of *Cryptosporidium* spp. in rabbits was 11.9% (28/235). This is similar to the infection rates of 11.2% (24/215) in a study of two rabbit farms in Heilongjiang Province, China [35], and 13.2% (14/106) in rabbits residing in Sydney drinking water catchments [36]. It is, however, higher than

infection rates found in rabbits from Australia (6.8% or 12/176 and 8.4% or 22/263) [37,38], and Nigeria (3.7% or 4/107) [19], but lower than the infection rate recorded in pet rabbits in Japan (21.9% or 21/96) [16]. Several reports from China showed low infection rates of 1.03% (3/290), 2.4% (9/378), 3.4% (37/1081), and 3.4% (11/321) [13,39–41]. The differences in the infection rates of *Cryptosporidium* spp. among studies may be attributed to differences in sample size, rabbit breeds, management systems, geographic regions, and sample collection seasons. In one study, the infection rate of *Cryptosporidium* spp. in dead juvenile rabbits suffering from diarrhea was significantly higher than healthy ones (30.3% vs 3.3%) [16]. Among the nine farms examined in the present study, the occurrence of *Cryptosporidium* spp. on El-Gharbia 2 (24%) was higher than other farms (0–21%), possibly because of the poor hygiene and management practices on the farm.

Like in other animals, the infection rate of *Cryptosporidium* spp. is significantly higher in rabbits of youngest age. In this study, rabbits of <3 months had a significantly higher *Cryptosporidium* infection rate than older rabbits. Our findings are in agreement with observations in earlier studies, where the highest prevalence of *Cryptosporidium* spp. was recorded in young rabbits [13,39]. Similar age-associated occurrence of *Cryptosporidium* spp. has been reported in humans, cattle, and bamboo rats [33,34,42]. In the present study, a higher detection rate of *Cryptosporidium* spp. was recorded in Hi-Plus rabbits than Rex and New Zealand ones, possibly because of the high number of samples and sampling of many young animals. In contrast to our results, Rex and New Zealand rabbits were more susceptible than other breeds to *Cryptosporidium* infection in some earlier studies in China [13,35].

Generally, few clinical signs have been associated with cryptosporidiosis in rabbits, especially adult ones, and the infection is often not recognized due to the asymptomatic oocyst shedding [39,41]. This is in line with our results, where most rabbits were apparently healthy. Although two reports observed fatality in outbreaks of diarrhea in rabbits due to cryptosporidiosis [16,43], *Cryptosporidium* was detected in only one of the four samples from animals with diarrhea.

All isolates of *Cryptosporidium* spp. detected in our study were genotyped as *C. cuniculus*, which is one of the causes of human cryptosporidiosis and has zoonotic significance [44]. In some countries, such as the UK, Australia and New Zealand, many sporadic cases of human cryptosporidiosis have been attributed to infections with *C. cuniculus* [22,24,25,45]. It was recognized as the third most important *Cryptosporidium* species causing cryptosporidiosis in humans in the UK during 2007 to 2008 and New Zealand during 2009 to 2019 [24,25]. It was also associated with a waterborne outbreak of cryptosporidiosis due to contamination of treated drinking water by wild rabbits [46,47]. Humans may be infected with *C. cuniculus* via contaminated water or direct contact with rabbits [22].

In this study, based on *gp60* sequence analysis, the *C. cuniculus* isolates belong to two subtypes (VbA19 and VbA33) in the Vb subtype family. Previously, the VbA19 subtype was isolated from rabbits in the Czech Republic [46], while the VbA33 subtype was detected in humans in the UK [48]. The two subtypes identified in the present study, however, differed from them by one nucleotide in the non-repeat region. Currently, Va and Vb are the only two subtype families within *C. cuniculus*. Between them, subtypes in the Va subtype family are more commonly seen in humans while those in the Vb subtype family are more commonly seen in rabbits [35]. The occurrence of similar subtypes of *C. cuniculus* in humans and rabbits supports the zoonotic potential of *C. cuniculus* [41].

In conclusion, to the best of our knowledge, this is the first study on the genetic identity of *Cryptosporidium* spp. in rabbits in Egypt. The results of this study suggest a common occurrence of *C. cuniculus* in farm rabbits in several areas of the country. The detection of *C. cuniculus* in this study supports the potential role of rabbits as a source of human infections. Further studies from other localities in Egypt are needed to improve our understanding of the clinical and public health significance of *Cryptosporidium* spp., in rabbits in Egypt.
