1. IntroductionEscherichia coli is widely regarded as a commensal bacterium [1]. Nonetheless, some E. coli pathotypes are pathogens causing intestinal infections ranging from mild to severe in animals and humans [2,3]. Common pathotypes include Enterohaemorrhagic E. coli (EHEC), Enterotoxigenic E. coli (ETEC), Enteropathogenic E. coli (EPEC), Enteroaggregative E. coli (EAEC), Enteroinvasive E. coli (EIEC), and diffusely adherent E. coli (DAEC) [4]. Among these pathotypes, the EHEC serotype O157:H7 is recognized as the most important zoonotic pathogen. Globally, between 2007 to 2015, EPEC- and ETEC-related yearly human deaths amounted to 37,000 and 26,000, respectively [5]. The World Health Organization (WHO) has estimated that one-third of the population in low-income countries suffers from foodborne diseases, of which pathogenic E. coli is one cause [5].In low- and middle-income countries, calf diarrhea caused by E. coli is common and causes a considerable number of deaths [6,7,8]. For instance, diarrheal diseases caused up to 15% of preweaning calf mortality in Uruguay [9] and 31% [10] and 63% [11] in Ethiopia, where calf diarrhea is an issue, though it is also an issue in high-income countries. For instance, in the USA, where about 57% of weaning calf mortalities have been reported due to diarrhea, with the majority of the cases in calves less than one month old [12]. About 9% mortality was reported in male calves with diarrhea in Canada [13]. In Norway, the overall calf mortalities were 280,000 heads per year, and the economic loss was estimated to be USD 10 million in 2006 [14]. Overall, calf diarrheal disease mainly seemed to be associated with poor farming practices [15,16]. Thus, an infection with E. coli pathotypes in calves poses a significant risk to animal health with a considerable impact on productivity and the farmers’ economy, in addition to being a public health issue due to their zoonotic potential. Given the generally poorer hygienic conditions and disease prevention capacity in rural areas in low-income countries, these risks may be accentuated in such settings. However, data on the occurrence of different E. coli pathotypes in low-income countries is sparse [17,18,19,20]. Here, we provide such data from diarrheic calves in a district in Ethiopia, the most livestock-rich country in Africa [21]. 2. Materials and Methods 2.1. Study Area and Period The study was conducted in Basona Werana woreda (District), Ethiopia (Figure 1). The Basona District is found in the north Shewa zone of the Amhara regional state, 130 Km from the capital, Addis Ababa. The district is divided into 22 small administrative subdistricts (Kebeles). The district has a population of 120,930 individuals living in 27,686 households. The households are distributed in urban and rural areas with an urban to a rural proportion of 0.01 [22]. It has an annual temperature range of 6 to 20 °C and the rainfall varies between 950 and 1200 mm. The farming practice is a mixed crop–livestock (crops, cattle, sheep, and goat) type where cattle predominate. 2.2. Study Design and Sampling Technique

A cross-sectional study design was used. The sample collection was carried out from October 2020 to March 2021, following the contact of veterinarians in each subdistrict. Farm owners were informed about the sampling through their milk union and requested to report the occurrence of diarrhea in calves. All types of farms with diarrheic calves were included. Each farm was visited once during the study period. Samples were taken from diarrheic calves between 0 and 10 weeks old. The age groups were 0–2, 2–4, 4–7, and 7–10 weeks. One fecal sample was collected from one calf per farm, except on one occasion when three diarrheic calves were sampled on a single farm. The sample size was set based on the financial constraints and a reasonable representation of the farms.

A total of 100 fecal samples from diarrheic calves were collected from 98 farms. Five ml feces were collected from the rectum of the calves. All samples were temporarily stored in phosphate-buffered saline (HiMEDIA, Mumbai, India) and transported to Addis Ababa University, the Institute of Biotechnology, health biotechnology laboratory, and processed within 24 h of the collection time.

2.3. Isolation and Characterization of E. coli The isolation and characterization of E. coli were performed using standard bacteriological procedures [23]. Briefly, all of the collected samples were enriched in tryptose soya broth (TSB) (HiMedia, Mumbai, India) at 37 °C for 18–24 h. Subsequently, a loopful of this bacterial culture was streaked onto McConkey agar (HiMedia, Mumbai, India) and was grown at 37 °C for 18–24 h. The pink color colonies (lactose fermenter) from the McConkey agar culture were picked and transferred onto an Eosin Methylene Blue (EMB) (HiMedia, Mumbai, India) agar medium and incubated to grow at 37 °C for 18–24 h. The E. coli colonies were presumptively identified based on their appearance of a greenish metallic sheen on Eosin Methylene Blue (EMB) (HiMedia, Mumbai, India) medium. Three well-isolated presumptive E. coli colonies were randomly selected from each EMB plate and pure cultures from the selected isolates were used for further testing. The E. coli isolates were identified using the biochemical tests Indole, Methyl Red, Vogues–Proskauer test, and Citrate utilization (IMViC) [23]. 2.4. DNA Extraction The DNA was extracted from the E. coli isolates as described by Moore et al. [24] with a minor modification. Briefly, a single colony was subcultured in Luria Bertani (LB) broth media (HiMEDIA, Mumbai, India). A 1.5 mL bacterial culture was pelleted via centrifugation and subsequently resuspended and washed in 100 µL sterile deionized water. The 150 µL bacterial suspensions were lysed at 100 °C and then frozen at −20 °C. Finally, the suspensions were centrifuged at 10,000 rpm for 10 min, and 100 µL of supernatant containing DNA from each preparation was transferred into fresh Eppendorf tubes. The DNA preparations were quantified using a Nano-drop spectrophotometer and a ratio of 260 nm/280 nm was used to estimate the quality of the DNA. Moreover, gel electrophoresis was used to check the intactness of the isolated DNA. 2.5. E. coli Pathotyping Using PCRThe DNA from the E. coli isolates was subjected to a PCR amplification of ten virulence genes; bundle-forming Pili (bfp), attaching and effacing (eae), adhesion (daaA-E); shiga-toxin-1/2 (stx-1/2), heat-stable (st) and heat-labile (lt) toxins, hemolysin toxin (hly), virulence protein transporter (aatA), and invasion-associated protein (ial) using the primers listed in Table 1 and the E. coli isolates were classified into one of the six pathotypes based on their virulence genes. All primers and PCR conditions used for the virulence gene amplification were performed as described previously [25,26,27,28,29,30,31,32,33]. All of the PCR reactions were performed in a singleplex platform in a Prima 96 plus Thermal Cycler (Himedia Laboratories, Mumbai, India).

Each PCR reaction was carried out in a 25 μL final volume reaction mixture containing nuclease-free water (Himedia, Mumbai, India), 10X PCR buffer with 17.5 mM of MgCl2 (Himedia, Mumbai, India), 0.35 mM of each dNTPs (Himedia, Mumbai, India), 100 pM for each of the virulence gene-specific forward and reverse primers (Bioneer; Daejeon, Republic of Korea), 500 U Taq polymerase enzyme (DELTA Biotechnology, Ethiopia), and a DNA template. DNA samples that carried the relevant virulence gene(s) and nuclease-free water (without template) were used as the positive and negative control, respectively. PCR amplification products were processed using 1.5% agarose gel electrophoresis at 100 Volt for one hour; the gels were stained with ethidium bromide in 1× TAE buffer (40 mM Tris-HCl, 20 mM Acetate, 0.5 mM EDTA, pH 8.3). A DNA ladder of 100 bp (Himedia, Mumbai, India) was run in parallel with the PCR products to determine the size of the amplicons. Separated PCR products were visualized, photographed, and documented in a gel documentation system (Bio-Rad, Feldkirchen, Germany).

2.6. Farm DataA structured questionnaire was prepared and used to collect the relevant farm data (Supplementary Table S1). In the questionnaire, farm owners were given closed-ended questions about the overall management of the calves. The questions covered the sex, age, and breed of the calf, colostrum and supplementary feeding practice, housing type, and flooring. In addition to the questionnaire, data about the hygiene was collected through a direct observational judgment by the first author. To grade the hygiene of the housing, a four-point scale ranging from 1 to 4, 1—very poor, 2—poor, 3—good, and 4—very good, was used. For the housing hygiene evaluation, a checklist was used to minimize the judgment biases, containing the following aspects: proper manure disposal, dryness, aeration, and the absence of nearby waste, each with 1 point (Supplementary Table S2). The questionnaire and observational data as well as the fecal samples from diarrheic calves were collected during the same visit to the farm. Each farm was visited once between October 2020 and March 2021. 2.7. Data Processing and Analysis

Data related to the sex, age, breed, colostrum feeding status, time of the colostrum feeding, type of supplementary feed, housing type, and hygiene were coded and entered into a Microsoft® Excel spreadsheet. The responses to the multiple-choice questions were coded into categorical variables. Similarly, the “Yes” or “No” responses were categorized as “1” or “0”, respectively. The housing hygiene “Very poor” to “Very good” responses were coded into numbers ranging from 1 to 4.

Descriptive statistics were used to analyze the occurrence of virulence genes and E. coli pathotypes. The questionnaire responses and their association with the proportion of detected E. coli pathotypes were analyzed using the chi-square test, with p < 0.05 considered as a significant association. A pairwise analysis of the co-occurrences of virulence genes was carried out using a two-tailed chi-square test, with p < 0.05 seen as a strong correlation. The software SPSS (IBM SPSS Statistics 28.0.0.0) was used for the analyses.

4. Discussion The present study intended to describe the occurrence of E. coli pathotypes and overall calving practices in a low-income setting. Livestock production is important in many Sub-Saharan countries like Ethiopia, for the national economy as well as for the livelihood of individual families, but the sector is facing various challenges. This study supports previous reports that diarrhea in calves is one of these challenges [15].Virulence genes are extensively used in determining E. coli pathotypes and have been found in multiple combinations [34,35]. In our study, stx1, stx2, and st VGs were more frequent than eae, hly, aatA, and lt. In another report, however, stx1 and eae were found at a higher rate than st and stx2 [35]. In the isolates where virulence genes were found in combination, stx2-st, sx1-st, stx1-stx2, and lt-st represented the dominant combinations. In addition, stx1-stx2-st and stx2-hly were identified in a low proportion of samples from diarrheic calves. None of the samples were positive for the bfp, ial, and daaE virulence genes, in contrast to a previous report that indicated the presence of these genes in E. coli from diarrheic calves [36]. When translating the VRG combinations into pathotypes, ETEC was detected from 32.3% of the diarrheic calves as a sole pathogenic isolate. This result was in accordance with Dall et al. [37] who described ETEC as the leading cause of severe neonatal calf diarrhea in Brazil. All of the ETEC pathotypes carried the st virulence gene. It has been reported that st encodes structurally stable and highly virulent toxins [38]. Other studies also suggested that ETEC carries multiple chromosomal and plasmid virulence gene combinations that enhance the fitness and transmission between hosts [39,40]. Moreover, a study in Egypt described an st-associated ETEC pathotype in diarrheic calves [41,42]. The present finding and the previous studies indicate that ETEC plays a major role in causing calf diarrhea. However, the results presented in this study contradicted that of Umpiérrez et al. [34] and Khawaskar et al. [43] who reported only 10% and 1.9% ETEC from diarrheic calves in Uruguay and India.In the present study, STEC was the second most frequent pathotype found in 23.2% of diarrheic calves. This is in agreement with previous reports that STEC is a key pathotype in calf diarrhea [2,43]. In addition, STEC has been detected at a higher rate in diarrheic calves [40] and other diarrheic animals than in healthy animals [44]. Several studies [44,45,46] have reported that STEC isolates with stx1 were more common than those that carried stx2, which contradicts the findings in the current study. Similar with the current study, Sobhy et al. [47] showed that the majority of STEC isolates had stx2 instead of stx1 and stx1-stx2 combinations. These inconsistent results indicate that stx1, stx2, and stx1-stx2 play a role in causing diarrhea among calves. In addition, STEC was reported in 34% of non-diarrheic water buffalo calves in Brazil [48] and 5.5% of healthy cattle calves in Korea [35]. This is in line with previous suggestions that animals are important reservoirs of STEC [35,49]. Moreover, the severity of the disease could be associated with the number and type of VGs present, suggesting that an infection by E. coli isolates with more VGs in various possible combinations are more likely to cause a disease. This study revealed that 18.2% of the sampled calves had mixed STEC and ETEC (STEC/ETEC) infections. This result was in line with Awad et al. [41] who reported ETEC/STEC (14.7%) and ETEC/EPEC (2.7%) in diarrheic calves in Egypt. Other researchers showed that diarrheic calves could be infected with a hybrid STEC-ETEC pathotype [50,51]. These pathogroups are characterized as hetero-pathotypes comprised of virulence genes representing two or more E. coli pathotypes [52]. Other studies reported that various hetero-pathotypes, EAEC-STEC and EPEC-STEC, successfully colonized newborn calves and caused persistent diarrheal diseases [53,54]. In the present study, EPEC, EAEC, and EHEC pathotypes were found only in 3, 2, and 1% of diarrheic calves, respectively. These low percentages are in line with the findings from diarrheic neonatal calves sampled in India: EPEC (2.9%), EAEC (0.9%), and EHEC (0.3%) [43]. A study of diarrheic water buffalo diarrheic calves in Brazil detected EPEC (3.4%) and EHEC (3.4%) [48]. However, the 2% EAEC detection rate in our study is lower than the 39% EAEC detection rate in the same study [48]. This difference could be due to differences in the study’s design or target population (the species, health status, and geographical location). In the current study, we attempted to relate the occurrence of pathotypes in the fecal samples to some basic farm and calf management practices. However, these observations must be interpreted with caution as the sampling focused on clinical cases with no healthy controls and was not randomized, which would have been needed for a solid epidemiological analysis. The incidence of E. coli pathotypes was higher in diarrheic calves sampled from medium sized farms. Regardless of the farm’s size, calves ≤ 2 weeks old were more frequently infected than other age groups. This is similar to previous studies that noted that younger calves were more frequently infected with different E. coli pathogroups than older calves [55,56]. This might be due to differences between younger and older calves regarding their immunity and rumen function. In addition, this could be due to age-related changes in the receptor expression for adhesins of the different pathotypes in the intestinal epithelial cells. In the present study, more female and crossbreed diarrheic calves were found to be infected with E. coli pathotypes. There is no plausible explanation for these differences, and the sex skewed finding is inconsistent with a report from India where the detection of STEC (8.0%) and ETEC (2.5%) were higher in samples from male neonatal calves. This was suggested due to the fact that male calves did not receive appropriate care as they were considered economically less important [43]. Our results show that E. coli pathotypes were more common in calves not fed colostrum than in calves fed colostrum. We also noted that calves fed colostrum 6–24 h after birth were more frequently infected by E. coli pathotypes than those fed within 0–6 h. Multiple studies show that the sufficient and timely colostrum feeding of calves improves the growth, immunity, and intestinal microbiota of the calves [57,58], as maternal antibodies in the colostrum prevent a bacterial attachment to the intestinal epithelium [59,60,61]. According to the findings in this study and multiple other studies, the colostrum feeding of calves should takes place within 6 h after their birth. Of the 17 diarrheic calves fed colostrum via hand, 47.1% (p = 0.008) and 11.8% (p = 0.003) were positive for STEC and EAEC, respectively. Similar research in Ethiopia has indicated that hand/bucket feeding was significantly correlated with a high incidence of E. coli [15]. The higher occurrence of pathotypes in samples from hand-fed calves compared with those that suckled colostrum may indicate a lack of appropriate hygiene routines during the hand-feeding of the colostrum to calves. This merits further study.The detection of E. coli pathotypes was slightly higher in calves housed in barns (with other animals) and on soil floors. This result was in agreement with Fesseha et al. [15]. We also found that calves housed with very good hygiene had a higher frequency of E. coli pathotypes; however, those farms were comparatively larger farms, where a higher infectious pressure with E. coli pathotypes could be assumed. The interaction between the farm’s size and hygiene may have prevented the full exploration of these aspects in our limited number of samples.