Isolation of Bacterial causes of Respiratory Infections in Calves in Smallholder farms in and around Gonder Town, Ethiopia

Background: Bovine respiratory disease (BRD) is considered as the major cause of severe respiratory tract infections in calves. Pasteurellosis is a multifactorial respiratory disease, which mainly affect calves within four weeks of weaning. A cross-sectional study was conducted from October 2017 to April 2018 in and around Gondar town, Amhara Regional State, North West of Ethiopia. The aim of the study was to isolate Mannheimia and Pasteurella species from calves up to six months old, and to assess the associated risk factors with the occurrence of respiratory disease. Sex, age (< 16 weeks and > 16 weeks), body condition status (poor, medium, good), breed (local and cross breed), livelihood (mixed crop and urban), farming systems (semi intensive and intensive), herd size (small medium, and large), maternity pens (present or absent), and method of colostrum feedings (hand bucket and suckling) were the examined risk factors. Results: A total of 84 nasopharyngeal swab samples were collected from calves with any signs of illness related to pasteurellosis. The overall isolation rate of the respiratory pathogens was 64/84 (76.2%) (95% CI=65.7-84.8), with 46.4% of Mannheimia haemolytica and 28.8% Pasteurella multocida isolates. The distribution of pathogens was statistically higher (P< 0.001) in calves with respiratory problems (93.6%; 95% CI= 82.5-98.7) compared to those with no symptoms of respiratory illness (54.1%; 95% CI= 36.9-70.5). Among the examined risk factors age, sex, breed, farming system were found to be potential risk factors and signicantly associated with Pasteurella infection of calves (p<0.05). The higher isolation rate of Mannheimia haemolytica indicated that it is the major cause of respiratory disease in the study area. Conclusion: The present nding revealed that pasteurellosis is one of the major diseases of calves in the study area in which M. haemolytica and P. multocida were found to be commonly involved in respiratory infections. Improved


Background
Bovine respiratory disease (BRD) is considered as the major cause of severe respiratory tract infections in calves [1]. It causes great economic losses such as reduced average daily gain, feed e ciency, and overall performance of beef calves and nally calves morbidity and mortality [2,3]. Several infectious agents are commonly isolated from the respiratory tracts of clinically sick and healthy animals [4]. Many factors can weaken the host's immune system and/or damage the lining of the respiratory tract to such an extent that these pathogens are able to progress deeper into the respiratory tract and cause disease [5][6].
Calf pneumonia refers to infectious respiratory disease in calves. It is primarily a problem in calves less than 6 months old with peak occurrence from 2-10 weeks, but it may be seen in calves up to 1 year of age [7]. Pasteurellosis is a multifactorial respiratory disease [6,8], which mainly affect calves within four weeks of weaning. Infected calves manifest the clinical signs when they are sorted and sold to different farms. The bacteria that causes pasteurellosis are part of the normal microbiota in the upper respiratory tract [9], making the disease di cult to prevent. mannheimia haemolytica and pasteurella multocida are most commonly associated with pneumonia in cattle calves [10,11]. These pathogens can easily spread between animals, especially when calves are crowded (as in shipment) or closely con ned (as in a dairy calf nursery) [12].
In practice, deep nasopharyngeal swabs (DNS) [13], transtracheal aspiration (TTA) [14], and broncho alveolar lavage (BAL) [15] have been used for sampling the respiratory tract. Deep nasopharyngeal swab is the easiest, fastest, and cheapest technique and therefore most suitable for sampling large numbers of animals [16].
In Ethiopia, calf pneumonia has been reported as the second most disease syndrome associated with calf morbidity and mortality next to calf diarrhea [17,18]. Even though bovine pneumonic pasteurellosis is one of the most economically important infectious diseases in Ethiopia [19], there is limited data on the status of the disease and potential risk factors. This is particularly important in the study area, because dairy farms are an integral part of the economy. Hence, the study was conducted to identify the main bacterial respiratory pathogens and to assess major risk factors in bovine calves managed under dairy farms of Gondar town.

Study area
Sample collection was held in and around Gondar town, which is the capital of central Gondar administration zone, Amhara regional state. The area is located in the north west of Ethiopia at 740 km from Addis Ababa, capital city of the country. It is situated between 12 o 36'N and 33 o 28'E at an altitude of about 2300 m above mean sea level with an average temperature of 20 o c and an average annual rain fall of 1800 mm.

Study Population
The study population constitutes bovine calves up to 6 months of age that shows any respiratory sign or having general illness (suspected to have respiratory pathogens) found in dairy farms located at urban as well as peri-urban, which is characterized by mixed crop livestock production system. Study animals are selected regardless of breed, sex, body condition, herd size, and weaning status.

Study Design and Animal Selection
A cross-sectional study was employed to isolate bacterial pathogens causing bovine respiratory disease of calves and assess its major risk factors from November 2017 to April 2018. Purposive type of sampling was applied based on the availability of calves and willingness of farmers to take the nasopharyngeal swabs for bacteriological culture of Pasteurella species. Thus, a total of 84 calves were included in the study.

Sampling Procedures
Animals were restrained by an assistant and then the external part of the nose was disinfected with 70% alcohol. After the alcohol has evaporated, a sterile cotton-tipped swab was inserted in to the nostril and rotated against the wall of the nasal cavity as described by [20]. The swab was then placed in a labeled sterile test tube containing 3 ml of tryptose soya broth. Samples were then kept in a box containing ice for transport to University of Gondar veterinary microbiology laboratory.

Isolation and identi cation of Pasteurella species
The isolation and identi cation was performed as described by [21]. Brie y, the specimen in tryptose soya broth was incubated for 24 hours at 37°c and a loop full of the broth culture was streak on petri-dish of blood agar base supplemented with 5% sheep blood then incubated aerobically at 37 °C for 24 hours. Then colonies showing typical gram's reaction and cellular morphology were further sub-culture on both blood and MacConkey agar plates. The general appearance of colonies, presence and nature of hemolysis, and the ability to ferment lactose was recorded. The suspected pure colonies from both blood and MacConkey agars were transferred onto nutrient agar slants for further identi cation by biochemical tests. M. haemolytica was characterized as able to produce a narrow zone of hemolysis on blood agar, able to grow on MacConkey agar, but unable to produce indole, whereas P. multocida was characterized as unable to produce hemolysis on blood agar, unable to grow on MacConkey agar, and able to produce indole as described by [22].

Data management and Statistical Analysis
Data was stored in Microsoft excel sheet for handling. After checking for its correctness, it was transferred to SPSS version 20.0 for analysis. Descriptive analysis, such as frequency and percentage were used to describe the proportion of calves affected by bacteria. Chi-square and regression analysis were employed to establish the association between cultured bacteria result and risk factors. Before regression analysis, the data was cheeked for ful llments of assumptions, such as correlation of each variables (not more than 0.7), correlation of independent variables with dependent variable (minimum of 0.3), and multi-collinearity tests (VIF (> 10) and Tolerance (< 0.1)). Thus, Age, sex, and breed were used in the nal model. The OR was computed at 95% CI and in all cases the difference between parameters were tested for signi cance at probability level of less than 0.05 (P ≤ 0.05).

Prevalence of Pasteurella species and Risk Factors
The result showed that M. haemolytica and P. multocida varied in proportion with different factors examined. Thus, sex and health status were signi cant associated (P < 0.05) with the occurrence of M. haemolytica, in that it was higher in male (56.8%) than female calves (35.0%) and animals with signs of respiratory illness (57.4%) than healthy one (32.4%). Moreover, the prevalence was relatively higher in calves of < 16 weeks old (50.9%) than older calves (37.0%) ( Table 5). With regard to P. multocida, the occurrence was higher in cross bred (33.3%) than local (22.2%) and calves with signs respiratory illness (36.2%) than healthy one (21.6%) ( Table 6).

Cultural and Biochemical Characteristics of Isolated Pasteurella Species
Mannheimia haemolytica were able to grow as small red colony on MacConkey agar and show βhemolysis on blood agar, while Pasteurella multocida were unable to grow on MacConkey agar and nonhemolytic (Table 7), but had mucoid colony and gram negative on gram staining. Different biochemical tests performed to identify the isolated pathogen and all the isolates were positive for oxidase, catalase, nitrate and phosphate and able to ferment sucrose, glucose and mannose. P. multocida isolates were positive for indole test but unable to ferment lactose and maltose (Table 7).

Discussion
In the present study the overall isolation rate of pasteurellosis was found to be 76.2% (95% CI = 65.7-84.8). The current nding in the prevalence of pasteurellosis is relatively higher than reports of previous studies in the country. Thus, [23], [24], [25] [26], and [27], who 50.2, 40.8, 39.2, 13, and 8.7%, respectively. This might be due to the difference in sampling that this study was conducted on cases and suspected of pneumonia, study area, time of sampling and farm management [23].
Analysis on sex related susceptibility showed that pasteurellosis was higher in male (84.1%) than female (67.5%) calves, in that the odd of being positive was 1.64 times higher in the former (OR = 1.64; 95% CI = 0.52-5.2). The possible explanation for sex related susceptibility is that less colostral immunoglobulin absorbed in male than female during neonatal life which leads to disease in male calves [28]. It is also worth mentioning that male calves are not as valuable to the dairy operation as females and therefore may not receive an attention the heifers do have, possibly accounting for the higher infection in males [29,30]. Meanwhile, the prevalence was signi cantly higher in younger calves (< 16 weeks) (82.5%) than older one (≥ 16 weeks) (63.0%) with younger calves have 3.1 times the chances of being affected than older (OR = 3.1; 95% CI = 1.0-9.6; p = 0.04). The possible explanation for the age-related susceptibility might be due to failure of passive immunity in hand feeding practices of young calves being able to predispose for bacterial infection and other predisposing etiological agents [21]. For instance, calves less than one month of age lack su cient postruminal digestive enzymes to break down most sugars and are limited in their ability to utilize starch, maltose, sucrose, or dextran [31]. Partial or complete failure of passive transfer of maternal antibodies is an important host factor related to development of pneumonia in young calves [7,32].There was signi cant association (p < 0.05) between pasteurellosis and breed of calves, with higher prevalence in cross breed (82.5%) than local (63%) calves. This deference might be linked to variation in environment adaptability. Therefore, local breed has high disease resistant capability. Similarly, the difference in the isolation rate of the two breeds might be due to the difference in feed access of the calves. Local breeds feed relatively less in amount and quality than cross breed calves which consume much amount and quality feed. So local breeds being less exposed to infection and therefore, have lower isolation rate than cross breed [33].
This study compares the level of isolation rate between animals kept under intensive and semi intensive management systems. The infection rate was higher in calves kept under intensive system (80%) compared to semi intensive one (70.6%). An intensive management system is mainly associated with con nements and predisposing calves are frequently contact to accumulations of urine and other wastes. This situation is likely to favors the spread of Pasteurella species among animals [29]. Similar observation was reported previously by [34; 35]. Moreover, the occurrence of pasteurellosis varied signi cantly (p = 0.035) among livelihood, in that it was higher in urban (82.8%) than mixed crop (61.5%) production system.
The overall proportion of Mannheimia haemolytica and Pasteurella multocida was 46.4% and 28.8%, respectively, indicating M. haemolytica was the major causative agent involved in calve pneumonic pasteurellosis in the study area. Although the infection rate varies, this nding is consistent with previous reports of [23]. Though Mannheimia haemolytica is a normal ora of the upper respiratory tract, suppressors of the host immune system favor the multiplication of Pasteurella species, leading to bronchopneumonia in purely pneumonic animals [36].The reasons for increased susceptibility to M. haemolytica infection in stressed animals are primarily attributed to the breakdown of innate pulmonary immune barriers by stressors [37,38]. Although the percentage of isolation was relatively low (28.8%) the possible role of P. multocida in the etiology and pathogenesis of calve pneumonia should not be under estimated [23].
Concerning the isolation rate of Pasteurella species in relation to the health status, it was higher in animals with respiratory illness (93.6%) than no respiratory illness (54.1%), among which Mannheimia haemolytica and Pasteurella multocida were isolated from 57.4% and 36.2% of calves with respiratory disease symptoms, while they were isolated from 32.4% and 21.6% of calves with no respiratory illness, respectively. In this study signi cant variation was observed in prevalence of Pasteurella species in calves with and without respiratory illness. This suggests the possible involvement of these bacteria in the genesis of pneumonia.

Conclusion
This present nding revealed that pasteurellosis is one of the major diseases of calves in the study area in which M. haemolytica and P. multocida were found to be commonly involved in respiratory infections. Age, breed, sex, and livelihood were the risk factors associated with occurrence of respiratory infection.
The study focuses on M. haemolytica and P. multocida. Therefore, further epidemiological investigation is needed to rule out the role of concurrent microbial causes of calf pneumonia, so as to design appropriate control and preventive measures at farm as well as national level.
Abbreviations BAL: brocho alveolar lavage; BCS: body condition scores; BRD: bovine respiratory disease; CI: con dence interval; DNS: deep nasopharyngeal swabs; LG: logistic regression; MCF: methods of colostrum feeding; TTA: transtracheal aspiration Declarations Ethics approval and consent to participate The investigators treated animals with kindness and took proper care by minimizing discomfort, distress or pain. Basically, the study was conducted by taking swab samples from the nasal cavity of animals. In addition, there were no human subjects and no private data was taken except history related with management of animals. Nevertheless, Verbal consent was obtained from calves owners for inclusion of their calves in the study. Con dentiality of the data obtained was strictly followed in the study periods as well as they were told that the data is not to be used for any other purpose that was not intention of the study. Finally, ethical clearance was obtained from Haramaya University ethical review board.

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analyzed during the study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that none of them have nancial or personal relationships with individuals or organizations that may have inappropriately in uenced them in writing this paper and, therefore, declare that there is no competing interest.

Funding
No funding was obtained for this study Author contributions TM participated in the study design, eld work and conducted the work. PW, AA and TF designed the study; provision test material, analyzed the statistics, performed data management and interpretation, prepared and wrote the manuscript. All authors read, evaluated and approved the nal manuscript.

Acknowledgments
The researchers would like to acknowledge University of Gondor for the provision of valuable resource and laboratory facility and technical supports during the study period. We would also like to acknowledge farm owners who are highly appreciated for their tremendous cooperation and allowing us to take samples.  Figure 1 Map showing the Study area (Q-GIS version 3.10, http://osgeo4w-oslandia.com)