Introduction

Bovine respiratory disease (BRD) is multietiological disease complex associated with a wide range of infectious disease pathogens in conjunction with alterations in environmental, climatic, and management practices (Griffin et al. 2010; Castro et al. 2021; Confer and Ayalew 2018; Rice et al. 2007). Animals with clinical BRD may terminate in some form of pneumonia if not adequately medicated. The clinical and pathological consequences of BRD frequently result in elevated cattle morbidity, mortality, and lethality worldwide. Viral disease agents frequently associated with the development of BRD include bovine viral diarrhea virus (BVDV), bovine respiratory syncytial virus (BRSV), bovine parainfluenza virus (BPIV3), bovine alphaherpesvirus 1 (BoAHV1), and bovine coronavirus (BCoV) (Castro et al. 2021; Fulton et al. 2009; Zhou et al. 2023). Additionally, there are accumulating evidence to suggest that the Macavirus, ovine gammaherpesvirus 2 (OvGHV2), the cause of sheep-associated malignant catarrhal fever, can serve as a viral pathogen for respiratory disease (Headley et al. 2023a, b; Frucchi et al. 2024). Bacterial agents frequently identified in BRD are Histophilus somni, Pasteurella multocida, Mannheimia haemolytica, and Mycoplasmopsis bovis (formerly, Mycoplasma bovis) (Griffin et al. 2010; Castro et al. 2021; Zhou et al. 2023).

Mycoplasmas are microorganisms within the genus Mycoplasmopsis, class Mollicutes, family Metamycoplasmataceae (Gupta et al. 2018). Most members of the Mycoplasmopsis genus are animal pathogens (Gupta et al. 2018), with M. bovis probably being the most widely studied organism within this genus. While the role of M. bovisin the development of calf pneumonia is well established (Radostits et al. 2017), the exact participation of Mycoplasmopsis bovirhinis (formerly, Mycoplasma bovirhinis) in the etiopathogenesis of respiratory disease in cattle remains obscure (Frucchi et al. 2024). Nevertheless, several studies have suggested that this organism may be associated with the development of pulmonary disease in cattle (Quinn et al. 2011; Radostits et al. 2017). In contrast, M. bovirhinis was detected in pulmonary sections of cattle without any associated disease pattern (Thomas et al. 2002), and is considered as an opportunistic organism of ruminants (Quinn et al. 2011; Radostits et al. 2017).

Our research group has demonstrated that M. bovirhinis was the most frequently identified organism in nasal samples from asymptomatic (55.3%; 57/103) and symptomatic (57.8%; 59/102) dairy calves from 15 dairy herds that were simultaneously infected by BCoV, OvGHV2, H. somni, and P. multocida (Frucchi et al. 2024). Similar studies have identified concomitant infections of M. bovirhinis with other recognized organisms of BRD within the bronchoalveolar fluids (Angen et al. 2009; Thomas et al. 2002), as well as in the nasal (McDaneld et al. 2022) and pulmonary (Zhou et al. 2023) samples of cattle. However, the authors have not identified any published study, within major English and Latin databases, that investigated the possible effects of nasal shedding of M. bovirhinis in calves with BRD. Accordingly, the objectives of this study were to investigate and predict the possible effects of concomitant infections due to BCoV, OvGHV2, H. somni, and P. multocida on the odds of nasal shedding of M. bovirhinis in dairy calves with BRD.

Materials and methods

Study location, animals, and sampling

This study was done with dairy calves from three farms located in two milk producing regions of southern Brazil. All farms were medium-sized milking producing dairy herds located within the municipalities of Arapoti (Farms #1 and 2) and Clevelândia (Farm #3), Paraná state, Brazil (Table 1). These three farms were selected for this investigation from a previous study (Frucchi et al. 2024) since they contained the most dynamic distribution of simultaneous and singular infectious agents of BRD in dairy calves as well as BRD negative dairy calves.

Table 1 Distribution of infectious disease agents diagnosed in diseased and asymptomatic dairy calves from the municipalities of Arapoti and Clevelândia, Paraná state, southern Brazil
Full size table

Nasal samples (NS) were collected from calves (n = 87) using commercially produced synthetic nylon swabs (16 cm) that were inserted into the ventral meatus of the nostril and were derived from a previous investigation of simultaneous infections due to M. bovirhinis and other agents of BRD in calves from 15 dairy herds (Frucchi et al. 2024). In that study (Frucchi et al. 2024), calves were characterized as BRD positive due to the manifestations of at least four clinical signs, including fever (> 40 °C), cough, nasal and/or ocular discharge, and dyspnea (Ferraro et al. 2021). In contrast, dairy calves without any of these clinical signs were classified as BRD negative (Frucchi et al. 2024).

Molecular detection and sequencing of BRD-associated pathogens from nasal samples of dairy calves

The extracted nucleic acids from the NS were used in molecular assays designed to amplify specific genes of the principal agents associated with BRD, including M. bovirhinis, H. somni, P. multocida, M. haemolytica, M. bovis, BVDV, BoAHV1, BRSV, BCoV, BPIV3, and OvGHV2, as described (Frucchi et al. 2024). The specific target genes, desired base pairs of the amplicon, primers sequences, and the related reference can be consulted (Supplementary Table 1). Positive controls consisted of nucleic acids of these organisms derived from previous studies (Headley et al. 2023a, b). Sterile, ultrapure water (Invitrogen Life Technologies, Carlsbad, CA, USA) served as negative control. Both controls were included in all molecular assays. All PCR products were separated by electrophoresis in 2% agarose gels, stained with ethidium bromide, and examined under ultraviolet light. The sequence confirmation of the infectious agents detected in these molecular assays was done as previously described (Frucchi et al. 2024), using standard molecular diagnostic procedures.

Statistical analyses

The Epiinfo program (version 7.2.3.1) generated all frequency tables and was used to determine a possible statistically significant association between the molecular detection of disease pathogens of BRD and the occurrence of clinical manifestations. The Yates-corrected Chi-square or Fisher´s exact test was used with a significance level of p < 0.05. In addition, when appropriate the obtained results were interpreted by using descriptive statistical analyses.

Statistical model designed to determine the risk of Mycoplasmopsis bovirhinis shedding in nasal samples of dairy calves

Input data for the statistical models were derived from a previous investigation (Frucchi et al. 2024) and are summarized in Table 1. The variables associated with individual risk of shedding M. bovirhinis were explored using a 2-step approach. Firstly, univariable analyses were performed using univariable logistic regression, during which unconditional odds ratio and associated 95% confidence intervals were noted. Then, in a second stage, variables with P-values < 0.2 were retained for the multivariable analysis. A logistic regression analysis using M. bovirhinis positivity as a dichotomous variable was used. A manual backward stepwise method was used until all variables remain with P-values < 0.05. All analyses were performed with the R software ( R Core Team 2021). The effect plots obtained from the multivariable model were obtained using sjPlot R package. The discrimination of the model was determined using the area under the receiver operating characteristics (AUC). Interpretation of AUC values were based on the most commonly reported benchmarks with 0.5–0.6 as poor, 0.6–0.7 as low, 0.7–0.8 as moderate, 0.8–0.9 as good, and 0.9-1.0 as excellent discrimination (de Hond et al. 2022).

Results

Infection dynamics observed in dairy calves

The distribution of the infectious disease agents diagnosed in dairy calves at the three farms is presented in Table 1. Although no significant statistical difference was observed between the occurrence of infectious disease pathogens between BRD-diseased and asymptomatic dairy calves (Table 1), BCoV (37.84%; 14/37) and P. multocida (37.84%; 14/37) were the most frequently diagnosed pathogens among all BRD-diseased calves. In contrast, OvGHV2 (85.71%; 6/7) and BCoV (65.38%; 17/26) were the most frequently diagnosed pathogens among asymptomatic dairy calves. Additionally, H. somni was only diagnosed in BRD-diseased calves from Farm #3. Moreover, specific nucleic acids of M. haemolytica, M. bovis, BVDV, BoAHV1, BRSV, and BPIV3 were not detected.

Effects of simultaneous infections on Mycoplasmopsis bovirhinis nasal shedding in dairy calves

The results of the prediction for the shedding of M. bovirhinis in dairy calves are provided in Table 2. This multivariable model demonstrated that two simultaneous infections in dairy calves had direct effects on the nasal shedding of M. bovirhinis: OvGHV2 and M. haemolytica. The likelihood of dairy calves being infected by M. bovirhinis was 2.59 (95%CI: 1.02–6.756) times more elevated in calves that were positive for OvGHV2 as compared with calves that were negative for OvGHV2. Additionally, the odds of dairy calves being infected by M. bovirhinis were 3.46 (1.30-9.636) times higher for calves that were M. haemolytica positive, as opposed to calves not infected by M. haemolytica.

Table 2 Univariable analyses with a multivariable model to predict the occurrence of infectious respiratory disease pathogens in BRD-asymptomatic and symptomatic dairy calves, using Mycoplasmopsis bovirhinis as the dependent variable
Full size table

These relationships are more easily appreciated in the effect plot graphs generated from the multivariable model (Fig. 1). The probability of dairy calves to shed M. bovirhinis varied from 30 to 75% when being simultaneously infected with M. haemolytica (A), while the probability of shedding M. bovirhinis in NS varied from 25 to 75% in dairy calves concomitantly infected by OvGHV2 (B). Furthermore, the multivariable model has demonstrated that simultaneous infections by H. somni, P. multocida, and BCOV demonstrated no positive effect on the nasal shedding of M. bovirhinis in dairy calves. However, the discrimination of the multivariable model was low (AUC = 0.689, 95%CI:0.580–0.799).

Fig. 1
figure 1

Graphical representation of the probability of dairy calves to be infected by Mycoplasmopsis bovirhinis due to concomitant infections by Mannheimia haemolytica (A) and Ovine gammaherpesvirus 2 (B)

Full size image

Discussion

Dairy calves infected by OvGHV2 and M. haemolytica were more likely to shed M. bovirhinis in nasal secretions

The results of the current study have demonstrated that the multivariable model used herein could be useful to predict the nasal shedding of M. bovirhinis in dairy calves. As far as the authors are knowledgeable, this is probably the first study to design a model to investigate nasal shedding of M. bovirhinis. Consequently, these findings are novel and important as they demonstrate that the probability of dairy calves to shed M. bovirhinis was increased in calves that were concomitant infected by M. haemolytica and OvGHV2. Therefore, it seems that infections by M. haemolytica and OvGHV2 may be possible risk factors for the occurrence of M. bovirhinis in dairy calves. In contrast, simultaneous infections by H. somni, P. multocida, and BCoV seem not to have any effect on the nasal shedding of M. bovirhinis in these dairy calves. However, the discrimination of the model was relatively poor, which means that on an individual basis we could not rely on the identification of M. haemolytica and OvGHV2 to accurately predict the M. bovirhinis status.

These initial results are however interesting for the dairy industry particularly in Brazil, and possibly in other countries, where subclinical infections by OvGHV2 are being diagnosed (Headley et al. 2022; Xavier et al. 2023; O'Toole et al. 2002). This is because most commercially available vaccines are efficient in protecting cattle against infections due to M. haemolytica, P. multocida, H somni, and BCoV. While there is no vaccine currently available for the control and/or mitigation of infections induced by OvGHV2. The local scenario relative to infections by OvGHV2 in dairy cattle is even worsened, since the results of a serological assay done within a geographical region of elevated milk production in Brazil has demonstrated seropositivity to this pathogen in 7.9% (29/367) of all cows and 37.2% (16/43) of all dairy farms evaluated (manuscript in preparation).

M. haemolytica is a well-known bacterial agent associated with the development of BRD in dairy and feedlot cattle worldwide (Griffin et al. 2010; Castro et al. 2021; Confer and Ayalew 2018; Rice et al. 2007). Unlike OvGHV2, M. haemolytica is a commensal organism of the respiratory tracts of ruminants (Confer and Ayalew 2018; Rice et al. 2007), and requires stress-induced conditions, such as alterations in the immune system of the host, viral infections, adverse environmental and climatic changes, to become pathogenic (Rice et al. 2007). In addition, this bacterium has a wide range of virulence factors that facilitate infection in the affected host (Confer and Ayalew 2018; Rice et al. 2007). However, as indicated above there are numerous immunization therapies to prevent this bacterial infection as well as efficient antibiotics to treat infected animals (Confer and Ayalew 2018; Rice et al. 2007).

No association was observed between the occurrence of the pathogens in BRD-positive and asymptomatic calves

As was expected, no significant statistical association was observed between the occurrence of the five pathogens identified in BRD-positive and asymptomatic dairy calves, considering that these data originated from a previous study with similar results (Frucchi et al. 2024). Although BCoV and P. multocida were the most frequently diagnosed pathogens of BRD in the dairy calves evaluated, these two pathogens seem to demonstrate no influence on the nasal shedding of M. bovirhinis. Additionally, since the infection dynamics associated with the simultaneous occurrence of these five pathogens were thoroughly described (Frucchi et al. 2024), the discussion above was based on the effects of simultaneous infections on the nasal shedding of M. bovirhinis.

Study limitations

Although these results are interesting in demonstrating the complex associations between infectious disease agents in the development of respiratory disease in dairy calves, a more robust model would have obtained more consistent results. Consequently, the concomitant identification of other members of the Mycoplasmopsis genus, such as M. bovis, in these dairy calves as well as the utilization of larger data set could probably have provided more stability to the model with a more robust AUC values. Nonetheless, these initial findings demonstrated the importance of disease modeling in veterinary medicine.

Conclusions

The multivariable statistical model demonstrated that concomitant respiratory infections of OvGHV2 and M. haemolytica had a positive effect on the nasal shedding of M. bovirhinis in dairy calves with BRD. In contrast, simultaneous respiratory infections with BCoV and P. multocida had no effect on the shedding of M. bovirhinis.