The growing interest in understanding the complex bovine udder microbiome has resulted in several published studies in recent years. These studies aimed to uncover how this microbiome influences the udder health and possibly has an important role during mastitis. In this study, we contribute to this knowledge by exploring the milk microbiota with a cross-sectional study of the milk microbiota as elucidated from over 400 quarter milk samples obtained from 60 lactating cows.
Sampling of udder milk for microbiome analysis is difficult and challenging15,16. In order to accomplish a more representative overview of the milk microbiota that colonize the upper interior part of the udder, the sampling method used in this study was different to those previously used to study milk microbiota2,6. The sampling was performed after the cow was regularly milked with the intent to remove microbial contaminations from the teat apex and to avoid the sampling of milk present in the cistern which might contain bacteria able to enter the udder between milking. The complete removal of contaminations from the teat apex or the environment cannot be ensured by the results obtained in this study, and additional experiments are needed to compare microbiota from milk samples obtained pre- and post-milking. However, although some taxa associated with the environment, such as Bacillaceae and Pseudomonadaceae, were detected in the dataset their abundance was much lower (< 1%) than that found in milk collected from bulk farm tanks which included organisms present in cows as well as milk equipment and the farm environment17,18,19. Interestingly, by using this sampling regime, we obtained a microbiota that was highly dominated by two families (Corynebacteriaceae and Staphylococcaceae). Several other taxa commonly associated with mastitis and as part of the milk microbiota were also detected. These included Aerococcaceae, families within the order Clostridiales, Streptococcaceae and Enterococcaceae. The sampling regime used in the study resulted in a similar or lower bacterial richness in our samples compared to previous studies20,21,22. However, comparisons with other studies cannot be conclusive due to different sampling procedures, filters and bioinformatics pipelines used between the studies. In particular, the method used for SV inferences increases the richness detected of microbiota studies compared to open-reference OTU clustering methods23.
The most common taxa found in the quarter milk samples are frequently reported in previous milk microbiota studies2,20. The family Corynebacteriaceae, which includes only the genus Corynebacterium, has been previously described as part of the skin-associated microbiome in humans and it is also found in other bovine body sites, such as the teat canal and uterus24,25,26,27. Therefore, it was not unexpected to find a high abundance and frequency of Corynebacteriaceae in quarter milk samples. This was similar to previous reports of milk microbiota2,6. Surprisingly, this taxon was found in almost all the quarter milk samples from two different farms and was detected in all the 60 cows. This suggests that Corynebacterium is part of the udder core microbiota revealed in this study. The role of Corynebacterium within the udder is still unknown and this might include a protective role against mastitis pathogens. Competition for niche adaptation between Corynebacterium and other taxa and inhibition of pathogen bacteria has been previously reported in microbiome studies28,29. Of particular interest was the interaction between Corynebacterium and Staphylococcus. Here, we found that the two main taxa of this study were negatively correlated with regards to their abundance. It is not possible to know if this negative correlation arose from the increase in abundance of one of the two taxa in the quarter or the decrease of one of them because of the lack of information of the absolute number of bacteria. However, competition between Corynebacterium and Staphylococcus has been shown previously in different environments and the negative correlation found in this study is further confirmation28,30,31. Another indication of a possible protective role of Corynebacterium against dysbiosis can be drawn from the analysis of the composition of the Corynebacterium population itself. The Corynebacterium population was different in dysbiotic quarters and this was shown in several cows. In most of these cases, this dysbiosis was caused by the presence of mastitis pathogens, such as Streptococcaceae and Enterococcaceae, which might outcompete and replace the dominant Corynebacterium SVs found in the other quarters of the same cow.
Several species of Corynebacterium inhabit the udder microbiome and Hoque et al.11 identified 12 of them using a metagenomics approach. In this study, the most common Corynebacterium SVs were assigned to Corynebacterium bovis, but several other species were detected by amplicon sequencing and MALDI-TOF identification of isolates. Corynebacterium bovis was previously described as part of the teat canal microbiota and Hiitio et al.12 also detected this species in the upper part of the gland. Corynebacterium bovis is classified as a minor mastitis pathogen and usually targeted during routine testing by mastitis laboratories. In our study, we found that C. bovis was present in a large number of quarter milk samples and several isolates were also obtained by culturing. By using standard culturing methods, Goncalves et al.32 detected C. bovis in only 15.8% of composite milk samples and using sequencing methods Taponen et al.2 detected that C. bovis covered between 50 and 73% of all Corynebacterium reads in milk samples that were PCR positive for C. bovis. The presence of C. bovis in milk has also been linked to an increase number of somatic cells in the cow’s milk and to the changed milk composition in subclinical mastitis32.
The genus Staphylococcus contains one of the most frequent and well-known mastitis pathogens, Staph. aureus. The samples analyzed in this study were obtained from cows with no signs of clinical mastitis and this pathogen was infrequently found in our samples and was not the most abundant Staphylococcus species isolated. On the other hand, the non-aureus staphylococci (NAS), which included several species of skin-related and the minor mastitis pathogens, were the most abundant group within the Staphylococcaceae family. Similar to Corynebacterium, NAS are commonly found in milk microbiota and in other microbiomes such as the bovine teat canal and skin12,33.
High correlations between the incidence of several families of the order Clostridiales were detected. These families accounted for more than 10% of the total microbiota in this study and are known to inhabit the cow rumen34,35. Their high positive correlation might indicate that their presence in the milk microbiota was connected. Rumen-related families within the Clostridiales order are often detected in milk microbiota studies. Vasquez et al.5 found that members of the Clostridiales such as Ruminococcaceae and Lachnospiraceae were among the dominant taxa in milk microbiota. However, the presence of these families in milk might originate from contamination during sampling procedure from the farm environment, the teat canal or from the translocation of cow microbiota through the hypothesized endogenous entero-mammary pathway3,36. Another aspect might be that families of the order Clostridiales are able to colonize the interior of the udder due to favorable anaerobic conditions and presence of nutrients. None of the over 1000 isolates obtained were classified within the order Clostridiales. This could be due to their fastidious growth requirements as this order includes obligative anaerobe spore-forming genera.
In this study, we collected milk samples from each quarter and a considerable difference in microbiota was found for samples collected from the same cow. This was detected in several individuals. Despite factors pertaining to the individual cow, such as immunity and genetics, that are known to influence the udder microbiota37, our results indicated that within-individual microbial variation is frequent. This is supported by the fact that certain taxa were often detected in only one quarter, rather than in all four quarters. Examples of these are found in this study where, in particular, the presence of Streptococcaceae and Enterococcaceae was detected in some quarters of the same cow but not in all four.
The milk microbiota structure was found to be influenced by farm and sampling period and several of the most abundant taxa were found to be influenced by these factors. The reasons for this variation in composition of quarter microbiota between the two farms studied may be mainly due to different farming practices, animal variability and genetics. Farming practices, such as bedding material, milking systems, housing system have previously shown to have an impact on the milk microbiota38,39. Whether or not the farming practices used in both farms in this study (e. g. cows regularly milked in an automatic milking system, open barn environment) influenced the microbiota inside the udder cannot be fully confirmed. However, since hygienic conditions have been found to be important for the transmission of mastitis pathogens within the herd and it can therefore be hypothesized that taxa able to colonize the udder can be spread within the herds, thus increasing the differences in diversity between farms40.
The variation in microbiota between sampling periods can be attributed both to the season when the sampling occurred and to the days of lactation of the animals. Interestingly, Streptococcaceae were significantly more abundant in the second sampling which was performed between late Aug-Sep. Species belonging to the Streptococcaceae family, such as Strep. dysgalactiae, Strep. uberis and Strep. agalactiae are common mastitis pathogens. Previous studies showed that cows have a greater risk of Streptococcus mastitis during the summer months compared to the winter months41,42. Therefore, it is not unexpected that this taxon was more abundant in the second sampling period compared to the first one. The presence of Streptococcus in milk samples was confirmed by the identification of the isolates.
The term dysbiosis can be defined as an alteration of the normal microbiota due to loss of diversity, reduction of commensal microbes and usually presence of pathogens which outcompete the existing microbiota. Several studies showed that disruption of diversity in udder microbiota is linked to a change in the health status of the udder and, in case of mastitis, the microbiota is dominated by the pathogen11,12. An important challenge for defining dysbiosis is the definition of a “normal” or “baseline” microbiota. Here, we decided to compare the microbial composition within the same individual and sampling to identify the “normal” microbiota and the presence of dysbiosis in only one quarter. By using this approach, we were therefore not able to identify dysbiosis that might occur in more than one quarter. However, we identified a high prevalence of dysbiosis in the bovine udder and a good correlation between the detected dysbiosis and isolation of bacterial species (within the same taxonomy) identified by maldi-TOF.
While the data presented here further support that a microbiota is present in the udder and might play a protective role, a few limitations of this study need to be considered. The somatic cell counts (SCC), used as indicator for the udder health, were not recorded at the time of sampling and indications of the health status of the quarters was obtained with culturing methods, which are biased by the conditions used for culturing and might therefore not reflect the true status. However, samples were collected from lactating cows, milked during regular farm practices with no sign of clinical mastitis during the collection days. Most microbiota studies suffer from the lack of absolute number of bacteria and results are presented in relative abundance where each feature is constrained to each other. Changes in abundance of one taxon will therefore impact the relative abundance of the others. The microbiota study presented here was performed using short-read sequencing, a method which achieves a lower resolution during taxonomical assignation of the reads. Furthermore, milk from cows with no sign of clinical mastitis harvest a low abundance microbial biomass and in case of an introduction of one taxon into the udder, this might have a big impact on the abundance of the resident microbiota of the udder. The cow breed has previously shown to be an important factor shaping the milk microbiota22. In this study, we selected 60 cows of Norwegian Red breed and therefore our results need to take into consideration possible breed-to-breed diversity when compared to other studies where more popular breeds are used (e. g. Holstein or Friesian). The microbiota of milk from the udder of Norwegian Red cows has not previously been reported.
Source: Ecology - nature.com
