The impact of the epimural microbiota on ruminant production

Milk composition

Milk composition is associated with rumen fermentation and the available nutrients absorbed through the rumen epithelium (Matthews et al., 2019). Microbial nitrogen metabolism in the rumen is vital in providing microbial protein to the host for milk production (Tadele and Amha, 2015). VFAs also affect milk fat component (Hurtaud et al., 1993). Most of the studies investigating the association between rumen microbiota and milk composition have focused on the liquid and solid phase-associated microbiota (Jami et al., 2014; Bainbridge et al., 2016; Zhang et al., 2017a,b). They have concluded that modulation of the rumen microbiota to achieve a larger proportion of casein or whey relative to non-protein nitrogen (N) in the rumen can eventually improve milk protein yield (Tadele and Amha, 2015).

Some of the epimural bacteria (members of Firmicutes, Proteobacteria, Actinobacteria, and Bacteroidetes) possess urease genes (Jin et al., 2017; Mann et al., 2018), and may play a role in the overall rumen N metabolism. The




Stage of life


Major phylotypes®

Chen et al. (2011)

Beef heifers

Corn: 97% hay, 3% concentrate Trt: transitioning from 60:40, 40:60, 25:75,15:85, 8:92 forage-to- concentrate ratio

8 months old


Bacteria: Proteobacteria, Firmicutes, Bacteroidetes

Petri et al. (201 3)

Angus heifers

Forage: 95% grass hay, 5% supplement

Mixed forage: 60% barley silage, 30% barley grain, 10% supplement High grain: 9% barley silage, 81% barley grain, 10% supplement

BW at

308 ± 35 kg

16S rRNA gene



Bacteria: Firmicutes, Proteobacteria, Bacteroidetes, Actinobacteria

Liu etal.(2015)

Boer x Yangtze River Delta White goats

Hay: 81% Chinese wildrye, 15% Alfa Ifa,

HG: 30% Chinese wildrye, 45% corn meal, 20% wheat meal, 1.1% soybean

2-3 years

16S rRNA gene MiSeq sequencing

Bacteria: Firmicutes, Bacteroidetes, Proteobacteria, Actinobacteria, Spirochaetae

Jiao et al. (2015)


1-40 days: 0.5 Lgoat milk, 0.04 kg DM/meal fresh grass, 0.12kg DM/ meal starter concentrate 40-70 days: 0.06 kg DM fresh grass/ meal, 0.17 kg starter concentrate/ meal

1-70 days

16S rRNA gene MiSeq sequencing

Bacteria: Proteobacteria, Firmicutes, Bacteroidetes, Actinobacteria, Fusoba cteria

Wetzels et al. (2017)

Holstein cows

Corn: 50% grass silage, 50% second- cut meadow hay

Trt: 20% grass silage, 20% second- cut meadow hay, 60% concentrate

BW at

710 ± 118 kg

16S rRNA gene MiSeq sequencing

Bacteria: Proteobacteria, Firmicutes, Bacteroidetes, Synergistetes, Elusimicrobia

Abecia et al. (2014b)


Alfalfa hay ad libitum NAT: remained with dam ART: milk replacer

1,3, 7,14,21, 28 days

16S rRNA pyrosequencing

Bacteria: Proteobacteria, Firmicutes, Bacteroidetes, Cyanobacteria

De Mulder et al. (2017)

Holstein- Friesian cows

70:30 forage-to-concentrate ratio


16S rRNA gene MiSeq sequencing

Bacteria: Firmicutes, Bacteroidetes, Fibrobacteres, Proteobacteria, Spirochaetes, TM7, Actinobacteria Archaea: Methanobacteria

AlZahal et al. (201 7)

Holstein cows

High-forage: 77:23 forage-to- concentrate ratio (w/wo yeast) High-grain: 49:51 forage-to- concentrate ratio (w/wo yeast)


16S rRNA gene MiSeq sequencing

Bacteria: Bacteroidetes, Firmicutes, Fibrobacteres, Proteobacteria, Tenericutes, Cyanobacteria, SR1, Spirochaetes

Mann et al. (201 8)

Holstein cows

Corn: 50% grass silage, 50% second- cut meadow hay

Trt: 60% concentrate, 40% forage

Non-lactating (3-4 parities)


Bacteria: Proteobacteria, Firmicutes, Bacteroidetes, Spirochaetes, and Actinobacteria Archaea: Methanocaldococcus Metha nobrevibacter

Frutos et al. (2018)

Merino lambs

43.3% barley, 15.0% corn, 23.7% soybean meal 44, 15.0% barley straw, 3.0% vitamin-mineral premix

Fattening period


Bacteria: Only reported total bacteria, Prevotella sp., Selenomonas rummantium, and methanogens

Li et al. (2019b)

Holstein bull calves

Corn: texturized, 35.3% starch, 25.3% NDF

Trt: pelleted, 42.7% starch, 15.1% neutral detergent fiber, (NDF)

1-17 weeks


Bacteria: Proteobacteria, Firmicutes, Fusobacteria, Actinobacteria, Bacteroidetes

Lin et al. (2019)

Hu sheep

Corn: breast milk only Trt: fed starter twice daily

10-56 days

rRNA gene MiSeq sequencing; TruSeq metagenome sequencing

Bacteria: Bacteroidetes, Firmicutes, Spirochaetae, Actinobacteria, Proteobacteria, Tenericutes Protozoa: Entodinium, Polyplastron, unclassified Trichostomatia, Ophryoscolex, Diplodinium, Isotricha

Neubauer et al. (2019)

Holstein cows

Corn: 50:50 hay:grass-silage mix Trt: 65% concentrate, 35% roughage


16S rRNA gene MiSeq

Bacteria: Proteobacteria, Firmicutes, Synergistetes, Actinobacteria

* Phylotypes were listed according to their relative abundance in each study; only those with more than 1 % were presented.

distinctive ureolytic bacterial composition and the urease genes observed from the epi mural microbial community may indicate functional segregation from that of lumen microbial community (Jin et al., 2017). Currently, the exact functions of the epimural ureolytic bacteria, their urease activities, and to what extent the epimural microbiota influence N metabolism in the rumen, remain unclear. Advanced omics methods such as metagenomics and metatranscriptomics may help to reveal the functions of the epimural microbiota in overall nitrogen metabolism, but this depends on the accurately isolating epimural microbiota.

In addition to milk protein, milk fat is another economically important trait in milk production. The de novo synthesis of fatty acids in the udder relies on VFAs being transported from the rumen (Palmquist, 2006). VFA absorption and transportation is associated with aspects of rumen wall physiology such as epithelial surfaces and activity of transporters (O'Shea et al., 2016). There is evidence that rumen epithelial bacterial abundance is associated with total ruminal VFA concentration (Chen et al., 2012), suggesting that this community may play a role in regulating VFA absorption. It may be worth exploring the role of active transporters and the epimural microbiome using meta-transcriptomics to identify active transportation pathways for the corresponding VFAs and their relationship with active microbial groups in the rumen wall (Mann et al., 2018).

Feed efficiency

Epimural microbiota have also been reported to be a factor affecting host feed efficiency. Kong (2016) compared the epimural microbiota in beef cattle with different residue feed intake (RFI), a common measurement for cattle feed efficiency (where low RFI indicates higher efficiency). The study found that Campylobacteraceae and Neisseriaceae had significantly greater cell activity in the L-RFI epithelium than that in the H-RFI epithelium (5.8% vs 2.4%, 10.8% vs 1.0%, respectively). It was suggested that the higher oxidase activity enabled by these two phylotypes may lead to higher oxygen scavenging capacity in L-RFI animals, thus maintaining an anaerobic environment for better microbial anaerobic fermentation, resulting in higher energy efficiency. Liang et al. (2017) studied the epimural microbiota in finishing Hu breed lambs and found the proportion of Butyrivibrio fibrisolvens and Escherichia coli was higher in H-RFI animals. The greater abundance of 6. fibrisolvens was reported to be associated with the lower propionate concentration and a greater acetate:propionate ratio, which coincided with lower feed efficiency. It was suggested that the higher E. coli population was the result of the greater fluctuation of the rumen pH and VFA concentrations within the rumen of H-RFI animals (Liang et al., 2017). It has been suggested that the effect of epimural microbiota on host feed efficiency which is unknown may rely on other functions rather than feed degradation. It would be useful to apply metagenome and/or metatranscriptome-based approaches to identify the active pathway(s) to better define the exact mechanisms that may influence host feed efficiency and to what extent the epimural microbiota contribute to cattle feed efficiency.

Methane emissions

Oxygen concentration in the rumen epithelium has been found to be much higher than that in the rumen lumen. A high population of anaerobic archaea has commonly been identified within the epimural microbiota. The Mbb. gottschalkii:Mbb. ruminantium ratio of the planktonic community has been claimed to be positively associated with host methane emissions (Danielsson et al., 2017), but this is not supported by other studies.

Although there is no direct evidence that the epimural microbiota affect methane emissions, there are possible links worth investigating, such as the role of rumen epimural microbiota in regulating lumen [H] availability. If marker epimural microbes are found to be associated with lumen [H] concentration and/or indirectly associated with methanogenesis, we may be able to develop novel methane mitigation methods that focus on manipulating these epimural microbiota in addition to lumen microbiota.

There is indirect evidence of host effect on the archaeal community. In a study of rumen content transplantation, Zhou et al. (2018) found that the archaeal community was unaffected regardless by donor phenotype, and that the archaeal community of most of the animals returned to its original status. They argued that the rumen epimural community may be the innate driving force in re-establishment of microbiota, particularly the archaea. Host effect on methane emissions was reported by Roehe et al. (2016), who ranked the rumen archaeal population and host methane emissions based on host progeny groups. They have also suggested that that rumen archaeal abundance is under host genetic control. Difford et al. (2018) examined the factors related to methane emissions in dairy cows and reported that host genetics was the top factor followed by the bacteria and archaea contributing to this process. The authors have suggested the potential to modulate rumen microbiota by genetically selecting animals. These studies highlight the need to study epimural archaea individually to develop better methane mitigation strategies.

Challenges and future trends

Better illustration of microbial functions

Although recent studies have applied multi-omics-based methods to reveal the composition and functions of the rumen microbiota (e.g. Stewart et al., 2018; Li et al., 2019a), the knowledge of the epimural microbiota is still very limited. Mann et al. (2018) employed a metatranscriptomic approach to study the gene expression and functional potentials of the rumen wall bacteria. They found that the genes involved in galactose, starch, sucrose, and energy metabolism were highly expressed by the epimural microbiota. This result showed the active roles of rumen epimural microbiota in providing hostrelevant metabolites through cross-rumen-wall transportation. For the first time, they have reported nitrogen fixation by the cattle rumen epimural microbiota. High levels of expressions of the genes involved in oxidative stress were also observed. They also identified the active archaeal and fungal community within the epimural microbiome, though theirfunctions are still not understood.

There is no doubt that more descriptive data on the rumen epimural microbiota will become available using multi-omics methods, but it is essential to perform accurate analyses to identify the TRUE and KEY' members of the microbial community. Factors such as sample collection methods (Paz et al., 2016), DNA/RNA extraction protocols (Villegas-Rivera et al., 2013; Henderson et al., 2015), sequencing methods, and bioinformatic modules for data processing (Neves et al., 2017) can all affect the outcome of identifying rumen epimural microbiota. In addition, the database used for data interpretation is also critical. The metagenome being assembled by Seshadri et al. (2018) was retrieved from the Hungate1000 collection, which is highly specialized for the rumen microbiota and claims to have covered approximately three quarters of the genus-level microbial taxa (JGI Hungate Collection). However, both compositional and functional segregation exist between rumen content microbiota and epimural microbiota. It is only possible to define the accurate microbial functions by properly assigning the sequence reads to the true phylotypes, but currently available tools cannot achieve this outcome yet.

Recently, Wilkins et al. (2019) developed a more efficient method for retrieving metagenome-assembled genomes (MAGs). Such methods offer the rumen microbiologist a new direction for defining the epimural MAGs that can be utilized as reference, thus allowing better resolution for future study. On top of identifying microbial composition, studying the metabolome of the microbiota may provide direct evidence on how microbial activities impact host performance. Although there is no metabolite-based study being performed on the epimural microbiota, a proteome-based study (Hart et al., 2018) on the rumen fluid microbiota has provided a direction for future examination of the epimural microbiota in terms of metabolites/enzymes/proteins, allowing us to obtain a complete description of their functions as well as their impact on the host. It should be noted, though, that microbial metabolites are dynamic over time. Whether the metabolome obtained from these samples is representative and sufficient to reflect the real conditions within the rumen environment remains an issue.

Better understanding of host-microbial interaction

Due to their proximity to the host, the epimural microbiota are considered to have close interaction with host animals. Chen et al. (2012) reported that the epimural bacterial population was positively correlated with TLR 4 expression in the rumen wall in acidosis resistant beef steers, but such a relationship was not seen in acidosis-susceptible animals. Liu et al. (2015) found that the expression of TLR2 was associated with ten epimural bacterial taxa while TLR4 was associated with one taxon belonging to Anaerolineaceae. These findings suggest that the abundance of epimural bacteria may stimulate host gene expression. Future studies should explore the association between host gene expression and the epimural microbiota, to better elucidate how microbiota colonization can influence host biology.

Lipopolysaccharide (LPS) is found in the outer layer of gram-negative bacteria (GNB; Wang and Quinn, 2010). The lysis of GNB in the rumen lumen is associated with increasing levels of rumen LPS (Nagaraja et al., 1978; Gozho et al„ 2005, 2007). The LPS released from the epimural GNB is considered as endotoxin with strong pro-inflammatory potential, thus causing rumen tissue lesion (Steele et al., 2011), altering the expression of TLRs and the tight junction structure of the rumen epithelium (Liu et al., 2015; McCann et al„ 2016).

Among these epimural-associated GNB, Fusobacterium necrophorum is one of the best understood species. F. necrophorum is an aerotolerant anaerobe commonly identified from the rumen wall, and has been found to be actively involved in feed degradation, metabolizing lactic acid and epithelial proteins (Li et al., 2019a). Linder normal rumen conditions, F. necrophorum mainly functions as a feed digester. However, in animals suffering from SARA, it becomes an opportunistic pathogen (Berg and Scanlan, 1982), proliferating on the rumen wall where the mucosa are affected by parakeratosis and prolonged SARA conditions in the host (Okada et al., 1999; Takayama et al., 2000). It can also penetrate the rumen epithelium, translocate into the bloodstream, and invade cattle liver, causing abscesses (Nagaraja and Titgemeyer, 2007; Steele et al., 2009; Tadepalli et al., 2009). It is vital to understand the interaction between commensal epimural microbiota and the host, which could provide the basis for future strategies to prevent the pathogenesis of F. necrophorum under high-grain diets and ruminant metabolic dysfunctions.

Host-microbial interactions have been extensively studied in the case of content/liquid microbiota. These interactions help explain the variation in host feed efficiency for beef (Zhou et al., 2009; Hernandez-Sanabria et al., 2012; Myer et al., 2015) and dairy cattle (Jami et al., 2014; Jewell et al., 2015), in host methane emissions (Carberry et al., 2014; Shi etal.,2014)and in milk production (Jami et al., 2014; Jewell et al., 2015). However, the role of the epimural community in explaining host phenotypes is less well understood. It is only recently that Mann et al. (2018) have examined and reported the expression of a wide range of microbial genes by the epimural microbiota. The products by these active genes may serve as the future targets for identifying their roles in influencing epithelium morphology, molecule transportation, signaling, and so forth, which provides direct evidence of microbial-host interactions.

More effective manipulation strategies

There have been many recent attempts to manipulate rumen microbiota to enhance fermentation efficiency and to create a healthier rumen environment. Such attempts to change rumen microbiota during adulthood have not shown consistent, long-term effectiveness. Abecia et al. (2014a,b) have proposed targeting the early-life rumen content microbiota for their long-term effect in reducing methane emission. As the epimural microbiota are exposed to stronger host-related selective pressures and are functionally diverse from the rumen microbes residing in other niches, future studies of host regulatory effects on the epimural microbiota may help to improve the chance for successful early- life interventions. There needs to be a meta-analysis of the existing multi-omics datasets generated from different studies to provide clues for novel targets such as the key phylotypes, key pathways, and key enzymes. We are currently working to properly allocate all the different types of data to a single analysis platform to better interpretthe meta-omesfrom multiple studies and to provide more accurate comparisons among studies. Application of probiotics may also be a future direction for manipulating the epimural microbiota, which requires more experimental evidence to support this approach.

Improved sample collection and data handling

While multi-omics technologies can provide high-resolution analysis of nucleic acids (DNA and RNA) and metabolites within the collected samples, the major roadblock of studying the epimural community is identifying 'TRUE' epimural microbiota. Unlike the easy accessibility of rumen fluid and digesta samples either through tubing or rumen cannula methods, epithelial samples are comparatively difficult to obtain, van Niekerk et al. (2018) have developed a rumen tissue sampling method through endoscopy, which allows repeated sampling without rumen cannulation. This method allows us to collect rumen tissue samples from young and adult ruminants more efficiently and humanely, and also to study changes in the epimural microbiota within the same animals when they receive different treatments.

Removing the non-adherent microbes from the rumen epithelial tissue is also a technical issue for sample handling. An adequate rinse step to remove any non-adherent materials from the epithelium samples is necessary (Chen et al., 2011). However, since the half-lives of microbial mRNAs can be short as seconds (Laalami et al., 2014), the rapid yet efficient way to remove the nonadherent microbes should be explored to ensure the integrity of the samples.

Contaminants such as reads identified from the blank control in commercial kits (Becker et al., 2016; Thoendel et al., 2017) and host reads identified from meta-omics data (e.g. Brown et al., 2019) should also be considered when interpreting the data to avoid false-positive results. Properly selecting the reaction kits as well as the data processing pipeline may help to limitthe impact of contamination.

< Prev   CONTENTS   Source   Next >