Metabolic health, the metabolome and reproduction in female cattle: a review

Abstract

Studies over the past 30 years have confirmed the important role of metabolic hormones and metabolic substrates in reproductive function in female cattle. The emergence of metabolomics is providing a deeper understanding of the role of specific metabolites, and clusters of metabolites, in reproduction and also health and disease. Dairy cows undergo major fluctuations in metabolic health and metabolomics is helping to better understand the changes in metabolite profiles associated with negative energy balance and ketosis. New knowledge that emerges from this work should lead to improved nutritional management of dairy cows. In reproduction, it is now possible to gain a metabolomic signature of ovarian follicular fluid and of developing embryos. This should likewise lead to improvements in both natural and assisted reproduction in cattle. Systems biology integrates genomics, transcriptomics, proteomics and metabolomics, and contributes to gaining an understanding of complex biological networks.

Highlights

• Metabolic hormones and metabolic substrates have a major influence on reproduction in female cattle.

• Negative energy balance and ketosis are associated with changes in the systemic and liver metabolome in dairy cows.

• The metabolome of ovarian follicular fluid influences oocyte quality and embryo development.

• Systems biology integrates genomics, transcriptomics, proteomics and metabolomics, and provides a deeper understanding of complex biological networks.

Keywords:

• Metabolic health
• metabolome
• reproduction
• female cattle

Introduction

Studies over the past 30 years have demonstrated the fundamental importance of metabolic hormones (GH, IGF1, insulin, leptin, thyroxine) and metabolic factors (glucose, fatty acids) in female cattle reproduction (Lucy 2000, 2008; Velazquez et al. 2008; Silva et al. 2009; Roche et al. 2011; Castro et al. 2012; Samadi et al. 2013, 2014; Lucy et al. 2014; Sartori et al. 2016; Meikle et al. 2018; D'Occhio et al. 2019). Insulin-like growth factor-I (Ghanipoor-Samami et al. 2018) has received particular attention given its central roles in ovarian follicular growth (Wandji et al. 1992; Spicer and Echternkamp 1995; Yuan et al. 1998; Lucy 2000, 2008, 2011; Spicer and Aad 2007) and function of the corpus luteum (Lucy et al. 1999; Woad et al. 2000). Follicles can produce IGF1 (Adashi 1998; Yuan et al. 1998) but an important source is IGF1 secreted by the liver and sequestered from blood by follicles (Clemmons and Underwood 1991; Lucy 2011; Zhang, Wu et al. 2013). Insulin-like growth factor-I production by the liver is influenced by nutrition and metabolic status (Clemmons and Underwood 1991; Guggeri et al. 2014). Dysfunction of the liver, due to metabolic stress or liver disease, is associated with reduced IGF1 (Fenwick et al. 2008) and a disruption of ovarian folliculogenesis in cattle (O’Doherty et al. 2014; Mokhtari et al. 2016). Metabolic health and liver health are therefore critical for normal ovarian function and fertility in cattle.

The relationships between body condition, metabolic status, liver function and reproduction, have been studied extensively in periparturient and early lactation dairy cows (Roche et al. 2007, 2009, 2011, 2013, 2018; LeBlanc 2012; Sundstrum 2015; Overton et al. 2017). In the period before calving, there is an increase in glucose demand for foetal growth and after calving glucose is required for milk production. This is often accompanied by a decrease in appetite which can lead to negative energy balance (NEB) and lowered amounts of glucose in blood (Shaw 1956; Lucy et al. 2014; www.farmhealthonline.com/US/disease-management/cattle-diseases/ketosis/). The rumen converts carbohydrates in feed to volatile fatty acids that include propionate, acetate and butyrate. Propionate is used by the liver to produce glucose that is used as a source of energy for oxidative processes in the liver. If carbohydrate intake and propionate levels are low due to decreased appetite, the liver can switch to energy production by utilising non-esterified fatty acids (NEFAs) that diffuse into blood from fat because of the NEB (Adewuyi et al. 2005). NEFAs are oxidised by the liver to acetyl-CoA. Acetyl-CoA can undergo complete oxidation through the tricarboxylic acid cycle (TCA), converted to very-low density lipoprotein (VLDL) and exported, or converted to triglycerides (TAG) and stored by the liver (White 2015). The liver has a limited capacity to metabolise acetyl-CoA and if complete oxidation is not possible then acetyl-CoA is converted to ketone bodies that are exported as acetone, acetoacetate and beta-hydroxybutyrate (Aschenbach et al. 2010). The latter causes the metabolic disease of ketosis (acetonemia) in dairy cows. Common features of ketosis are low blood concentrations of glucose, insulin and IGFI, with elevated NEFAs and ketone bodies (Walsh et al. 2007; LeBlanc 2010; Bisinotto et al. 2012; Esposito et al. 2014). At the same time, excess TAG is stored by the liver giving rise to ‘fatty liver’ syndrome. This metabolic picture is typically associated with anoestrus (Bisinotto et al. 2012). Major disruption of glucose homeostasis before calving can cause a clinical condition in cows similar to pregnancy toxaemia in smaller ruminants (Marteniuk and Herdt 1988; Rook 2000). Ketosis and pregnancy toxaemia are uncommon in beef cows.

Metabolomics

The study of relationships between metabolic condition and reproduction in cattle entered a new era with the advent of metabolomics (Ryan et al. 2013; Canovas et al. 2014; Fontanesi 2016; Goldansaz et al. 2017; Li et al. 2017). Metabolomics is the study of the metabolome which comprises the myriad of low molecular weight metabolites (lipids, amino acids, vitamins) that influence cellular, tissue and organ function (Patti et al. 2012; Dona et al. 2014; Goldansaz et al. 2017). The metabolome is downstream of the genome, transcriptome and proteome, and is considered the closet ‘omic’ to the phenotype (Figure 1) (Fontanesi 2016; Pantophlet et al. 2017). The latter has led to the suggestion that the metabolome is a particularly important indicator of changes in biological function. Another significant feature of the blood metabolome is that it represents the integration of external (e.g. diet) and internal (e.g. genotype) factors that influence metabolism (Figure 2). Hence, the metabolome acts as an integrator of endogenous and exogenous processes to shape the phenotype. This includes the reproductive phenotype. The following sections provide a synthesis of how metabolomics is leading to a deeper understanding of relationships between metabolic condition and reproduction in cattle.

Figure 1. Diagram illustrating the positioning of the metabolome downstream of the genome and positioned closely to the phenome. The myriad of low molecular weight metabolites that comprise the blood metabolome is derived from endogenous processes and exogenously from the diet and activity of the gut microbiome. The environment and microbiome can also impact the genome through mutations or epigenetic effects.

Figure 2. Diagram illustrating the contribution of the ‘exogenous metabolome’ and ‘endogenous metabolome’ to the systemic and tissue metabolome.

Ruminal metabolome

As illustrated in Figure 2, feed consumed by cattle impacts the ruminal microbial metabolome (Saleem et al. 2013; Khiaosa-ard and Zebeli 2014) which ultimately influences the systemic metabolome. Holstein dairy cows fed either corn stover or a mixture of alfalfa hay and corn silage showed differences in ruminal fluid levels of key metabolic factors such as acetate, glucose and propionate (Zhao et al. 2014). In a second study in Holstein dairy cows, low concentrate and high concentrate diets produced differences in the ruminal amino acid profile that included alanine, leucine and glycine (Zhang, Zhu et al. 2017). Dairy cows fed a high grain diet early in lactation tend to have a higher incidence of metabolic disorders. It was found that Holstein cows on a total mixed ration (TMR) with a high grain diet (30 or 45% barley on a dry matter basis) from day 60 of lactation had an altered ruminal microbiome compared with cows fed a low grain diet (15% barley), and the former cows also had differences in ruminal levels of short-chain fatty acids and a range of amino acids (Saleem et al. 2012). In several studies, high grain diets (TMR and 30 or 45% barley on a dry matter basis) increased the ruminal levels of potentially toxic and inflammatory compounds such as putrescine and methylated amines in lactating Holstein cows (Ametaj et al. 2010; Saleem et al. 2012). Lactating Simmental cows with subacute ruminal acidosis from a grain-rich diet had an altered systemic metabolome response to intramammary challenge with lipopolysaccharide (Aditya et al. 2018). However, this was not observed in a second study (Humer et al. 2018). Lactating Friesian cows that grazed ryegrass and white clover pastures had greater ruminal concentrations of isoacids compared with contemporary cows that were fed a TMR based on maize silage (O’Callaghan et al. 2018). Isoacids are branched-chain fatty acids (isobutyric, isovaleric) and straight-chain valeric acid, which serve as nutrients for ruminal cellulolytic bacteria (Andries et al. 1987). The ruminal metabolite profile of Holstein-Friesian cows differed to Hanwoo steers independent of diet, and this was interpreted to suggested that genotype can have a predominant influence on the ruminal metabolome in cattle (Lee et al. 2012).

Diet influences the milk metabolome in dairy cows in addition to the systemic metabolome (Sun et al. 2015a, 2015b). In a recent study, the systemic metabolome in Holstein cows was predictive of the milk protein profile (Wu et al. 2018). However, an earlier study reported that milk was a distinct metabolic compartment and had a metabolite composition different to blood (Maher et al. 2013). Further studies are required to ascertain relationships between the blood and milk metabolomes. This is important as accurate indices are required to determine the metabolite profile of milk in relation to human nutrition and health. In one study, the ruminal metabolome was related to feed efficiency in Angus crossbred steers (Artegoitia et al. 2017). Similar to ruminants, diet influenced the systemic metabolome in a monogastric species, the pig (Sun et al. 2015a, 2015b).

The above presents only a summary of changes in the ruminal and systemic microbiomes in response to diet, and in all studies the complete picture is far more complex. This underscores the necessity for advanced metabolomics analysis platforms in order to understand the biology of very large metabolite datasets (Dona et al 2014).

Metabolome and ketosis

The nutritional management of dairy cows to prevent or mitigate ketosis is important for production and animal welfare (Littledike et al. 1981; Miettinen and Setala 1993; Lu et al. 2013). The ability to use metabolomics to identify cows susceptible to ketosis is a preferred strategy, compared with the application of metabolomics as a diagnostic technology (Kenez et al. 2016; Ceciliani et al. 2018). In Holstein-Friesian cows, a milk glycerophosphocholine/phosphocholine ratio of ≥ 2.5 early in lactation was associated with a very low risk of developing ketosis (Klein et al. 2012). Plasma metabolomic profiling was able to distinguish between lactating clinically normal cows, cows with subclinical ketosis, and cows with clinical ketosis (Zhang, Davis et al. 2013; Li et al. 2014; Sun et al. 2014). Hepatic lipidosis could also be diagnosed using the plasma metabolome in transition Holstein-Friesian and Red-Holstein cows (Imhasly et al. 2014) and early lactation Holstein-Friesian and Simmental cows (Humer et al. 2016). The systemic metabolome was able to identify Holstein dairy cows at-risk for a retained placenta (Dervishi et al. 2018) and metritis (Zhang, Deng et al. 2017). A relatively large study with lactating Danish Holstein-Friesian and Jersey cows revealed an association between the milk metabolome and somatic cell count (Sundekilde et al. 2013). Other studies have shown changes in the systemic metabolome of Holstein-Friesian calves after experimental infection (De Buck et al. 2014) and vaccination (Gray et al. 2015).

Transcriptome and negative energy balance

Whilst the metabolome is the focus of this review, it is relevant to consider the transcriptome within the context of metabolic health in cattle. Using RNA-seq technology, major differences in liver RNA linked to fatty acid metabolism were found between mild NEB and severe NEB in lactating Holstein-Friesian cows (McCabe et al. 2012). Differences were also reported for liver micro RNA (miRNA) expression related to NEB severity in lactating Holstein-Friesian cows (Fatima et al. 2014a, 2014b). In lactating Lacaune ewes, the systemic transcriptome was altered by NEB (Bouvier-Muller et al. 2017). The incorporation of transcription factors and miRNAs in an integrated computational approach yielded information on gene networks involved in NEB in cattle (Mozduri et al. 2018).

The science of systems biology (Woelders et al. 2011) brings together transcriptomic, proteomic and metabolomic information to build integrated gene networks that underpin biological processes linked to phenotypes (Homuth et al. 2012; Krumsiek et al. 2012; Widmann et al. 2013, 2015; Cho et al. 2014; Wang et al. 2019). Application of this approach in dairy cows has been reviewed in relation to nutrition (Loor et al. 2013), lactation (Li et al. 2017), breeding (Berry et al. 2011) and fertility (Ceciliani et al. 2017). The integration of ‘omics’ technologies has the potential to provide new knowledge that informs changes in management that enhance both production and wellbeing in dairy cattle and other livestock (Crowe et al. 2018).

Metabolome and reproduction

Ovarian follicles

Ovarian function in mammals is acutely sensitive to metabolic homeostasis, and the important role of the GH-IGF1 axis was noted above. It is now emerging that the metabolome, both systemic and follicular, influences follicle growth, oocyte quality and embryo developmental competency (Singh and Sinclair 2007; Nel-Themaat and Nagy 2011; Collado-Fernandez et al. 2012; Wallace et al. 2012; Bertoldo et al. 2013; Gerard et al. 2015; Gu et al. 2015; Krisher et al. 2015). Follicular fluid provides a metabolomic micro-environment that supports oocyte growth and development (Gerard et al. 2002; Pinero-Sagredo et al. 2010; Sirard 2011; Hennet and Combelles 2012; Leroy et al. 2012; Dumesic et al. 2015; El-Hayek and Clarke 2016; Guerreiro et al. 2018). In a study that utilised abattoir cow ovaries, palmitic acid and total fatty acids were reduced, and linoleic acid increased, in follicular fluid of follicles that contained competent oocytes (Matoba et al. 2014). Differences in follicular fluid concentrations of saturated fatty acids between Holstein-Friesian heifers and lactating cows were associated with differences in fertility (Bender et al. 2010). Predominantly, Holstein cows with either a positive or negative estimated progeny difference (EPD) for fertility had differences in follicular fluid content of saturated fatty acids, mono-unsaturated fatty acids and poly-unsaturated fatty acids (Moore et al. 2017). Lactating Holstein-Friesian cows had different profiles of amino acids and fatty acids in follicular fluid compared with non-lactating cows and heifers (Forde et al. 2016). Follicular fluid influences oocyte development through the cumulus layer (Zhang et al. 1995) and the metabolome profile of cumulus undergoes changes during follicular growth in cattle (Uhde et al. 2018). These studies are providing new insight into the metabolite environment of follicles that is optimal for oocyte development and should lead to targeted nutritional strategies that enhance fertility in cattle.

Assisted reproduction

Metabolomics has also been combined with assisted reproduction in cattle. Metabolomic analysis of spent culture media of IVF produced cattle embryos was able to distinguish between male and female embryos (Uyar et al. 2013; Gomez et al. 2016). The accuracy in predicting sex using spent culture media of bovine IVF embryos increased from early blastocysts (74%) to expanded blastocysts (86%) (Munoz et al. 2014b). Metabolomics, proteomics and miRNA have also been applied to assess stage of embryo development and embryo quality (Al Naib et al. 2009; Rodgaard et al. 2015). In a recent study, IVF and ICSI derived cattle embryos were associated with spent culture media with a different metabolomic signature (Li et al. 2018). Dual assessment of the systemic metabolome of recipient cows, together with the metabolome of spent culture media, could predict the pregnancy outcome for transferred IVF embryos (Munoz et al. 2014a) and conventional superovulated embryos (Munoz et al. 2014c). Notwithstanding recent progress with application of metabolomics in IVF, limitations have been identified and the field remains at an early stage (Cuperlovic-Culf et al. 2010; McRae et al. 2013; Uyar and Seli 2014; Munoz et al. 2014c; Krisher et al. 2015; Rodgaard et al. 2015).

Conclusions and future direction

The important role of metabolic hormones and metabolic substrates in health and reproductive function in female cattle has been clearly established over the past 30 years. The advent of metabolomics has brought this field into a new phase that will deliver an unprecedented increase in knowledge of the role of individual metabolites, and networks of metabolites, in reproduction. Metabolomics will similarly allow major advances in understanding of the biology of metabolic health and metabolic disease. Maintaining livestock in good metabolic health is important for production and reproduction, and it is also vital for animal wellbeing.

For the future, metabolomics will be integrated together with genomics, transcriptomics and proteomics into a systems biology framework (Dumas 2012). The unique strength of metabolomics is that biochemical pathways and networks are largely known, which means that information on the metabolome can guide gene discovery and also provide information on gene function (Gauguier 2016). Parallel study of the metabolomes of the rumen microbiome and host will provide new knowledge on the impact of the environment on metabolic health and disease, and gene function (Aguiar-Pulido et al. 2016; Scharen et al. 2018).

Acknowledgements

The authors acknowledge with gratitude the many postgraduates who contributed to research and thinking in our laboratories that is included in this review.

Disclosure statement

All authors declare there is no conflict of interest.