Novel Metabolite Biomarkers for Early Detection of Fatty Liver Disease in Dairy Cows

Fatty liver disease is one of the most common metabolic disorders affecting dairy cows during the periparturient and early postpartum period. It primarily develops due to negative energy balance, excessive mobilization of body fat reserves, and intense puerperal stress, which together severely compromise subsequent milk production and future reproductive performance. Within the first month after calving, approximately 5–10% of dairy cows develop severe fatty liver disease, while 30–40% exhibit mild to moderate hepatic fat infiltration. Notably, the first two weeks postpartum represent the period of greatest susceptibility to metabolic disorders^1,2,3. 

From a clinical perspective, fatty liver disease rarely occurs as a standalone condition. It is frequently complicated by infectious diseases associated with immunodeficiency as well as other metabolic and reproductive disorders, leading to increased culling rates during the perinatal period. Consequently, fatty liver disease contributes to substantial economic losses in the dairy industry, estimated to reach tens of millions of dollars annually^1,4,5. 

Challenges in Current Diagnostic Approaches 

In production settings, fatty liver disease is commonly identified using ultrasonographic imaging and traditional serum biochemical indicators such as aspartate aminotransferase (AST), glucose (GLU), insulin (INS), and non-esterified fatty acids (NEFA). However, these approaches are limited by low diagnostic sensitivity and high false-negative rates, particularly in subclinical cases. At present, liver biopsy remains the only reliable diagnostic method; however, its application in commercial dairy farms is impractical due to the need for specialized training, increased risk of infection, and additional animal stress¹. Moreover, invasive sampling contradicts animal welfare principles and may aggravate the condition of already compromised cows. 

Despite the high prevalence and impact of fatty liver disease, there is still a lack of effective, minimally invasive diagnostic tools that can be routinely applied under field conditions¹. 

Metabolomics as a Strategy for Early and Non-Invasive Detection 

In recent years, circulating metabolites (serum or plasma) and terminal metabolites (milk, urine, feces) have gained attention as potential biomarkers for screening metabolic disorders such as ketosis, retained placenta, metritis, and lameness in dairy cattle, while also offering insight into metabolic adaptations during the transition period^1,4,6,7,8. However, only a limited number of molecular markers have been identified specifically for fatty liver disease. 

Because hepatic fat accumulation often precedes other metabolic disorders, the early identification of fatty liver disease is critical. It has therefore been hypothesized that inherent genetic and metabolic regulatory mechanisms govern the development of fatty liver disease in dairy cows. The primary objectives of this study were to identify novel, non-invasive metabolite biomarker panels with improved diagnostic sensitivity and specificity and to elucidate the common biological pathways underlying fatty liver disease during the parturition period. To achieve this, circulating metabolites from serum and terminal metabolites from urine and feces were analyzed using gas chromatography–mass spectrometry (GC–MS)¹. 

Metabolic disorders such as fatty liver disease and ketosis remain as prevalent today as they were two decades ago¹. Accordingly, metabolomics approaches have been increasingly applied to identify metabolite panels capable of distinguishing healthy cows from those affected by metabolic disorders, including ketosis, retained placenta, metritis, lameness, mastitis, and displaced abomasum^1,5,6,7,8,9. Using this approach, non-targeted metabolomics enabled the identification of predictive and diagnostic biomarker panels for fatty liver disease, along with the metabolic pathways involved in disease onset and progression¹. 

Identification and Validation of Novel Metabolite Biomarker Panels 

Metabolomics is defined as the comprehensive quantitative analysis of all detectable metabolites, particularly small-molecule compounds, within a biological sample, providing a global overview of metabolic status and disease-related alterations. Nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS) are the most commonly used analytical platforms in metabolomics research^1,7,9. Compared with NMR, MS offers superior sensitivity and selectivity, with detection limits in the femtomolar to attomolar range. Among MS-based techniques, GC–MS is widely used due to its high throughput, strong reproducibility, reliable quantitative accuracy, and detailed structural information¹. 

Building on these principles, GC-TOFMS-based metabolomics was applied to serum, urine, and feces. A rigorous, multi-step experimental design was implemented to ensure the reliability of identified biomarkers. First, liver biopsy–confirmed cows were included as a discovery set to identify candidate biomarkers. Second, highly suspected cows with serum biochemical profiles consistent with biopsy-confirmed cases were used as a test set for further screening. Third, systematic multivariate and univariate statistical analyses were applied to refine biomarker selection. Differential expression between healthy and diseased cows was confirmed using violin plot analyses. Finally, receiver operating characteristic (ROC) curves were generated to evaluate diagnostic performance¹. These biomarkers demonstrated superior diagnostic potential compared to traditional indicators, supporting their utility in early detection. 

Beyond their high diagnostic accuracy, these biomarkers offer practical advantages, including non-invasiveness, rapid detection, and suitability for use with feces and urine samples. These features align well with animal welfare considerations and the goals of sustainable dairy production. 

Dysregulated Fatty Acid and Amino Acid Metabolism in Fatty Liver Disease 

Metabolic profiling revealed that fatty liver disease is characterized by increased fatty acid levels and decreased amino acid levels across serum, urine, and feces¹.  

Elevated circulating non-esterified fatty acids are closely associated with fatty liver disease during the perinatal period and primarily result from negative energy balance. As lactation progresses postpartum, glucose demand increases, often leading to insufficient energy supply and enhanced mobilization of adipose tissue. Excessive NEFAs enter the liver, where limited oxidative capacity results in triglyceride accumulation due to reduced export efficiency in cattle. Consequently, elevated circulating fatty acids serve as a key indicator of fatty liver syndrome ¹. 

In parallel, reduced amino acid levels were associated with hepatic dysfunction. For example, L-α-aminobutyric acid, involved in cysteine and methionine metabolism, was significantly reduced in diseased cows, suggesting increased hepatic steatosis. Previous studies have shown that disruptions in methionine metabolism can lead to fatty liver and other liver diseases. Additionally, nitrotyrosine, a marker of oxidative stress and mitochondrial damage, was associated with liver injury through its role in inhibiting respiratory enzymes and promoting lipid peroxidation¹. Together, these findings indicate that enhanced fatty acid mobilization and impaired metabolic capacity are central pathological features of fatty liver disease. 

Common Biological Pathways Underlying Disease Pathogenesis 

Pathway enrichment analysis revealed that fatty liver disease is associated with widespread metabolic reprogramming involving fatty acid, amino acid, and bile acid metabolism. Key pathways included biosynthesis of unsaturated fatty acids, primary bile acid biosynthesis, and branched-chain amino acid degradation and biosynthesis. Collectively, these pathways reflect perturbations in energy metabolism, oxidative stress regulation, and inflammatory responses¹. 

Amino acid metabolism pathways, such as arginine and proline metabolism, valine, leucine, and isoleucine metabolism, and glutathione metabolism, were consistently enriched across serum, urine, and feces. These pathways are closely linked to gluconeogenesis, adipogenesis, and TCA cycle activity, and their dysregulation contributes to abnormal ketone body production and elevated aminotransferase levels, indicating liver injury. Glutathione metabolism, in particular, was disrupted across multiple biological matrices, reinforcing its role in oxidative stress and hepatic dysfunction¹. 

Translational Potential and Broader Implications 

Fatty liver disease in dairy cows shares several metabolic features with non-alcoholic fatty liver disease (NAFLD) in humans, including elevated fatty acids, altered insulin signaling, increased aminotransferase activity, and hepatic lipid accumulation. Although differences exist in lipid origin and insulin resistance, dairy cows may serve as a valuable animal model for studying NAFLD pathogenesis. Given the growing global burden of NAFLD, insights gained from bovine metabolomics may help advance understanding of liver disease mechanisms across species^{1,10,11,12,13}. 

Conclusion 

This multi-channel metabolomics study demonstrates that non-targeted GC–MS analysis of serum, urine, and feces can identify robust, non-invasive biomarker panels for the early detection of fatty liver disease in dairy cows. The identified biomarkers not only outperform traditional diagnostic indicators but also provide mechanistic insight into disrupted energy, lipid, and amino acid metabolism during the transition period. These findings support the integration of metabolomics-based screening into herd health programs, offering a practical and welfare-friendly approach to improving metabolic disease management in modern dairy production systems. 

References 

1. Zhang X, Liu T, Hou X, Hu C, Zhang L, Wang S, Zhang Q, Shi K. Multi-channel metabolomics analysis identifies novel metabolite biomarkers for the early detection of fatty liver disease in dairy cows. Cells. 2022 Sep 15;11(18):2883. https://www.mdpi.com/2073-4409/11/18/2883 

2. Shi, K.R.; Li, R.R.; Xu, Z.J.; Zhang, Q. Identification of crucial genetic factors, such as PPARγ, that regulate the pathogenesis of fatty liver disease in dairy cows is imperative for the sustainable development of dairy industry. Animals 2020, 10, 639. https://www.mdpi.com/2076-2615/10/4/639 

3. Shi, K.R.; Niu, F.; Zhang, Q.; Ning, C.; Yue, S.J.; Hu, C.Z.; Xu, Z.J.; Wang, S.X.; Li, R.R.; Wang, Z.H. Identification of whole-genome significant single nucleotide polymorphisms in candidate genes associated with serum biochemical traits in chinese holstein cattle. Front. Genet. 2020, 11, 163. https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2020.00163/pdf 

4. Zhang, L.T.; Liu, T.J.; Hu, C.Z.; Zhang, X.; Zhang, Q.; Shi, K.R. Proteome analysis identified proteins associated with mitochondrial function and inflammation activation crucially regulating the pathogenesis of fatty liver disease. BMC Genom. 2021, 22, 640. https://link.springer.com/content/pdf/10.1186/s12864-021-07950-2.pdf 

5. Shen, Y.; Chen, L.; Yang, W.; Wang, X. Exploration of serum sensitive biomarkers of fatty liver in dairy cows. Sci. Rep. 2018, 8, 13574. https://www.nature.com/articles/s41598-018-31845-0.pdf 

6. Eckel, E.F.; Zhang, G.; Dervishi, E.; Zwierzchowski, G.; Mandal, R.; Wishart, D.S.; Ametaj, B.N. Urinary metabolomics fingerprinting around parturition identifies metabolites that differentiate lame dairy cows from healthy ones. Animal 2020, 14, 2138–2149. https://www.sciencedirect.com/science/article/pii/S1751731120001172 

7. Zhang, G.; Zwierzchowski, G.; Mandal, R.; Wishart, D.S.; Ametaj, B.N. Serum metabolomics identifies metabolite panels that differentiate lame dairy cows from healthy ones. Metabolomics 2020, 16, 73. https://link.springer.com/content/pdf/10.1007/s11306-020-01693-z.pdf 

8. Zhang, L.T.; Hu, C.Z.; Zhang, X.; Zhang, Q.; Yan, Z.G.; Wei, Q.; Wang, S.-X.; Xu, Z.-J.; Li, R.-R.; Liu, T.-J.; et al. Protein acetylation in mitochondria plays critical functions in the pathogenesis of fatty liver disease. BMC Genom. 2020, 21, 435. https://link.springer.com/content/pdf/10.1186/s12864-020-06837-y.pdf 

9. Zhang, G.; Tobolski, D.; Zwierzchowski, G.; Mandal, R.; Wishart, D.S.; Ametaj, B.N. A targeted serum metabolomics GC-MS approach identifies predictive blood biomarkers for retained placenta in Holstein dairy cows. Metabolites 2021, 11, 633. https://www.mdpi.com/2218-1989/11/9/633 

10. Nasr, P.; Ignatova, S.; Kechagias, S.; Ekstedt, M. Natural history of nonalcoholic fatty liver disease: A prospective follow-up study with serial biopsies. Hepatol. Commun. 2018, 2, 199–210. https://aasldpubs.onlinelibrary.wiley.com/doi/pdf/10.1002/hep4.1134 

11. Younossi, Z.M.; Loomba, R.; Rinella, M.E.; Bugianesi, E.; Marchesini, G.; Neuschwander-Tetri, B.A.; Serfaty, L.; Negro, F.; Caldwell, S.H.; Ratziu, V.; et al. Current and future therapeutic regimens for nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. Hepatology 2018, 68, 361–371. https://onlinelibrary.wiley.com/doi/pdf/10.1002/hep.29724 

12. Zhou, F.; Zhou, J.H.; Wang, W.X.; Zhang, X.J.; Ji, Y.X.; Zhang, P.; She, Z.G.; Zhu, L.; Cai, J.; Li, H. Unexpected rapid increase in the burden of NAFLD in China from 2008 to 2018: A systematic review and meta-analysis. Hepatology 2019, 70, 1119–1133. https://onlinelibrary.wiley.com/doi/pdf/10.1002/hep.30702 
13. Caussy, C.; Hsu, C.; Lo, M.T.; Liu, A.; Bettencourt, R.; Ajmera, V.H.; Bassirian, S.; Hooker, J.; Sy, E.; Richards, L.; et al. Link between gut-microbiome derived metabolite and shared gene-effects with hepatic steatosis and fibrosis in NAFLD. Hepatology 2018, 68, 918–932. https://onlinelibrary.wiley.com/doi/pdf/10.1002/hep.30702