Predicting subacute ruminal acidosis from milk mid-infrared estimated fatty acids and machine learning on Canadian commercial dairy herds.

Our objective was to validate the possibility of detecting SARA from milk Fourier transform mid-infrared spectroscopy estimated fatty acids (FA) and machine learning. Subacute ruminal acidosis is a common condition in modern commercial dairy herds for which the diagnostic remains challenging due to its symptoms often being subtle, nonexclusive, and not immediately apparent. This observational study aimed at evaluating the possibility of predicting SARA by developing machine learning models to be applied to farm data and to provide an estimated portrait of SARA prevalence in commercial dairy herds. A first data set composed of 488 milk samples of 67 cows (initial DIM = 8.5 ± 6.18; mean ± SD) from 7 commercial dairy farms and their corresponding SARA classification (SARA+ if rumen pH <6.0 for 300 min, else SARA-) was used for the development of machine learning models. Three sets of predictive variables: i) milk major components (MMC), ii) milk FA (MFA), and iii) MMC combined with MFA (MMCFA) were submitted to 3 different algorithms, namely Elastic net (EN), Extreme gradient boosting (XGB), and Partial least squares (PLS), and evaluated using 3 different scenarios of cross-validation. Accuracy, sensitivity, and specificity of the resulting 27 models were analyzed using a linear mixed model. Model performance was not significantly affected by the choice of algorithm. Model performance was improved by including fatty acids estimations (MFA and MMCFA as opposed to MMC alone). Based on these results, one model was selected (algorithm: EN; predictive variables: MMCFA; 60.4, 65.4, and 55.3% of accuracy, sensitivity, and specificity, respectively) and applied to a large data set comprising the first test-day record (milk major components and FA within the first 70 DIM of 211,972 Holstein cows (219,503 samples) collected from 3001 commercial dairy herds. Based on this analysis, the within-herd SARA prevalence of commercial farms was estimated at 6.6 ± 5.29% ranging from 0 to 38.3%. A subsequent linear mixed model was built to investigate the herd-level factors associated to higher within-herd SARA prevalence. Milking system, proportion of primiparous cows, herd size and seasons were all herd-level factors affecting SARA prevalence. Furthermore, milk production was positively, and milk fat yield negatively associated with SARA prevalence. Due to their moderate levels of accuracy, the SARA prediction models developed in our study, using data from continuous pH measurements on commercial farms, are not suitable for diagnostic purpose. However, these models can provide valuable information at the herd level.

Relationship between farm management strategies, reticuloruminal pH variations, and risks of subacute ruminal acidosis.

Low reticuloruminal pH (rpH), often observed in subacute ruminal acidosis (SARA), may negatively affect rumen health and animal performance. To investigate the variability of rpH and the prevalence of SARA on commercial farms, we conducted an observational study on 110 early-lactation Holstein cows of different parities from 12 farms selected to cover a broad range of farm management strategies. The rpH of each cow was continuously monitored for 50 d using wireless boluses. To study the effects of animal and farm management characteristics on rpH, we used a multivariable mixed model analysis with the animal and farm as random effects. Automatic milking system and presence of corn silage in the ration were associated with a decrease in rpH of 0.37 and 0.20 pH units, respectively, whereas monensin supplementation was associated with an increase of 0.27 pH units. The rpH increased by 0.15 pH units during the first 60 d in milk. We defined a SARA-positive day as rpH below 5.8 (SARA5.8) or 6.0 (SARA6.0) for at least 300 min for 1 d. Using those definitions, during our study, a total of 38 (35%) and 65 (59%) cows experienced at least one episode of SARA5.8 and SARA6.0, respectively. The proportion of cows with at least one SARA-positive day varied among farms from 0 to 100%. Automatic milking system was associated with an increased risk of SARA5.8 (odds ratio: 10) and SARA6.0 (odds ratio: 11). The use of corn silage was associated with an increased risk of SARA5.8 (odds ratio: 21), whereas the use of monensin was associated with a decreased risk of SARA5.8 (odds ratio: 0.02). Our study shows that rpH is quite variable among farms, but also among animals on the same farm. We also show that multiple animal and farm characteristics are associated with rpH variability and the risk of SARA under commercial conditions.

Cumulus cell gene expression associated with pre-ovulatory acquisition of developmental competence in bovine oocytes.

The final days before ovulation impact significantly on follicular function and oocyte quality. This study investigated the cumulus cell (CC) transcriptomic changes during the oocyte developmental competence acquisition period. Six dairy cows were used for 24 oocyte collections and received FSH twice daily over 3 days, followed by FSH withdrawal for 20, 44, 68 and 92 h in four different oestrous cycles for each of the six cows. Half of the cumulus-oocyte complexes were subjected to in vitro maturation, fertilisation and culture to assess blastocyst rate. The other half of the CC underwent microarray analysis (n=3 cows, 12 oocyte collections) and qRT-PCR (n=3 other cows, 12 oocyte collections). According to blastocyst rates, 20 h of FSH withdrawal led to under-differentiated follicles (49%), 44 and 68 h to the most competent follicles (71% and 61%) and 92 h to over-differentiated ones (51%). Ten genes, from the gene lists corresponding to the three different follicular states, were subjected to qRT-PCR. Interestingly, CYP11A1 and NSDHL gene expression profiles reflected the blastocyst rate. However most genes were associated with the over-differentiated status: GATM, MAN1A1, VNN1 and NRP1. The early period of FSH withdrawal has a minimal effect on cumulus gene expression, whereas the longest period has a very significant one and indicates the beginning of the atresia process.

The effects of LH inhibition with cetrorelix on cumulus cell gene expression during the luteal phase under ovarian coasting stimulation in cattle.

Cumulus cells have an important role to play in the final preparation of the oocyte before ovulation. During the final phase of follicular differentiation, FSH levels are low and LH maintains follicular growth; however, it is not known if at that time LH has an influence on cumulus cells inside the follicle. In humans, LH is often inhibited to avoid a premature ovulatory LH surge. This procedure provides a tool to investigate the role of LH in follicular development. In this study, we investigated the impact of suppressing LH using the GnRH antagonist cetrorelix during an ovarian coasting stimulation protocol on the transcriptome of bovine cumulus cells (CC). Oocytes were collected twice from 6 dairy cows. For the first collection, the cows received FSH twice daily for 3 d, followed by FSH withdrawal for 68 h as a control protocol. For the second collection, the same stimulation protocol was used, but the cows also received, starting on day 2 of FSH stimulation, a GnRH antagonist once a day until recovery of the cumulus-oocyte complexes (COC). Half of the COC were subjected to in vitro maturation, fertilization, and culture to assess blastocyst rates. The other half of the COC underwent microarray analysis (n = 3 cows, 2 treatments, 6 oocyte collections) and qRT-PCR (n = 6 cows: 3 microarray cows +3 other cows, 2 treatments, 12 oocyte collections). The differential expression of specific genes was confirmed by RT-qPCR: decrease of ATP6AP2, SC4MOL, and OSTC and increase of PTGDS in the LH-inhibited condition. The global transcriptomic analysis of cumulus cells demonstrated that the inhibition of LH secretion may decrease survival and growth of the follicle. Moreover, the results suggested that LH may be important to cumulus for the maintenance of cellular mechanisms such as global RNA expression, protein and nucleic acid metabolism, and energy production. These results support the hypothesis that LH support is important during the final part of follicle maturation through its influence on the cumulus cells.

Individual bovine in vitro embryo production and cumulus cell transcriptomic analysis to distinguish cumulus-oocyte complexes with high or low developmental potential.

Studying cumulus cell (CC) transcriptome is of great interest as it could provide a noninvasive method to assess oocyte quality. In cattle, the search for quality markers has not been done with cumulus-oocyte complexes (COCs) cultured individually from maturation to blastocyst stage. Here, differences between high- and low-potential COCs were examined by transcriptomic analysis of CC biopsies obtained from COCs of 2 to 6 mm follicles (n = 249; eight replicates) before individual in vitro maturation, fertilization, and culture until Day 8 after fertilization. Each COC was individually tracked and categorized based on his fate: embryo at blastocyst stage (CC-Blast) or embryo arrested at 2- to 8-cell stage (CC-2-8-cells). Average blastocyst rates were 27.7% for individual culture and 31.2% for group control (not significantly different). For transcriptomic analysis, five cumulus biopsies per replicate were pooled for each fate. Three CC replicates underwent transcriptomic analysis using RNA microarray assay. Some clear differences in gene expression between the CC-Blast and the CC-2-8-cell groups were identified. Considering a 1.5-fold change (P < 0.05), 68 genes were differentially expressed between the CC-Blast and CC-2-8-cells. Quantitative reverse transcription-polymerase chain reaction validations were performed for 12 selected genes: six upregulated genes for each COC fate. Higher expression of 1-acylglycerol-3-phosphate O-acyltransferase 9 (AGPAT9) (lipid metabolism), Chloride intracellular channel 3 (CLIC3), Keratin 8 (KRT8), and Lumican (LUM) (molecular transport) was observed in CC-2-8-cells (P < 0.05). The CC-Blast fate analysis revealed a significantly higher expression of Glycine amidinotransferase (L-arginine:glycine amidinotransferase) (GATM) (posttranslational modification, amino acid metabolism, and free radical scavenging). This newly identified set of genes could provide new markers to distinguish COCs associated with good quality embryos from COCs with limited developmental potential.