藉由微生物質體及代謝質體系統性研究荷蘭牛乳房炎

Abstract

全世界每年皆因乳房炎造成嚴重的酪農經濟損失。目前的研究指出乳房炎為 局部性器官發炎，感染原因不限於細菌性的感染，亦有研究指出消化道菌相組成和 代謝體學為影響宿主誘發發炎反應的重要因素之一。根據本研究室先前的研究發 現罹患乳房炎的泌乳牛，瘤胃中的菌相及代謝體與健康泌乳牛是具有差異性的，故 本研究目的為從健康泌乳及乳房炎荷斯登乳牛之瘤胃微生物及瘤胃代謝體中，找 出關鍵生物標記，再利用體外試驗模擬瘤胃系統驗證這些生物標記菌群與代謝物 作為生物補充製劑的可能性。 本試驗樣品為根據獸醫師的診斷、生乳中的體細胞數及酵素與血清中的細胞 激素結果，篩選出 30 頭泌乳牛，並將其分為健康組和乳房炎組。試驗中使用次世 代定序法 (NGS) 與超高效液相色譜-串聯質譜儀 (UHPLC-MS / MS) 分析健康組 和乳房炎組瘤胃液中的微生物菌相和代謝體之組成。在菌相定序的結果中發現健 康組和乳房炎組泌乳牛的瘤胃菌相豐富度與多樣性相似，進一步使用偏最小平方 判別分析 (PLS-DA) 分析，發現兩組間的菌相組成是具有差異性的，因此利用線 性判別分析效力量鑑別分析 (LEfSe)，鑑定兩組瘤胃中菌門至菌種的生物標記菌群， 結果顯示，在健康和乳房炎泌乳牛瘤胃菌相中共鑑定出 17 項菌屬和 3 項菌種可作 為微生物標記菌群。在乳房炎組中，菌種階層之生物標記菌群為 Unidentified rumen bacterium RNF82;健康組中鑑定出 Ruminococcus flavefaciens 與 Bifidobacterium longum subsp. longum 作為菌種階層之生物標記菌群。 接著透過 UHPLC-MS / MS 和 Student's t-test 分析瘤胃中的代謝物，總共鑑定 出 12,709 種代謝物，其中的 1,181 種代謝物在兩組間具有顯著差異的含量，其中 包含 65 種已知代謝物 (VIP> 1，P < 0.05) 。此外，運用 Spearman 相關係數評估瘤胃微生物相與代謝體間之相關性，3-methylcytosine、1- (3-aminopropyl) -4- aminobutanal、putrescine、piperidine 與 xanthurenic acid，此 5 種代謝物皆與健康組 的生物標記菌種呈現顯著正相關性。 因此，使用體外瘤胃發酵系統確認生物標記菌種與代謝物的相關性，透過額外 添加 Ruminococcus flavefaciens 與 Bifidobacterium longum subsp. longum 進行發酵試 驗，結果發現，額外添加菌群不會影響 pH 值及營養物質的消化率，但會顯著增加 (VFA) 中乙酸、乳酸與氨態氮 (NH3-N) 的濃度，亦能回復乳房炎組瘤胃生物標記 代謝物的濃度，達到與健康組濃度無顯著差異的效果。 綜合上述，本研究對瘤胃菌相及代謝體的探討驗證，生物標記菌種可能具有調 節瘤胃發酵，促進體內代謝效率之功效，希望能將生物標記菌種與代謝物發展為對 泌乳牛有益之飼料補充劑。

Dairy cow mastitis is one of the major economic challenges to the dairy industry worldwide. Typically, mastitis is an inflammation of the mammary gland, but not limited in bacterial infection. Our previous studies demonstrated that the composition of microbiota and metabolites in ruminal fluid between healthy and mastitis Holsten cows was different. Thus, the purpose of the present study is to identify the specific biomarkers (microorganism and metabolites) between healthy and mastitis Holstein dairy cows. The correlation between the biomarkers was also clarified and verified by analyzing data of microbiome/metabolome and in-vitro rumen fermentation, respectively. We expect that the biomarkers could be applied as a novel feed supplement for preventing dairy cow mastitis. First, thirty dairy cows were selected and separated into healthy and mastitis groups according to the results of veterinary diagnosis, milk somatic cells, milk enzyme and serum cytokines. We then evaluated compositions of microbiota and metabolites in ruminal fluids from healthy and mastitis cows by next generation sequencing technology (NGS) and ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS), respectively. The results of NGS indicated that the community richness and diversity indices were not different between groups. Further examination by the partial least squares discriminant analysis (PLS-DA) showed a clear differentiation of healthy and mastitis groups. We then evaluated the compositions of ruminal microbiota and identified the bacterial biomarkers using linear discriminant analysis effect size (LEfSe) analysis.Results indicated that microbiome biomarkers including seventeen genera and three species were identified in the healthy and mastitis groups. For mastitis group, 1 species, Unidentified rumen bacterium RNF82, was identified as biomarker. While, Ruminococcus flavefaciens and Bifidobacterium longum subsp. longum were found at the species level as biomarkers of healthy group. Further ruminal metabolomic analysis showed that 1181 metabolites had significant difference between groups, including 65 identified metabolites (VIP > 1, P < 0.05). Additionally, the relationship between ruminal microbiota and metabolites was also evaluated by Spearman’s correlation coefficient. The results showed that Ruminococcus flavefaciens and Bifidobacterium longum subsp. longum have positive correlation and significant difference with 5 metabolites including 3-methylcytosine, 1-(3-aminopropyl)-4-aminobutanal, putrescine, piperidine and xanthurenic acid. Therefore, in-vitro rumen fermentation was conducted to confirm the correlation of microbiome and metabolome. Results indicated that the groups with Ruminococcus flavefaciens and Bifidobacterium longum subsp. longum wouldn’t affect the pH value and the digestibility of nutrients but would significantly increase the concentration of VFA and NH3-N. The biomarker metabolites in the mastitis groups were also restored as compared with the healthy groups. In conclusion, we found that the specific microorganisms which are negatively related with mastitis, could affect the rumen metabolome. The finding could provide a better understanding for the dairy cow mastitis with the rumen microbiota, and these biomarkers could be used as a novel feed supplement to prevent mastitis in the near future.