Article Cite This: J. Agric. Food Chem. 2018, 66, 12889−12897
pubs.acs.org/JAFC
Metabolomics Reveals that Crossbred Dairy Buffaloes Are More Thermotolerant than Holstein Cows under Chronic Heat Stress Zhaobing Gu,† Lin Li,† Shoukun Tang,‡ Chuanbin Liu,‡ Xianhai Fu,‡ Zhengxiang Shi,*,§ and Huaming Mao*,†,∥ †
Faculty of Animal Science and Technology, Yunnan Agricultural University, Kunming 650201, China Bureau of Animal Husbandry and Veterinary Medicine, Mangshi 678499, China § Department of Agricultural Structure and Bioenvironmental Engineering, College of Water Resources and Civil Engineering, China Agricultural University, Beijing 100083, China ∥ Yunnan Provincial Key Laboratory of Animal Nutrition and Feed Science, Kunming 650201, China
Downloaded via UNIV OF KANSAS on January 1, 2019 at 13:55:16 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
‡
S Supporting Information *
ABSTRACT: Heat stress (HS) threatens the worldwide dairy industry by decreasing animal production performance and health. Holstein cows and dairy buffaloes are the most important dairy animals, but their differences in the metabolic mechanism of thermotolerance remain elusive. In this study, we used serum metabolomics to evaluate the differences in thermotolerance between Holstein cows and crossbred dairy buffaloes under chronic heat stress (HS) and thermal-neutral conditions. In response to HS, the body temperatures and respiratory rates were increased more for Holstein cows than for dairy buffaloes (38.78 vs 38.24 °C, p < 0.001; 43.6 vs 32.5 breaths/min, p < 0.001). HS greatly affected serum metabolites associated with amino acids, fatty acids, and bile acids. The enriched metabolic pathways of these serum metabolites are closely related to HS. We demonstrated that buffaloes adapt to HS by adopting a metabolism of branched-chain amino acids and ketogenic amino acids and gluconeogenesis, but Holstein cows decrease the effect of HS with citrulline and proline metabolism. Both physiological parameters and serum metabolic profiles indicate that dairy buffaloes are more thermotolerant than Holstein cows, providing the feasibility to vigorously develop the buffalo dairy industry in tropical and subtropical regions. KEYWORDS: dairy buffaloes, heat stress, holstein cows, metabolomics, thermotolerance
■
INTRODUCTION The trend of global warming is obvious in recent decades.1 Among other concerns, it poses a great threat to the global dairy industry,2,3 especially considering that most dairy animals have low thermotolerance, and about 60% of raw milk is supplied by tropical and subtropical countries. Holstein cows ( Bos taurus), which are a temperate breed originating from The Netherlands, are the largest source of raw milk, and they can withstand extremely low temperatures (−12 °C); however, they are sensitive to temperatures exceeding 25 °C.4 As the second important source of raw milk, dairy buffaloes (Bubalus bubalis) are mainly distributed in tropical and subtropical regions,5,6 and the optimum temperatures and humidity for their growth and reproduction are 13−18 °C and 55−65%, respectively.7 Heat stress (HS) occurs when heat production exceeds dissipation. HS has become a widespread risk for the dairy industry in tropical, subtropical, and arid regions.8 Holstein cows have more thermal sensitivity due to their high metabolic rates and heat production.9 The temperature−humidity index (THI), calculated by the ambient temperature and relative humidity, is an indicator of HS levels for animals.10,11 When the THI exceeds 72 (or air temperature >25 °C), Holstein cows are unable to maintain heat balance, causing low dry material intake, low milk yield, and poor health.12−17 About 30% reduction of dry material intake causes 27.6% loss of milk production, and 50% of milk loss can be attributed to HS.18 © 2018 American Chemical Society
The greater thermal sensitivity of Holstein cows versus buffaloes is supported by the measurement of thermoneutral (TN) zones, which are the temperature tolerance ranges for homeothermal animals without additional energy to maintain normal body temperature. The TN zone is below −20 to 25 °C for Holstein cows and 10−36 °C for buffaloes.4 Conversely, buffaloes have fewer sweat glands and are often considered to be more sensitive compared with other domestic ruminants to HS.19−21 Despite the contradictory evidence, the differences in the molecular mechanisms of thermotolerance for Holstein cows and dairy buffalos have not yet been characterized. Metabolites, which play important regulatory roles in animal health and production, tend to be similar in different genuses.22 Thus, it is informative to discriminate the molecular mechanism for different species and genuses with metabolomics.23 In consideration of the high upper critical temperature, low rectal temperature, and small effect of HS on the conception rate and productivity,24−28 we speculated that buffaloes may be more thermotolerant than Holstein cows. Here, we used metabolomics to reveal the differences in thermotolerance between Holstein cows and dairy buffaloes Received: Revised: Accepted: Published: 12889
June 1, 2018 November 23, 2018 November 24, 2018 November 24, 2018 DOI: 10.1021/acs.jafc.8b02862 J. Agric. Food Chem. 2018, 66, 12889−12897
Article
Journal of Agricultural and Food Chemistry
Figure 1. Rhythms of indoor THI, body temperature, and respiration rates for Holstein cows and dairy buffaloes: (A) temperature humidity indexes (THI), (B) body temperatures (measured vaginally), and (C) respiratory rates (RR). HH, heat stressed Holstein cows (red); HB, heat stressed buffaloes (black); TNH, thermal-neutral Holstein cows (blue); TNB, thermal-neutral buffaloes (green). min, 65% B from 1 to 12 min, 40% B from 13 to 17 min, and 85% B from 18 to 22 min. To monitor the stability and repeatability of instrument analysis, quality-control samples were prepared by pooling 10 μL of each sample and analyzing them together with other samples. In MS-only acquisition, the instrument was set to acquire over the m/z range 60−1000 Da, and the accumulation time for the TOF MS scan was set at 0.20 s/spectra. The capillary voltage was maintained at ±5500 V for positive-mode and negative-mode detection. The source temperature was set at 600 °C. In automatic MS/MS acquisition, the instrument was set to acquire over the m/z range 25−1000 Da, and the accumulation time for the product ion scan was set at 0.05 s/ spectra. The product ion scan was acquired using informationdependent acquisition with the high-sensitivity mode selected. Parameters were set as follows: the collision energy was fixed at 35 ± 15 eV, the declustering potential was 60 V (+) and −60 V (−), isotopes were excluded within 4 Da, and six candidate ions were monitored per cycle. Data Deconvolution and Processing. Raw MS data (.wiff scan files) were converted to .mzXML files using ProteoWizard before doing peak matching, retention time alignment, and peak area extraction with XCMS software. Compound identification of metabolites was performed by comparing the accuracy of m/z values (1 were further analyzed by the Student’s t test at the univariate level to measure the significance of each metabolite. Pairwise comparisons of
under chronic HS, expecting that the differences in the molecular mechanisms of thermotolerance can provide theoretical guidance to develop the buffalo dairy industry in tropical and subtropical regions.
■
MATERIALS AND METHODS
Animals. Animals were cared for according to the principles of the Animal Welfare Act of 2007, issued by the Animal Ethics Committee of Yunnan Province, China. On the basis of the average THI values exceeding 72 from June to August, calculated with meteorological data from recent decades, a field trial of heat stress was conducted from July 20 to August 4, 2016. Six multiparous Holstein cows (designated HH, daily average milk yield 18.4 kg) and six Nili-Ravi × Murrah × Local crossbred buffaloes (designated HB, daily average milk yield 4.5 kg) with similar lactation were used in heat stress experiment. Another six multiparous Holstein cows (TNH, daily average milk yield 23.2 kg) and six dairy buffaloes (TNB, daily average milk yield 6.1 kg) with similar lactation were used as control groups under thermal-neutral (TN) conditions from January 3 to 18, 2017. All animals were loose housed, and they can obtain fresh drinking water and total mixed rations (80% whole-plant corn silage ad libitum, 12.5% concentrate feeding, 7.5% corn protein powder). The nutritional value of the three feed ingredients is shown in Table S1 of the Supporting Information (SI). Environmental and Physiological Parameter Measurements. Thermometers (±0.2 °C, Testo 175H1) were placed 2.0 m above the bedding area floor to record the ambient temperature and relative humidity at 30 min intervals in order to calculate the THI:11 THI = (1.8 × T + 32) − [(0.55 − 0.0055 × RH) × (1.8 × T − 26)], where T is the ambient temperature (°C) and RH is the relative humidity (%). For body temperature collection at 30 min intervals, a micro temperature sensor (DS1922L, ±0.5 °C) was placed in the vagina by CIDR (blank controlled internal drug release) without P4 (progesterone). Respiration rates were collected with a stopwatch at 08:00, 13:00, and 18:00 h. Sample Collection and Preparation. Blood samples were collected via jugular venipuncture with vacutainer tubes. Serum was obtained by centrifugation at 1400g for 10 min and stored at −80 °C until further analysis. Aliquots of 100 μL of serum were thawed at 4 °C and mixed with 400 μL of cold methanol/acetonitrile (1:1, v/v). The mixtures were stored at −20 °C for 20 min and centrifuged for 15 min (14000g, 4 °C). The supernatants were dried in a vacuum centrifuge. The dried samples were redissolved in 100 μL of acetonitrile/water (1:1, v/v) solvent and then centrifuged for 15 min (14000g, 4 °C). UHPLC−QTOF-MS Analysis. UHPLC−QTOF-MS (ultrahighperformance liquid chromatography equipped with quadrupole timeof-flight mass spectrometry) analysis was performed with an Agilent 1290 UHPLC system combined with a Q-TOF mass spectrometer (ESI/Triple TOF 5600; AB Sciex, Concord, Canada). The chromatographic separation was performed on an ACQUITY UPLC BEH Amide column (1.7 μm, 2.1 mm × 100 mm column). A 2 μL serum sample was injected into a column maintained at 25 °C, with the flow velocity of 0.3 mL/min. The mobile phase consisted of A = 25 mM ammonium hydroxide and 25 mM ammonium acetate in water and B = acetonitrile. The gradient elution program was 85% B from 0 to 1 12890
DOI: 10.1021/acs.jafc.8b02862 J. Agric. Food Chem. 2018, 66, 12889−12897
Article
Journal of Agricultural and Food Chemistry metabolite concentrations were also analyzed by the Student’s t test, and p values less than 0.05 were considered to be statistically significant. The univariate data analysis also included fold-change analysis.
■
RESULTS Physiological Parameters of Holstein Cows and Buffaloes under HS or TN Conditions. To reveal the differences in thermotolerance between Holstein cows and dairy buffaloes under chronic HS, we housed cows and buffaloes in parallel. Under TN conditions the barns of both Holstein cows (TNH) and dairy buffaloes (TNB) had THI values between 45 and 60, which is considered to be a comfortable thermal state. Furthermore, under HS conditions, the THI values for barns of the Holstein cows (HH) and dairy buffaloes (HB) were similar, and most values exceeded the threshold of HS (72) during the summer field trial (Figure 1A). To evaluate the responses in the body temperature, we measured the average vaginal temperatures every 30 min. All of the cows exhibited body temperature rhythms that reflected the changes in daily temperatures, with overall lower values around midday, and higher values from 18:00 to 22:00 h. However, the body temperatures of the HH group were significantly higher than those of the HB group (38.78 vs 38.24 °C, p < 0.001), despite the similar housing conditions. The body temperatures of the HH group were also statistically higher than those of the TNH group (38.54 °C, p < 0.001), but the body temperatures of the HB group were not statistically greater than those of the TNB group (38.16 °C, p > 0.05) (Figure 1B). These findings indicate that Holstein cows reacted to HS with elevated body temperatures. We also evaluated the respiration rates (RR, breaths/min) as an additional indicator of physiological response. The HH cows had greater RRs than the HB buffaloes did at 08:00, 13:00, and 18:00 h (p < 0.001). The RRs were also significantly greater in HB vs TNB groups (p < 0.001) and in HH vs TNH groups at each of the time points (Figure 1C; p < 0.001). These results indicate that both the Holstein cows and buffaloes reacted to HS with increased RR, but that the Holstein cows reacted more severely. Comparison of the Metabolic Profiles of Holstein Cows and Buffaloes. We further examined the metabolic profiles of blood serum from Holstein cows and buffaloes to determine whether the increased heat tolerance of the buffaloes might be explained by metabolic differences. OPLSDA plots of the serum metabolomics data show a clear separation with no overlap for HH vs HB (Figure 2A,B), HB vs TNB (Figure 2C,D), and HH vs TNH (Figure 2E,F), indicating that the serum metabolic profiles of the above three pairwise comparisons were distinct. OPLS-DA models with 7-fold cross-validation were also established using SIMCA-P 14.1 software. The UHPLC−QTOF-MS of serum metabolomics data for these three pairwise comparisons identified one predictive component and two orthogonal components. Satisfactory explanatory and predictive values for the intercepts (R2, Q2) indicate that OPLS-DA models were valid (Figure S1A−F, SI). These results indicate that both the Holstein cows and the buffaloes reacted to HS conditions with metabolic changes; however, the specific metabolic responses of the two types of dairy animals differed. Serum Metabolome Differences between Holstein Cows and Buffaloes under HS. To further evaluate the
Figure 2. Differentiation of the metabolic profiles of the HH vs HB, HB vs TNB and HH vs TNH groups using multivariate analysis. (A− F) OPLS-DA plots of the LC−MS data for the serum metabolome in both the positive and negative ionization mode.
differences in metabolic profiles between Holstein cows and buffaloes under HS, we examined the patterns of specific metabolites identified by UHPLC−QTOF-MS. Hierarchical clustering showed that the metabolites could be grouped according to their differential enrichment, with some increased and some decreased in HH relative to HB (Figure S2, SI). Examination of the differential metabolites in Table 1 revealed that many of them were enriched in KEGG pathways for pyrimidine metabolism, bile secretion, and biosynthesis of fatty acids and amino acids. Critical metabolites (2′-deoxyuridine, cytidine, cytosine, deoxycytidine, uracil, thymidine, and uridine) closely related to pyrimidine biosynthesis and catabolism were increased by 1.40−2.70-fold in the HH group compared with HB group (p < 0.05). Chenodeoxycholate and taurine, which are involved in bile acid secretion, were increased in the HH compared with the HB group by 1.84- and 1.98-fold, respectively (p < 0.05). Concentrations of oleic acid, arachidic acid, and linolenic acid, which are involved in the biosynthesis pathway of unsaturated fatty acids, were decreased by 0.68-, 0.29- and 0.75-fold, respectively (p < 0.05), and the concentrations of arachidonic acid and dodecanoic acid were increased by 1.26−2.38-fold, respectively (p < 0.05). Concentrations of leucine (Leu), lysine (Lys), tryptophan (Trp), valine (Val), and isoleucine (Ile), which are related to the pathway of the biosynthesis of amino acids, were decreased 0.48−0.68-fold in the HH group compared with the HB group (p < 0.05), but citrulline (Cit) was increased by 1.57-fold. To verify these findings, we performed UHPLC−MRM-MS analysis to quantify individual protein metabolites using standard solutions and optimized conditions. Consistent with the above data, the bile secretion pathway protein carnitine showed significantly lower levels in the HH group than in the HB group, as did acetylcarnitine and deoxycholic acid (p < 0.01) (Figure 3A). HH cows had higher concentrations of taurine than the HB group did (1.4 vs 1.2 μM), but no 12891
DOI: 10.1021/acs.jafc.8b02862 J. Agric. Food Chem. 2018, 66, 12889−12897
Article
Journal of Agricultural and Food Chemistry
Table 1. Distinct Pathway Metabolites Identified by UHPLC−QTOF-MS and Comparison of Their Levels between the HH and HB Groups in Positive or Negative Ionization Modea adduct (M (M (M (M (M (M (M M M M M
+ + + + + + +
+ + + +
H)+ H)+ H)+ H)+ H)+ H)+ H)+
H H NH4 H
M−H M−H M + CH3COO 2M − H M−H M M M M M M M
+H +H +H +H −H −H −H
M+H
metabolite
m/z
Pyrimidine Metabolism 2′-deoxyuridine 229.0805 cytidine 244.0914 cytosine 112.05 deoxycytidine 228.0965 thymidine 243.0962 uracil 113.0338 uridine 245.0753 Bile Secretion carnitine 162.1117 acetylcholine 146.1169 chenodeoxycholate 410.3252 taurine 126.0210 Biosynthesis of Unsaturated Fatty Acids arachidonic acid C20:4 303.2329 dodecanoic acid C12:0 199.1705 oleic acid C18:1 341.2689 arachidic acid C20:0 623.5977 linolenic acid C18:3 277.2169 Biosynthesis of Amino Acids glutamate 148.0594 leucine 132.1013 lysine 147.1116 tryptophan 205.0958 valine 116.0727 isoleucine 130.0885 citrulline 174.0897 betaine
118.0856
rt (s)
p-value
FC
201.91 445.34 381.07 380.93 174.44 291.53 291.51
