Advertisement

Effectiveness of the Australian breeding value for heat tolerance at discriminating responses of lactating Holstein cows to heat stress

Open AccessPublished:July 22, 2022DOI:https://doi.org/10.3168/jds.2021-21741

      ABSTRACT

      Heat stress has negative consequences for milk production and reproduction of dairy cattle. These adverse effects are likely to increase because of climate change and anticipated increases in milk yield. Some of the variation among cows in ability to resist effects of heat stress is genetic. The current objective of this observational study was to assess the effectiveness of the Australian breeding value for heat tolerance (ABVHT) based on the decline in milk yield with heat stress for predicting cow differences in effects of heat stress on regulation of body temperature, milk production, and reproductive function. Genomic breeding values for heat tolerance were calculated for 12,487 cows from a single California dairy farm. Rectal temperature in the afternoon (1100–2045 h) was measured on a subset of 626 lactating cows with ABVHT ≥102 (heat tolerant) or <102 (heat sensitive). Rectal temperature was 0.12°C lower for heat-tolerant cows than heat-sensitive cows. Vaginal temperatures were measured every 15 min for 5 d in 118 cows with ABVHT ≥108 (extreme heat tolerant) or <97 (extreme heat sensitive). Vaginal temperature was 0.07°C lower for extreme heat–tolerant cows than extreme heat–sensitive cows. Lactation records for 4,703 cows with ABVHT were used to evaluate seasonal variation in first 90-d milk yield, fat percent, and protein percent for each ABVHT quartile. Overall, cows with higher ABVHT had lower milk yield, fat percentage, and protein percentage and higher first service pregnancy rate. There was no summer depression in production or reproduction or interactions between season and ABVHT quartile. We observed that ABVHT can successfully identify heat-tolerant cows that maintain lower body temperatures during heat stress. The lack of a pronounced seasonality in milk production or reproduction precluded evaluation of whether ABVHT is related to the magnitude of effect of heat stress on those traits.

      Key words

      INTRODUCTION

      Climate change is a growing concern for food production systems across the globe. Historically hot regions are facing record high temperatures for longer periods, and a greater proportion of the globe is experiencing temperatures characteristic of heat stress (

      IPCC. 2014. Clim. Change. 2014 (Synthesis Report.).

      ). Heat stress in livestock impairs growth, production, and reproductive performance (
      • Hansen P.J.
      Heat stress and climate change.
      ). In dairy cattle, there is a decline in milk, fat, and protein yields in warm seasons (
      • Bohmanova J.
      • Misztal I.
      • Cole J.B.
      Temperature-humidity indices as indicators of milk production losses due to heat stress.
      ;
      • Hammami H.
      • Bormann J.
      • M'hamdi N.
      • Montaldo H.H.
      • Gengler N.
      Evaluation of heat stress effects on production traits and somatic cell score of Holsteins in a temperate environment.
      ;
      • Guinn J.M.
      • Nolan D.T.
      • Krawczel P.D.
      • Petersson-Wolfe C.S.
      • Pighetti G.M.
      • Stone A.E.
      • Ward S.H.
      • Bewley J.M.
      • Costa J.H.C.
      Comparing dairy farm milk yield and components, somatic cell score, and reproductive performance among United States regions using summer to winter ratios.
      ). Heat stress also reduces the ability of cattle to establish and maintain pregnancy because of damage to the oocyte, fertilization failure, and early embryonic mortality (
      • Hansen P.J.
      Reproductive physiology of the heat-stressed dairy cow: Implications for fertility and assisted reproduction.
      ). Heat abatement management strategies, including fans, sprinklers or soakers, cooling ponds, and shade cloth, can be applied on farms experiencing heat stress to reduce negative effects of heat stress on dairy cattle (
      • Collier R.J.
      • Dahl G.E.
      • Vanbaale M.J.
      Major advances associated with environmental effects on dairy cattle.
      ). These strategies often reduce, but do not eliminate, negative consequences of heat stress. Similar to many challenges faced by the livestock sector, breeding for a novel trait, such as heat tolerance, offers an additional potential solution to the challenges of heat stress.
      Rectal temperature, a trait that is a component of heat tolerance, has been shown to have a moderately high heritability of 0.17 (
      • Dikmen S.
      • Cole J.B.
      • Null D.J.
      • Hansen P.J.
      Heritability of rectal temperature and genetic correlations with production and reproduction traits in dairy cattle.
      ). Thus, it might be possible to select cattle genetically for resistance to heat stress. Several approaches have been used to estimate the genetic component of thermotolerance based on the response of milk production to increasing temperature-humidity index (THI;
      • Ravagnolo O.
      • Misztal I.
      Genetic component of heat stress in dairy cattle, parameter estimation.
      ;
      • Hayes B.J.
      • Bowman P.J.
      • Chamberlain A.J.
      • Savin K.
      • van Tassell C.P.
      • Sonstegard T.S.
      • Goddard M.E.
      A validated genome wide association study to breed cattle adapted to an environment altered by climate change.
      ;
      • Sánchez J.P.
      • Misztal I.
      • Aguilar I.
      • Zumbach B.
      • Rekaya R.
      Genetic determination of the onset of heat stress on daily milk production in the US Holstein cattle.
      ).
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-Mariam M.
      • Pryce J.E.
      • Hayes B.J.
      Genomic selection for tolerance to heat stress in Australian dairy cattle.
      developed an Australian breeding value for heat tolerance (ABVHT) in Holstein and Jersey cows based on the magnitude of decline of milk, fat, and protein yield per unit increase in THI. The heritabilities for ABVHT for milk, fat, and protein yield were 0.22, 0.20, and 0.23, respectively (
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-Mariam M.
      • Pryce J.E.
      • Hayes B.J.
      Genomic selection for tolerance to heat stress in Australian dairy cattle.
      ). The ABVHT has a negative genetic correlation with production traits and a positive genetic correlation with reproduction traits. Using climate-controlled chambers,
      • Garner J.B.
      • Douglas M.L.
      • Williams S.R.O.
      • Wales W.J.
      • Marett L.C.
      • Nguyen T.T.T.
      • Reich C.M.
      • Hayes B.J.
      Genomic selection improves heat tolerance in dairy cattle.
      showed that cows with higher ABVHT had less of a decline in DMI and milk yield under climate-controlled heat wave conditions than cows with lower ABVHT. High ABVHT cows also maintained lower rectal and vaginal temperatures and elevated skin temperatures, indicating the ability to better maintain body temperature under heat stress. The ABVHT was incorporated into Australian national genetic evaluations in December 2017 (
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-mariam M.
      • Nieuwhof G.J.
      • Hayes B.J.
      • Pryce J.E.
      Short communication: Implementation of a breeding value for heat tolerance in Australian dairy cattle.
      ).
      The intercountry correlation between genomic estimates of dairy cattle in the United States with cattle in Australia for production traits has been estimated as 0.72 (
      • Interbull
      MACE August 2021 for Production.
      ). Differences between countries in sire rankings are probably mostly due to differences in management. For example, Australian farms are primarily pastoral based (extensive), whereas US farms are primarily intensive, freestall operations. Cattle in intensive systems are likely to be affected differently by heat stress than cattle in extensive systems because animal housing for the former system is often designed to include features to abate some of the consequences of heat stress. Nonetheless, it is possible that many of the genetic variants that make cattle resistant to heat stress in Australia would function similarly in other locations and that the ABVHT could be informative in identifying animals that are genetically superior in resistance to heat stress.
      The objective of the present study was to evaluate the effectiveness of ABVHT at predicting and selecting animals that are more heat tolerant under intensive conditions. It was hypothesized that core body temperature would be lower, and the reduction in milk production and reproduction during the summer less severe, for lactating females with high ABVHT than for those with low ABVHT.

      MATERIALS AND METHODS

      Study Location

      Data were collected from a commercial dairy farm in Riverdale, California (36°32′06.57′′ N 120°01′23.8′′ W; elevation 56.4 m). The dairy was chosen because it is in a hot region of the country and makes extensive use of genotyping necessary to calculate ABVHT. The cows were Holstein, milked twice daily, and housed in freestall barns with fans, soakers, and shade-cloth curtains for heat abatement. Fans were programmed to turn on at 20°C and increase in speed with increasing dry bulb temperature. Cows were cooled by 2-stage feed line soakers. Soakers turned on at a dry bulb temperature of 24°C for 30 s at 5-min intervals. When dry bulb temperature reached 29.5°C, the interval between 30 s of soaking was reduced to 3 min. Cows had access to outdoor dirt lots until 0200 h (night milking) to encourage cows to eat during the cooler night temperatures. The farm had 8 barns with 4 pens of ∼100 cows/barn. Data for this study were obtained from cows in 20 of the 32 pens, representing all 8 barns. None of the cows used for the study were from fresh pens or the hospital pen.
      As of September 2020, the farm had a rolling herd average of 11,323 kg of milk with a 3.86% fat test and 3.15% protein test. Herd records were obtained from the farm and accessed using DHI-Plus (Amelicor). In addition to current herd records, archived records were used for production and reproduction analyses. The study was approved by the University of Florida Institutional Animal Care and Use Committee (Project 202010982).

      Genotypes

      Genotypes (n = 12,684) obtained using Clarifide genetic testing platforms (Zoetis) and pedigrees from animals on the farm from 2003 to 2019 were sent to DataGene (Melbourne, Australia) and included in the August 2020 official genetic evaluation run. A standard set of 45,685 SNP genotypes is used for routine evaluations by DataGene (
      • Nieuwhof G.J.
      • Beard K.T.
      • Konstantinov K.V.
      • Bowman P.J.
      • Hayes B.J.
      Implementation of genomics in Australia.
      ), and missing genotypes were imputed by DataGene to satisfy this requirement. In brief, the procedure was as follows: samples with a call rate less than 0.9 or with more than 40% of markers heterozygous were removed. Animals with parentage or sex inconsistencies between the pedigree and genotype were also excluded. After quality control, genomic ABVHT were calculated following the methodology of
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-Mariam M.
      • Pryce J.E.
      • Hayes B.J.
      Genomic selection for tolerance to heat stress in Australian dairy cattle.
      ,
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-mariam M.
      • Nieuwhof G.J.
      • Hayes B.J.
      • Pryce J.E.
      Short communication: Implementation of a breeding value for heat tolerance in Australian dairy cattle.
      ) for 12,487 cows. The estimated reliability of ABVHT is 0.38 (
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-mariam M.
      • Nieuwhof G.J.
      • Hayes B.J.
      • Pryce J.E.
      Short communication: Implementation of a breeding value for heat tolerance in Australian dairy cattle.
      ).

      Rectal and Vaginal Temperatures

      The experiment was performed during 2 wk in August 2020. During the study, wildfires resulted in much denser smoke during wk 2. Of the 12,487 cows with ABVHT, 2,925 cows were in the herd at the time the experiment was initiated (of 3,613 lactating cows total). Rectal temperatures were measured once per cow over the span of the experiment using digital rectal thermometers from a total of 1,078 cows in the 30 min after they returned from milking (range of sampling time was 1100–2045 h). Cows were selected for measurements of rectal temperature at random and without an a priori knowledge of the ABVHT. A total of 452 cows were not genotyped; therefore, an ABVHT was not available. An ABVHT was calculated for the remaining cows (n = 626). The distribution of ABVHT values is shown in Figure 1. The average ABVHT was 102, and cows were divided into 2 groups for statistical analysis. Those cows with an ABVHT from 102 to 109 were termed heat tolerant (n = 354), and those with an ABVHT from 95 to 101 were termed heat sensitive (n = 272).
      Figure thumbnail gr1
      Figure 1Distribution of Australian breeding value for heat tolerance (ABVHT) for 12,487 genotyped cows located on a California dairy from 2003 to 2019.
      Cows used to measure vaginal temperature were selected from genotyped cows only and, in particular, from cows chosen to be the most heat tolerant (ABVHT ≥108; termed extreme heat tolerant) and least heat tolerant (ABVHT ≤97; termed extreme heat sensitive) on the farm (Figure 1). A blank controlled internal drug release (CIDR) containing an iButton 1922L (Maxim Integrated) was placed intravaginally for 5 d to record temperature every 15 min. The iButtons were set to 0.5°C accuracy. The experiment was performed with 40 extreme heat–tolerant cows and 23 extreme heat–sensitive cows in wk 1 and with a separate 26 extreme heat–tolerant cows and 29 extreme heat–sensitive cows in wk 2. Cows were milked at times ranging from 1100 to 1430 h and 2300 to 0230 h.
      Dry bulb temperature and relative humidity were measured every 15 min for the duration of the study using Hobo U23 Pro v2 temperature and relative humidity data loggers (Onset) from the following 3 locations at the farm: exterior parking lot, the barn identified as the hottest by the farm manager, and the barn identified as the coolest. The THI was calculated using the following equation of
      • NRC
      A Guide to Environmental Research on Animals.
      :
      THI = (1.8 × Tdb + 32) − [(0.55 − 0.0055 × RH) × (1.8 × Tdb − 26.8)],


      where Tdb = dry bulb temperature (°C) and RH = relative humidity.

      Statistical Analysis

      Statistical analysis was performed using R version 4.0.3 (https://www.r-project.org/). Rectal temperature was analyzed using the following model:
      y = µ + + ε,


      where y is a vector of rectal temperature; µ = population mean; β is a vector of fixed effects including barn temperature (covariate of barn closest to the location of the cows), test-day milk yield closest to temperature measurements (covariate), DIM (<60, 60–119, 120–179, 180–249, and >249), parity (primiparous or multiparous), day of measurement (7 classes), and ABVHT class; X is a design matrix of the fixed effects; and ε = error.
      Vaginal temperature was analyzed with 2 models. First, average vaginal temperature was analyzed using the following model:
      y = µ + + ε,


      where y is a vector of vaginal temperature averaged across 5 d; β is a vector of fixed effects including ABVHT class, week (2 levels), pen (19 levels), test-day milk yield (covariate), parity (primiparous or multiparous), and DIM (120–179, 180–249, and >249); and X is a design matrix for fixed effects.
      Second, vaginal temperature was analyzed with time of day in the model. Data for wk 1 and 2 were analyzed separately using the following mixed model:
      y = µ + + Zg+ ε,


      where y is a vector of vaginal temperature; β is a vector of fixed effects including ABVHT class, day (5 levels), time of day (0000 h to 2400 h by 15-min intervals), day by ABVHT class interaction, time by ABVHT class interaction, test-day milk yield (covariate), parity (primiparous of multiparous), and DIM (120–179, 180–249, and >249); g is a vector of random effects including cow nested within ABVHT; and X and Z are design matrices for fixed and random effects, respectively.

      Seasonal Variation in Milk Production and Reproduction

      An analysis was performed to determine whether seasonal differences in milk production and reproduction were affected by ABVHT. For the analysis, summer was defined as May through September, and winter was defined as November through March. For production responses, only records from cows in which the first 90 DIM occurred in 1 season only were analyzed. A total of 4,703 cows with genotypes and farm records from 2015 to 2020 met this criterion. Limiting the analysis of milk yield to cows where the first 90 DIM were in summer or winter maximized the probability that lactation records were collected during hot or cool conditions only. Each cow's most recent lactation record was used. For reproduction, first service conception rate was compared for cows inseminated in summer and winter; records for 3,597 cows genotyped for ABVHT were analyzed. Cows were sorted into quartiles based on ABVHT for the analysis with ABVHT1 (lowest ABVHT, most heat sensitive; <100), ABVHT2 (100 and 101), ABVHT3 (102–104), and ABVHT4 (highest ABVHT, most heat tolerant; >104).
      The model for analysis of production traits was as follows:
      y = µ + + ε,


      where y is a vector of mean first 90 DIM yield, fat percent, or protein percent; β is a vector of fixed effects including ABVHT quartile, season (summer or winter), interaction of ABVHT quartile and season, average test DIM, parity (primiparous or multiparous), and calving year; and X is a design matrix of the fixed effects. Average test DIM was included because milk yield was measured on test dates only; therefore, average test DIM varied between cows.
      A generalized linear model with a binomial distribution to analyze pregnancy per AI was as follows:
      y = µ + + ε,


      where y is a vector of pregnancy outcome after first service (pregnant or not pregnant); β is a vector of effects including ABVHT quartile, season (summer or winter), interaction of ABVHT quartile and season, DIM at first breeding, parity (primiparous or multiparous), and calving year; and X is a design matrix of the fixed effect.
      In all analyses, statistical significance was declared when P < 0.05.

      RESULTS AND DISCUSSION

      Distribution of ABVT and Genomic Correlations with Other Traits

      Breeding values for 12,487 ABVHT exhibited a normal distribution with a SD of 3.6 (Figure 1). This result is similar to the distribution of ABVHT in Australia, where the within breed SD is standardized to 5 (
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-mariam M.
      • Nieuwhof G.J.
      • Hayes B.J.
      • Pryce J.E.
      Short communication: Implementation of a breeding value for heat tolerance in Australian dairy cattle.
      ). The farm average ABVHT was 102, which is slightly higher than the population average of 100 in Australia. Perhaps, the farm inadvertently selected cows for heat tolerance, or cows better adapted to heat stress have survived better in Southern California.
      Genomic correlations were calculated between the ABVHT and US genomic EBV for major traits including milk yield, fat yield, fat percent, protein yield, protein percent, net merit $, daughter pregnancy rate, productive life, and SCS (Table 1). There were significant negative correlations between ABVHT and milk yield (r = −0.275), fat yield (r = −0.420), protein yield (r = −0.470), net merit $ (r = −0.395), productive life (r = −0.062), and SCS (r = −0.033). Genomic breeding value for ABVHT was positively (i.e., favorably) correlated with daughter pregnancy rate (r = 0.160). Correlations were in the same direction as reported earlier by
      • Nguyen T.T.T.
      • Bowman P.J.
      • Haile-Mariam M.
      • Pryce J.E.
      • Hayes B.J.
      Genomic selection for tolerance to heat stress in Australian dairy cattle.
      for Australian cows, although the magnitude of correlations was often smaller than earlier reported. Differences are likely to be partially due to the intercountry genetic correlation between the United States and Australia for milk production traits being ∼0.72 (
      • Interbull
      MACE August 2021 for Production.
      ).
      • Ravagnolo O.
      • Misztal I.
      Genetic component of heat stress in dairy cattle, parameter estimation.
      and
      • Bohmanova J.
      • Misztal I.
      • Tsuruta S.
      • Norman H.D.
      • Lawlor T.J.
      National genetic evaluation of milk yield for heat tolerance of United States Holsteins.
      also found negative correlations between milk yield and heat tolerance and positive correlations with fertility and productive life.
      Table 1Genomic correlations between Holstein Australian breeding value for heat tolerance and the US genomic PTA for cows located on a commercial California dairy (n = 2,131)
      Pro = protein; NM$ = net merit $; DPR = daughter pregnancy rate; PL = productive life.
      ItemUS genomic breeding values
      PTA milkPTA fatPTA fat%PTA ProPTA Pro%PTA NM$PTA DPRPTA PLPTA SCS
      Breeding value for heat tolerance−0.275−0.420−0.129−0.470−0.250−0.3950.160−0.062−0.033
      P-value2.20E-162.20E-162.638E-092.20E-162.20E-162.20E-161.20E-130.004380.126
      1 Pro = protein; NM$ = net merit $; DPR = daughter pregnancy rate; PL = productive life.
      The direction of correlations between ABVHT and other traits makes biological sense. Increases in milk yield are associated with increased metabolic heat production (
      • Mesgaran S.D.
      • Eggert A.
      • Höckels P.
      • Derno M.
      • Kuhla B.
      The use of milk Fourier transform mid-infrared spectra and milk yield to estimate heat production as a measure of efficiency of dairy cows.
      ) and higher rectal temperatures during heat stress (
      • Dikmen S.
      • Khan F.A.
      • Huson H.J.
      • Sonstegard T.S.
      • Moss J.I.
      • Dahl G.E.
      • Hansen P.J.
      The SLICK hair locus derived from Senepol cattle confers thermotolerance to intensively managed lactating Holstein cows.
      ). Therefore, it is reasonable that one genetic component of a cow's resistance to heat stress would be lower heat production associated with milk yield. Care must be taken when selecting for high ABVHT to not also select for low milk yield. One of the reasons for reduced fertility during heat stress is damage to the oocyte and embryo by elevated body temperature (
      • Hansen P.J.
      Effects of heat stress on mammalian reproduction.
      ). Therefore, cows that are genetically more resistant to heat stress would be expected to be genetically superior for fertility in hot climates. Farmers wanting superior milk production and heat tolerance can identify bulls with positive EBV for both of these traits.

      Rectal Temperature

      Rectal temperatures adjusted for all terms in the model were lower (P = 0.032) for cows designated as heat tolerant (ABVHT ≥ 102) than cows designated as heat sensitive (ABVHT < 102; Figure 2A). The mean rectal temperature was 38.46°C and 38.58°C for heat-tolerant and heat-sensitive cows, respectively. The difference between the 2 ABVHT groups remained if data were not adjusted for milk yield (P = 0.034; mean rectal temperature = 38.53°C for heat-tolerant cows vs. 38.65°C for heat-sensitive cows). Variation in rectal temperature as a function of dry bulb temperature for heat-tolerant and heat-sensitive cows is shown in Figure 2B. As expected (
      • Dikmen S.
      • Khan F.A.
      • Huson H.J.
      • Sonstegard T.S.
      • Moss J.I.
      • Dahl G.E.
      • Hansen P.J.
      The SLICK hair locus derived from Senepol cattle confers thermotolerance to intensively managed lactating Holstein cows.
      ), rectal temperature increased as dry bulb temperature increased. Greatest differences between heat-tolerant and heat-sensitive cows occurred when dry bulb temperatures were less than 33°C (Figure 2B). This result indicated that differences in ability to regulate body temperature between heat-tolerant and heat-sensitive cows decline as heat stress becomes more severe. The results also confirm that differences in rectal temperature between heat-sensitive and heat-tolerant cows still exist even after accounting for milk production level.
      Figure thumbnail gr2
      Figure 2Differences in rectal temperature between heat-tolerant (n = 354) and heat-sensitive cows (n = 272). Heat-tolerant cows had an Australian breeding value for heat tolerance (ABVHT) ≥102, and heat-sensitive cows had an ABVHT < 102. (A) Mean rectal temperature ± SEM after adjustment (Adj.) for barn temperature, milk yield, parity, and day. The ABVHT affected rectal temperature (P = 0.032). (B) Variation in rectal temperature as a function of barn temperature for heat-tolerant (blue) and heat-sensitive cows (orange). Results from individual cows are indicated by the circles or squares, whereas the lines represent Lowess regression curves for each heat tolerance group.
      The upper critical temperature for dairy cattle has been reported to vary from 25 to 28.4°C (
      • Berman A.
      • Folman Y.
      • Kaim M.
      • Mamen M.
      • Herz Z.
      • Wolfenson D.
      • Arieli A.
      • Graber Y.
      Upper critical temperatures and forced ventilation effects for high-yielding dairy cows in a subtropical climate.
      ;
      • Dikmen S.
      • Hansen P.J.
      Is the temperature-humidity index the best indicator of heat stress in lactating dairy cows in a subtropical environment?.
      ). However, average rectal temperature for both the heat-tolerant and heat-sensitive cows was not elevated until barn dry bulb temperature exceeded 32°C (Figure 2B). Perhaps, cows in California have adapted to chronic heat stress, or the cooling techniques used by the dairy (e.g., sprinklers, fans, shade cloth, access to dry lots) were successful in mitigating heat stress effects. It should be noted that many cows were hyperthermic at lower dry bulb temperatures than 32°C, including between 26 and 27°C. Such a result highlights the variation among cows in ability to regulate body temperature.

      Vaginal Temperature

      Cows were selected for the experiment based on an ABVHT indicating extreme heat tolerance (ABVHT ≥ 108) or extreme heat sensitivity (ABVHT ≤ 97). Vaginal temperature was measured at 15-min intervals for 5 d for 2 separate groups of cows on different weeks.
      When data from both weeks were analyzed together, vaginal temperatures were lower for extreme heat–tolerant cows (ABVHT ≥ 108) than for extreme heat–sensitive cows (ABVHT ≤ 97). Data were analyzed with and without test-day milk yield as a covariate in the model and were shown to be consistent. After fitting milk yield as a covariate, average vaginal temperature was 39.02°C for extreme heat–tolerant cows and 39.09°C for extreme heat–sensitive cows (P < 0.001). Without the covariate, average vaginal temperature was 39.03°C for extreme heat–tolerant cows and 39.11°C for extreme heat–sensitive cows (P < 0.001). These results were similar to those of
      • Garner J.B.
      • Douglas M.L.
      • Williams S.R.O.
      • Wales W.J.
      • Marett L.C.
      • Nguyen T.T.T.
      • Reich C.M.
      • Hayes B.J.
      Genomic selection improves heat tolerance in dairy cattle.
      , who used climate-controlled chambers in Australia and found heat-tolerant cows (i.e., with high ABVHT) had significantly lower body temperatures during a heat challenge than heat-sensitive cows (low ABVHT).
      There was a large effect of week on vaginal temperature (P < 0.001), with higher vaginal temperatures in wk 1 than wk 2 (39.07°C vs. 39.04°C). Accordingly, analysis of vaginal temperature data to examine possible interactions between ABVHT and time of day were analyzed for wk 1 separately from wk 2. Extreme heat–tolerant cows had a lower average vaginal temperature than extreme heat–sensitive cows during both weeks regardless of milk yield (Table 2).
      Table 2Adjusted vaginal temperature (°C) ± SEM for each week calculated with and without milk yield in the model
      Extreme heat–tolerant cows defined as ABVHT ≥108 (n = 40 cows for wk 1 and 26 cows for wk 2); extreme heat–sensitive cows defined as ABVHT <97 (n = 23 cows for wk 1 and 29 cows for wk 2).
      Vaginal temperature differed (P < 0.001) between heat-tolerant and heat-sensitive cows for both models and for both weeks.
      ItemWk 1 (August 17–22)Wk 2 (August 25–29)
      With milk yieldWithout milk yieldWith milk yieldWithout milk yield
      Extreme heat tolerant39.00 ± 0.11839.02 ± 0.11738.91 ± 0.09738.91 ± 0.096
      Extreme heat sensitive39.10 ± 0.12339.05 ± 0.12239.04 ± 0.09239.05 ± 0.089
      1 Extreme heat–tolerant cows defined as ABVHT ≥108 (n = 40 cows for wk 1 and 26 cows for wk 2); extreme heat–sensitive cows defined as ABVHT <97 (n = 23 cows for wk 1 and 29 cows for wk 2).
      2 Vaginal temperature differed (P < 0.001) between heat-tolerant and heat-sensitive cows for both models and for both weeks.
      Effect of time of day on vaginal temperature for extreme heat–tolerant and extreme heat–sensitive cows adjusted for milk yield is shown in Figure 3. For wk 1, there was a significant effect of ABVHT (P < 0.001) and time (P < 0.001), but there was no interaction between time and ABVHT (Figure 3A). Vaginal temperatures were high in the early morning, declined to a nadir at 1100 to 1200 h, and then increased in the afternoon, peaking soon after the daily peak in THI (Figure 3C). In wk 2, there was an effect of time (P < 0.001) and an interaction between ABVHT and time of day (P < 0.001) but no significant effect of ABVHT (P = 0.225). The daily variation in vaginal temperature was different than for wk 1 (Figure 3B). Temperatures were low in the morning (although lower for extreme heat–tolerant cows than extreme heat–sensitive cows) and rose to a peak around 1100 to 1200 h. Then, vaginal temperatures declined slightly to a nadir around 1400 to 1500 h (corresponding to most cows returning from the parlor and drinking large amounts of water) and then either remained constant in extreme heat–tolerant cows or increased in heat–sensitive cows. Thus, for most of the day, except after the late-morning peak, extreme heat–tolerant cows experienced lower vaginal temperatures than extreme heat–sensitive cows.
      Figure thumbnail gr3
      Figure 3Daily variation in vaginal temperature recorded at 15-min intervals for extreme heat–tolerant cows [Australian breeding value for heat tolerance (ABVHT) ≥108] and extreme heat–sensitive cows (ABVHT ≤ 97). Cows were milked at times ranging from 1100 to 1430 h and 2300 to 0230 h. (A) Mean vaginal temperature ± SEM adjusted (Adj.) for milk yield, days fresh, and parity during wk 1 (n = 40 extreme heat–tolerant cows and 23 extreme heat–sensitive cows). Temperature was affected by ABHVT (P < 0.001) and time (P < 0.001) but not by the interaction. (B) Mean vaginal temperature ± SEM adjusted for milk yield, days fresh, and parity during wk 2 (n = 26 extreme heat–tolerant cows and 29 extreme heat–sensitive cows). Temperature was affected by time (P < 0.001) and the ABVHT × time interaction (P < 0.001). (C, D) Daily variation in mean dry bulb barn temperature (left y-axis), and temperature-humidity index (THI; right y-axis) at 15-min intervals for wk 1 (C) and wk 2 (D).

      Seasonal Variation for Production and Reproductive Traits

      Means ± SEM for milk yield, fat percent, and protein percent during the first 90 DIM and first service pregnancy rate for summer and winter are reported in Table 3. There were significant differences between ABVHT quartiles (P < 0.001) for milk yield, fat percent, and protein percent, with higher quartiles of ABVHT associated with reduced yield, fat percent, and protein percent. These results are consistent with the negative genetic correlations between ABVHT and production traits (Table 1). Similar results were obtained if complete lactation data were analyzed (results not shown).
      Table 3Effect of Australian breeding value for heat tolerance (ABVHT) on summer and winter means (± SEM) for average daily milk yield, fat percent, and protein percent in the first 90 d of lactation and pregnancy per AI at first service
      n = 4,703 for milk production traits; n = 3,597 for pregnancy/first service.
      Genotyped cows were divided into quartiles with ABVHT1 = lowest ABVHT, most heat sensitive (<100), ABVHT2 (100 and 101), ABVHT3 (102–104) and ABVHT4 = highest ABVHT, most heat tolerant (>104).
      ItemABVHT1ABVHT2ABVHT3ABVHT4
      Milk
      Means were adjusted for average DIM, parity (primiparous or multiparous), and calving year. Each variable was affected by ABVHT quartile (P < 0.001). Season affected first 90-d fat percent (P < 0.001) and protein percent (P = 0.023) but not milk yield. There were no interactions between ABVHT quartile and season.
      (kg/d)
       Summer43.6 ± 0.3943.3 ± 0.4142.0 ± 0.3141.4 ± 0.31
       Winter42.7 ± 0.4342.3 ± 0.4141.8 ± 0.3241.2 ± 0.30
      Fat
      Means were adjusted for average DIM, parity (primiparous or multiparous), and calving year. Each variable was affected by ABVHT quartile (P < 0.001). Season affected first 90-d fat percent (P < 0.001) and protein percent (P = 0.023) but not milk yield. There were no interactions between ABVHT quartile and season.
      (%)
       Summer3.72 ± 0.033.74 ± 0.033.69 ± 0.023.61 ± 0.02
       Winter3.67 ± 0.033.64 ± 0.033.60 ± 0.023.59 ± 0.02
      Protein
      Means were adjusted for average DIM, parity (primiparous or multiparous), and calving year. Each variable was affected by ABVHT quartile (P < 0.001). Season affected first 90-d fat percent (P < 0.001) and protein percent (P = 0.023) but not milk yield. There were no interactions between ABVHT quartile and season.
      (%)
       Summer3.06 ± 0.013.03 ± 0.023.02 ± 0.012.99 ± 0.01
       Winter3.09 ± 0.023.07 ± 0.023.06 ± 0.013.04 ± 0.01
      Pregnancy per AI at first service
      Means were adjusted for DIM at first service, parity (primiparous or multiparous), and calving year. There was no significant effect of ABVHT quartile or interaction between ABVHT quartile and season. Season did affect pregnancy per AI at first service (P < 0.001).
       Summer0.32 ± 0.020.36 ± 0.030.36 ± 0.020.35 ± 0.02
       Winter0.27 ± 0.030.32 ± 0.030.37 ± 0.020.36 ± 0.02
      1 n = 4,703 for milk production traits; n = 3,597 for pregnancy/first service.
      2 Genotyped cows were divided into quartiles with ABVHT1 = lowest ABVHT, most heat sensitive (<100), ABVHT2 (100 and 101), ABVHT3 (102–104) and ABVHT4 = highest ABVHT, most heat tolerant (>104).
      3 Means were adjusted for average DIM, parity (primiparous or multiparous), and calving year. Each variable was affected by ABVHT quartile (P < 0.001). Season affected first 90-d fat percent (P < 0.001) and protein percent (P = 0.023) but not milk yield. There were no interactions between ABVHT quartile and season.
      4 Means were adjusted for DIM at first service, parity (primiparous or multiparous), and calving year. There was no significant effect of ABVHT quartile or interaction between ABVHT quartile and season. Season did affect pregnancy per AI at first service (P < 0.001).
      Season affected first 90-d fat percent (P < 0.001) and protein percent (P = 0.023) but not first 90-d milk yield (Table 3). Fat percent and protein percent were higher in summer for all ABVHT quartiles. Season also affected first service pregnancy rate (P < 0.001). This latter effect resulted from the poor fertility of heat-sensitive cows during November and December (considered winter months in the analysis). Although the interaction was not significant, reduced fertility in winter was only apparent for ABVHT1 and ABVHT2. The reason for lower fertility in the winter is not known but could involve delayed effects of heat stress at time of calving. With an average voluntary waiting period of 74.3 d, the cows being bred for the first time in November and December calved in August and September, 2 very hot months for California's Central Valley. It has been shown that heat stress events can have delayed effects on pregnancy establishment through carryover effects on ovarian function (
      • Roth Z.
      • Meidan R.
      • Braw-Tal R.
      • Wolfenson D.
      Immediate and delayed effects of heat stress on follicular development and its association with plasma FSH and inhibin concentration in cows.
      ,
      • Roth Z.
      • Meidan R.
      • Shaham-Albalancy A.
      • Braw-Tal R.
      • Wolfenson D.
      Delayed effect of heat stress on steroid production in medium-sized and preovulatory bovine follicles.
      ;
      • de Torres Júnior, J.R.S.
      • Pires M.F.A.
      • de Sá W.F.
      • Ferreira A.M.
      • Viana J.H.M.
      • Camargo L.S.A.
      • Ramos A.A.
      • Folhadella I.M.
      • Polisseni J.
      • de Freitas C.
      • Clemente C.A.A.
      • de Sá Filho M.F.
      • Paula-Lopes F.F.
      • Baruselli P.S.
      Effect of maternal heat-stress on follicular growth and oocyte competence in Bos indicus cattle.
      ). The lack of large effects of season on production traits or fertility is probably a reflection of the specific environment that cows faced (high air temperature but low humidity) and the effectiveness of the farm's heat abatement strategy. Specific features of that strategy responsible for minimalizing seasonal variation cannot be determined after the fact. It is likely that providing access to lots open to the sky increased heat exchange via radiation (because the sky is cold compared with the barn). Vaginal temperature did not drop precipitously at night, perhaps because cows also increased eating behavior.

      CONCLUSIONS

      Cows with a high heat tolerance breeding value had lower body temperatures under heat stress than cows with a low heat tolerance value. Thus, at least some genetic variants controlling resistance to thermotolerance in Australia control resistance to thermotolerance under other conditions. Furthermore, the ABVHT can identify heat-tolerant cows with superior ability to regulate body temperature under US conditions. Further studies with nationwide data will give insight as to the value of ABVHT for minimizing effects of heat stress on production and reproduction traits for US breeding programs. Moreover, the likelihood of genotype × environmental interactions means that calculation of breeding values for heat tolerance using US data would potentially be of even greater value.

      ACKNOWLEDGMENTS

      The authors acknowledge Zoetis Animal Health (Kalamazoo, MI) and Mitchell Blanding for providing blank controlled internal drug releases, Zoetis Genetics for providing the genotypes, and Amelicor (Provo, UT) for providing access to DHI-Plus management software. Additionally, thanks are extended to DataGene (Melbourne, Australia) for calculating the ABVHT. Special thanks are also extended to Steve Maddox, Juan Garcia, Daniella Demetrio, and all personnel at Maddox Dairy (Riverdale, CA) for providing access to the cows, genotypes, and assistance in data collection. Research was supported by the Southeast Milk Checkoff Program and the L.E. “Red” Larson Endowment. L. M. Jensen is supported by the Doak Graduate Fellowship from the National Association of Animal Breeders. The authors have not stated any conflicts of interest.

      REFERENCES

        • Berman A.
        • Folman Y.
        • Kaim M.
        • Mamen M.
        • Herz Z.
        • Wolfenson D.
        • Arieli A.
        • Graber Y.
        Upper critical temperatures and forced ventilation effects for high-yielding dairy cows in a subtropical climate.
        J. Dairy Sci. 1985; 68 (4019887): 1488-1495
        • Bohmanova J.
        • Misztal I.
        • Cole J.B.
        Temperature-humidity indices as indicators of milk production losses due to heat stress.
        J. Dairy Sci. 2007; 90 (17369235): 1947-1956
        • Bohmanova J.
        • Misztal I.
        • Tsuruta S.
        • Norman H.D.
        • Lawlor T.J.
        National genetic evaluation of milk yield for heat tolerance of United States Holsteins.
        Interbull Bull. 2005; 0: 160-162
        • Collier R.J.
        • Dahl G.E.
        • Vanbaale M.J.
        Major advances associated with environmental effects on dairy cattle.
        J. Dairy Sci. 2006; 89 (16537957): 1244-1253
        • de Torres Júnior, J.R.S.
        • Pires M.F.A.
        • de Sá W.F.
        • Ferreira A.M.
        • Viana J.H.M.
        • Camargo L.S.A.
        • Ramos A.A.
        • Folhadella I.M.
        • Polisseni J.
        • de Freitas C.
        • Clemente C.A.A.
        • de Sá Filho M.F.
        • Paula-Lopes F.F.
        • Baruselli P.S.
        Effect of maternal heat-stress on follicular growth and oocyte competence in Bos indicus cattle.
        Theriogenology. 2008; 69 (17953981): 155-166
        • Dikmen S.
        • Cole J.B.
        • Null D.J.
        • Hansen P.J.
        Heritability of rectal temperature and genetic correlations with production and reproduction traits in dairy cattle.
        J. Dairy Sci. 2012; 95 (22612974): 3401-3405
        • Dikmen S.
        • Hansen P.J.
        Is the temperature-humidity index the best indicator of heat stress in lactating dairy cows in a subtropical environment?.
        J. Dairy Sci. 2009; 92 (19109269): 109-116
        • Dikmen S.
        • Khan F.A.
        • Huson H.J.
        • Sonstegard T.S.
        • Moss J.I.
        • Dahl G.E.
        • Hansen P.J.
        The SLICK hair locus derived from Senepol cattle confers thermotolerance to intensively managed lactating Holstein cows.
        J. Dairy Sci. 2014; 97 (24996281): 5508-5520
        • Garner J.B.
        • Douglas M.L.
        • Williams S.R.O.
        • Wales W.J.
        • Marett L.C.
        • Nguyen T.T.T.
        • Reich C.M.
        • Hayes B.J.
        Genomic selection improves heat tolerance in dairy cattle.
        Sci. Rep. 2016; 6 (27682591): 34114-34119
        • Guinn J.M.
        • Nolan D.T.
        • Krawczel P.D.
        • Petersson-Wolfe C.S.
        • Pighetti G.M.
        • Stone A.E.
        • Ward S.H.
        • Bewley J.M.
        • Costa J.H.C.
        Comparing dairy farm milk yield and components, somatic cell score, and reproductive performance among United States regions using summer to winter ratios.
        J. Dairy Sci. 2019; 102 (31606220): 11777-11785
        • Hammami H.
        • Bormann J.
        • M'hamdi N.
        • Montaldo H.H.
        • Gengler N.
        Evaluation of heat stress effects on production traits and somatic cell score of Holsteins in a temperate environment.
        J. Dairy Sci. 2013; 96 (23313002): 1844-1855
        • Hansen P.J.
        Effects of heat stress on mammalian reproduction.
        Philos. Trans. R. Soc. Lond. B Biol. Sci. 2009; 364 (19833646): 3341-3350
        • Hansen P.J.
        Heat stress and climate change.
        in: Moo-Young M. Comprehensive Biotechnology. 2nd ed. Vol. 4. Elsevier, 2011: 477-485
        • Hansen P.J.
        Reproductive physiology of the heat-stressed dairy cow: Implications for fertility and assisted reproduction.
        Anim. Reprod. 2019; 16 (32435293): 497-507
        • Hayes B.J.
        • Bowman P.J.
        • Chamberlain A.J.
        • Savin K.
        • van Tassell C.P.
        • Sonstegard T.S.
        • Goddard M.E.
        A validated genome wide association study to breed cattle adapted to an environment altered by climate change.
        PLoS One. 2009; 4 (19688089): e676
        • Interbull
        MACE August 2021 for Production.
        https://interbull.org/static/web/prod2108r.pdf
        Date: 2021
        Date accessed: August 17, 2021
      1. IPCC. 2014. Clim. Change. 2014 (Synthesis Report.).

        • Mesgaran S.D.
        • Eggert A.
        • Höckels P.
        • Derno M.
        • Kuhla B.
        The use of milk Fourier transform mid-infrared spectra and milk yield to estimate heat production as a measure of efficiency of dairy cows.
        J. Anim. Sci. Biotechnol. 2020; 11 (32399210): 43
        • Nguyen T.T.T.
        • Bowman P.J.
        • Haile-mariam M.
        • Nieuwhof G.J.
        • Hayes B.J.
        • Pryce J.E.
        Short communication: Implementation of a breeding value for heat tolerance in Australian dairy cattle.
        J. Dairy Sci. 2017; 100 (28711268): 7362-7367
        • Nguyen T.T.T.
        • Bowman P.J.
        • Haile-Mariam M.
        • Pryce J.E.
        • Hayes B.J.
        Genomic selection for tolerance to heat stress in Australian dairy cattle.
        J. Dairy Sci. 2016; 99 (27037467): 2849-2862
        • Nieuwhof G.J.
        • Beard K.T.
        • Konstantinov K.V.
        • Bowman P.J.
        • Hayes B.J.
        Implementation of genomics in Australia.
        Interbull Bull. 2010; 42: 35
        • NRC
        A Guide to Environmental Research on Animals.
        Natl. Acad. Sci., 1971
        • Ravagnolo O.
        • Misztal I.
        Genetic component of heat stress in dairy cattle, parameter estimation.
        J. Dairy Sci. 2000; 83 (11003247): 2126-2130
        • Roth Z.
        • Meidan R.
        • Shaham-Albalancy A.
        • Braw-Tal R.
        • Wolfenson D.
        Delayed effect of heat stress on steroid production in medium-sized and preovulatory bovine follicles.
        Reproduction. 2001; 121 (11427162): 745-751
        • Roth Z.
        • Meidan R.
        • Braw-Tal R.
        • Wolfenson D.
        Immediate and delayed effects of heat stress on follicular development and its association with plasma FSH and inhibin concentration in cows.
        Reproduction. 2000; 120 (11006149): 83-90
        • Sánchez J.P.
        • Misztal I.
        • Aguilar I.
        • Zumbach B.
        • Rekaya R.
        Genetic determination of the onset of heat stress on daily milk production in the US Holstein cattle.
        J. Dairy Sci. 2009; 92 (19620687): 4035-4045