It is well established that heat stress has an adverse effect on dairy cows that manifests itself through behavioral change and decline in performance (West, Reference West2003; Honig et al., Reference Honig, Miron, Lehrer, Jackoby, Zachut, Zinou, Portnick and Moallem2012). Lactating cows are particularly susceptible to heat stress, due to the high metabolic heat production associated with increased milk production (West, Reference West2003). The Israeli herd data book (2015) shows a drop of approximately 20% in conception rates between winter and summer, which indicates a major pitfall in the successful continuation and profitability of the dairy cattle herd. In order to relieve heat stress dairy farmers implement various management tactics such as environment modification, mainly by various means of cooling (Flamenbaum et al., Reference Flamenbaum, Wolfenson, Mamen and Berman1986; Her et al., Reference Her, Wolfenson, Flamenbaum, Folman, Kaim and Berman1988; Flamenbaum and Galon, Reference Flamenbaum and Galon2010). In Israel, the widely used cooling system is based on sessions comprised of direct watering of the cows, followed by forced air ventilation.
During heat load cows exhibit reduced feed intake in order to reduce metabolic heat (West, Reference West2003). In many mammals, including dairy cows, reduced feed intake is followed by a negative energy balance which leads to reduced leptin secretion (Liefers et al., Reference Liefers, Veerkamp, te Pas, Delavaud, Chilliard and van der Lende2003) and increased ghrelin levels (Bradford and Allen, Reference Bradford and Allen2008; Muccioli et al., Reference Muccioli, Lorenzi, Lorenzi, Ghe, Arnoletti, Raso, Castellucci, Gualillo and Meli2011). Leptin is predominantly synthesized in adipose tissue and is a protein hormone with important effects on eating behavior, energy expenditure, and body weight (Budak et al., Reference Budak, Fernandez Sanchez, Bellver, Cervero, Simon and Pellicer2006; de la Hoya, Reference de la Hoya MPGTR, Contreras, Diaz, Saucedo and Saucedo2015). Leptin serves as a signal of body energy status to the brain, whereas it acts on the hypothalamus to inhibit food intake (Budak et al., Reference Budak, Fernandez Sanchez, Bellver, Cervero, Simon and Pellicer2006; de la Hoya, Reference de la Hoya MPGTR, Contreras, Diaz, Saucedo and Saucedo2015).
Ghrelin is a gut—brain peptide composed of 27 amino acid in ruminants (Dickin et al., Reference Dickin, Thue and Buchanan2004). In cattle it is synthesized by abomasal and ruminal tissues (Hayashida et al., Reference Hayashida, Murakami, Mogi, Nishihara, Nakazato, Mondal, Horii, Kojima, Kangawa and Murakami2001; Gentry, Reference Gentry, Willey and Collier2003). Ghrelin has two known isoforms, des-acyl ghrelin and acylated ghrelin that binds to the growth hormone secretagogue receptor (GHS-R1A) (Kojima et al., Reference Kojima, Hosoda, Date, Nakazato, Matsuo and Kangawa1999; van der Lely et al., Reference van der Lely, Tschop, Heiman and Ghigo2004; Fernandez-Fernandez et al., Reference Fernandez-Fernandez, Martini, Navarro, Castellano, Dieguez, Aguilar, Pinilla and Tena-Sempere2006). Acyl ghrelin is known to be secreted in an oscillatory manner, concentrations increase prior to scheduled meals and in response to fasting while feeding suppresses its secretion (Hayashida et al., Reference Hayashida, Murakami, Mogi, Nishihara, Nakazato, Mondal, Horii, Kojima, Kangawa and Murakami2001; Miura et al., Reference Miura, Tsuchiya, Sasaki, Kikuchi, Kojima, Kangawa, Hasegawa and Ohnami2004; Wertz-Lutz et al., Reference Wertz-Lutz, Knight, Pritchard, Daniel, Clapper, Smart, Trenkle and Beitz2006). Interestingly, ad libitum feeding in ruminants diminishes this oscillation (Sugino et al., Reference Sugino, Yamaura, Yamagishi, Ogura, Hayashi, Kurose, Kojima, Kangawa, Hasegawa and Terashima2002; Borner et al., Reference Borner, Derno, Hacke, Kautzsch, Schaff, Thanthan, Kuwayama, Hammon, Rontgen, Weikard, Kuhn, Tuchscherer and Kuhla2013). These observations suggest that acyl ghrelin release is affected by serum modification of nutritional factors and function as circulating signal for energy insufficiency. Previously, we found that acyl ghrelin levels measured during heat stress in summer were lower than those in the winter (Honig et al., Reference Honig, Ofer, Elbaz, Kaim, Shinder and Gershon2016). Des-acyl ghrelin circulates at much greater concentrations than acylated ghrelin (Hosoda et al., Reference Hosoda, Kojima, Matsuo and Kangawa2000; ThidarMyint et al., Reference ThidarMyint, Yoshida, Ito and Kuwayama2006), and seems to be less studied. Its effects on food intake and metabolism are less characterized.
Little is known about ghrelin isoforms and leptin levels under conditions of heat stress in dairy cows. In this study, we demonstrate that increasing the number of cooling sessions from 5 a day to extensive cooling management comprised of 8 cooling sessions a day, during heat stress in the summer, significantly elevated acyl ghrelin and leptin levels whereas total ghrelin levels decreased in dairy cows.
Materials and methods
Cows and treatments
All experiments were approved by the Agricultural Research Organisation Animal Care Committee. The experiments were conducted at the Agricultural Research Organisation experimental farm at Bet Dagan, Israel, during the summer (August). 40 Israeli-Holstein dairy cows were housed in covered loose pens with an adjacent outdoor yard, that were equipped with a real-time electronic individual feeding system. Cows were fed a typical Israeli lactating-cow ration made up of 1.78 Mcal NEL, 16.5% CP, and 29.8% NDF. The cows were allocated to two treatment groups subjected to different cooling schedules carried out in the holding area of the milking parlor. One group was subjected to the ordinary cooling management preformed in Bet Dagan dairy farm comprised of 5 cooling sessions per day (5 CS). Cooling was carried out at 0415, 0945, 1215, 1645, and 1945 h. The second group was subjected to extensive cooling comprised of 8 cooling sessions per day (8 CS). In this case cooling was carried out at 0415, 0645, 0945, 1245, 1445, 1645, 1945, and 2230 h. Both groups were subjected to a cooling session before each milking. During the day, between milkings, the 5 CS cows were brought twice and the 8 CS cows were brought 5 times for further cooling sessions. Each cooling session lasted 45 min, comprised of repeated cycles of 60 s of showering and 4.0 min of only forced ventilation.
Cows were milked three times a day (0500, 1300, 2030 h). Milk yields were recorded electronically at each milking and cow weight was recorded automatically after each milking, with a walk-in electronic scale (S.A.E. Afikim, Kibbutz Afikim, Israel).
During the experiment period the cows dry matter intake (DMI), metabolic hormones blood levels, vaginal temperature, digestibility and rumination, activity and lying time were measured as described in the online Supplementary File.
Blood collection and handling
Blood samples were collected from 20 multiparous cows (10 cows from each group), every 2 to 3 d until the end of the experiment. The collection occurred at the time of food distribution approximately at 1030 h, and plasma was extracted after centrifuging at 10 000 rpm for 10 min. 1 N HCl and 0.5 mg PMSF (Sigma, Rehovot, Israel) dissolved in iso-propanol were immediately added to plasma samples designated for acyl ghrelin measures. Samples were stored at −80°C until further analysis until determinations of acyl and total ghrelin as well as leptin blood levels as described in the online Supplementary File.
Statistical analysis
The data were analyzed by two-way analysis of variance (ANOVA) with Cooling [5 or 8 sessions], periods [week of treatment], and their interaction as main fixed effects and with repeated-measures in one factor (period, also called mixed-model ANOVA). Each cow served as its own control because the hormone parameters for each cow were measured at all time periods. The Tukey—Kramer honest significant difference test was used to test the separation of the means, in comparing the periods within each cooling regime. These statistical analyses were conducted with JMP software (SAS Institute). All results are presented as mean ± standard error of the mean.
Results
Characterization of the cows at the two cooling management groups
The descriptive statistics of days in milking, dry matter intake (DMI), lactation number, age, milk yield, and weight, content of milk fat, protein and lactose between the two different cooling methods are presented in Table 1. No between-group differences in days in milking, DMI, milk yield, milk fat, protein and lactose contents were observed, but ECM was significantly higher in the cows exposed to 8 CS as compared to the cows from the 5 CS group (P < 0.05; Table 1).
Mean, sem and P-value of days in milking, dry matter intake (DMI), milk yield, energy corrected milk (ECM), content of milk fat, protein and lactose of all cows under different cooling managements
8 cooling sessions per day (8 CS), 5 cooling sessions per day (5 CS). * P < 0.01, ** P < 0.001, *** P < 0.0001.
Measuring the vaginal temperature of the cows at the two cooling management groups
First we calculated the THI index of the environment in order to approve the heat conditions in the area (online Supplementary File Fig. S1). Next, we measured the vaginal temperature of the cows in order to examine the effectiveness of the 8 cooling sessions as compared to the 5 CS. The temperature was measured during the second and fourth weeks of the experiment, for the duration of approximately 3 d each time. The cumulative number of hours that the body temperature of cows in the 5 CS group was above 39.4°C heat-stress threshold (Burfeind et al., Reference Burfeind, Suthar and Heuwieser2012) was significantly higher as compared to the 8 CS group (online Supplementary File Table S1) at all measurements.
Measuring the DMI of the cows at the two cooling management groups
We compared DMI in cows exposed to 5 or 8 cooling sessions per day and found that the DMI in the group of 8 CS was elevated every week of the experiment, while in the group of the 5 CS only one elevation in the DMI intake was observed between week 2 and 3 (Fig. 1 and Table 2). In the following weeks no changes in the DMI was observed (Fig. 1 and Table 2). No significant difference was detected between the DMI of the 8 CS group and the DMI measured in the 5 CS treated cows (Table 2). In addition, no correlation was found between the DMI and the weeks (Table 2).
Metabolic hormones levels and DMI throughout the study period in 5 CS vs. 8 CS. Mean ± sem of acyl ghrelin, Total ghrelin, Leptin levels and DMI through the study period of multiparous cows submitted to 8 cooling sessions per day (8 CS) and multiparous cows submitted to 5 cooling sessions per day (5 CS). *P < 0.05 between 8 CS and 5 CS treatments.
Least Square Mean ± Standard error of acyl ghrelin, Total ghrelin, Leptin levels and DMI at interaction between coolings and week of the experiment, of all cows under different cooling managements
** Levels not connected by same letter (A, B, C, D,E) are significantly different between interactions within a week.
Acyl ghrelin
We compared blood levels of acetylated ghrelin during the weeks of the experiment in cows exposed to 5 or 8 cooling sessions per day. In both groups there is a significant rise in the hormone levels (Fig. 1 and Table 2). At the last 3 weeks of the experiment acyl ghrelin levels were significantly higher in 8 CS group compared to the cows in the 5 CS group, which was also reflected in the overall means (Table 1). No correlation was found between the DMI and the weeks (Table 2).
Total ghrelin
We further compared blood levels of total ghrelin in cows exposed to 5 or 8 cooling sessions per day. We found that the total ghrelin levels increased in both group throughout the experiment (Fig. 1) with significant higher levels of total ghrelin in 5 CS group (Fig. 1 and Table 2). Overall, total ghrelin levels were significantly higher in 5 CS group compared to the cows in the 8 CS group (Table 1). Furthermore, a significant correlation was found between the DMI and the weeks (Table 2).
Leptin
Leptin levels were significantly higher in 8 CS group during the whole experiment as compared to the 5 CS group (Fig. 1 and Tables 1 and 2). Leptin levels were significantly elevated each week in both treatment groups (Fig. 1 and Tables 2). No correlation was found between the DMI and the weeks (Table 2).
Welfare parameters of multiparous cows subjected to 8 CS and 5 CS
The welfare parameters rumination, lying time and activity are presented in Table 3. In all these parameters there was a significant difference between groups. The 8 CS group cows ruminated and lay down more time during a day and simultaneously had better activity time. Furthermore, the DM digestibility percentage was numerically higher in the 8 CS group as compared to the 5 CS cows (P = 0.07, Table 3). Finally, no between-group differences in digestibility were observed (0.671 ± 0.004 at 5 CS compare to 0.674 ± 0.003 pv 0.25).
Mean, sem and P-value of welfare parameters: rumination, lying time, activity and DM digestibility, of the synchronized multiparous cows under different cooling managements
8 cooling sessions per day (8 CS), 5 cooling sessions per day (5 CS). ** P < 0.001.
Discussion
The regulation of nutritional state occurs at the hypothalamus. Such a complex regulation requires the activity of more than one regulator. One such regulator with a pivotal role in the regulation of food intake is ghrelin. There are two known isoforms of ghrelin, des-acyl ghrelin and acylated ghrelin (Gutierrez et al., Reference Gutierrez, Solenberg, Perkins, Willency, Knierman, Jin, Witcher, Luo, Onyia and Hale2008; Yang et al., Reference Yang, Brown, Liang, Grishin and Goldstein2008). In our study, both groups were in ad libitum feeding during heat load conditions. Although the cows were not in scheduled or restricted food conditions, we speculate that as result of scheduled fresh food distribution in the morning, especially after milking and cooling, acyl ghrelin level in the time around food distribution would rise in a resemble manner to scheduled meal. Thus the blood samples were collected during the cows morning meal time. We found that in synchronized multiparous cows from the 8 CS group acyl ghrelin concentration was higher. Since acyl ghrelin is associated with food intake, we examined the DMI. In our study, the 8 CS group consumed significantly more food. This result is in agreement with other studies that showed an association between high levels of acyl ghrelin and increased food consumption (Wertz-Lutz et al., Reference Wertz-Lutz, Knight, Pritchard, Daniel, Clapper, Smart, Trenkle and Beitz2006; Foote et al., Reference Foote, Hales, Lents and Freetly2014).
In similarity to the results presented here, our previous study (Honig et al., Reference Honig, Ofer, Elbaz, Kaim, Shinder and Gershon2016) that compared levels of acyl ghrelin in the different seasons demonstrated that acyl ghrelin levels in winter were higher compared to summer. Combining this data with the results presented here, allows us to infer that 8 CS promotes higher physiological levels of acyl ghrelin and improved coping ability for the deleterious heat stress effects. The combined results of our studies further support the notion that intensive cooling management mimics at least in part the winter condition and can relieve heat stress during the hot season.
Compare to acyl ghrelin, des-acyl ghrelin seems to be less studied and its effects on food intake and metabolism are less characterized. Primary studies in cattle reported a lower rumen pH in heat stressed dairy cows (Collier et al., Reference Collier, Beede, Thatcher, Israel and Wilcox1982). Interestingly, in our study, coincidentally to higher acyl ghrelin levels, 8 CS had lower levels of total ghrelin, which indicates lower levels of des-acyl ghrelin. In that respect, we surmise that as a consequence of low rumination time, due to heat stress conditions, less saliva would buffer the PH in the rumen and it may relate to the levels of des-acyl ghrelin in the circulation.
In our study we demonstrate that 5 CS group spent more time in heat stress conditions and it is evident that they spent less time ruminating, their des-acyl ghrelin level were higher and they consumed less food. Therefore, it can be speculated that there is a negative link between levels of des-acyl ghrelin and food consumption, in resemblance to studies shown in rodents (Asakawa et al., Reference Asakawa, Inui, Fujimiya, Sakamaki, Shinfuku, Ueta, Meguid and Kasuga2005; Chen et al., Reference Chen, Inui, Asakawa, Fujino, Kato, Chen, Ueno and Fujimiya2005). Further studies should be conducted in order to understand whether the lower levels of total ghrelin in 8 CS is due to higher acylation rate toward acyl ghrelin or due to reduction in the production of des-acyl ghrelin itself, or both.
Previous studies suggested leptin as a candidate hormone that operates in collaboration with ghrelin. Leptin is an important mediator in eating behavior, energy expenditure, and body weight. Our results demonstrate that in synchronized multiparous cows from 8 CS group, leptin concentration was about 1.4 times higher compare to 5 CS, suggesting that the 8 CS cows were in better nutritional state. This notion is further supported by the acyl-ghrelin levels measured in this study and described above.
Heat stress further disrupts the productivity and welfare of dairy cows. At heat load state cows reduce DMI, rumination time, activity and lying time together with longer standing time and panting (Silanikove, Reference Silanikove2000; De Rensis and Scaramuzzi, Reference De Rensis and Scaramuzzi2003; West, Reference West2003; Berman, Reference Berman2006; Allen et al., Reference Allen, Hall, Collier and Smith2015; Moretti et al., Reference Moretti, Biffani, Chessa and Bozzi2017). In overview it seems that the extra cooling sessions in the 8 CS group positively affected the welfare of the cows. In the 8 CS group the accumulated time that the vaginal temperature measured exceeded a critical temperature of 39.4°C was shorter, the cows spend more time lying down, ruminating, walking around the pen and consumed more food.
Supporting our results, a previous study also compared five to eight cooling sessions, and found similar differences between the two cooling regimes for rumination time, lying time and ECM yield (Honig et al., Reference Honig, Miron, Lehrer, Jackoby, Zachut, Zinou, Portnick and Moallem2012). However, these authors reported higher milk yield and DMI of cows receiving the 8 coolings regime, whereas in the current study, no significant differences were found in milk production and DMI between the two groups. This difference in results between the two studies may be due to the different facilities used: large fans were added into the barn used in the present experiment and the duration of Honig's study was longer and under higher THI conditions.
In conclusion, we have shown that adding three extra cooling sessions during the day helps to mitigate the detrimental effect of heat stress on ghrelin and leptin levels, hormones with paramount importance to food intake and metabolism. Adding extra cooling also improves welfare and production. From this study it can be speculated that 8 CS cooling can assist in reducing the negative effects of heat stress on lactating dairy cows, without harming the level of existing production and even improving it to some extent, taking into account the welfare of the cow.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029919001055
Acknowledgements
The authors would like to thank the team at the Volcani Center's experimental dairy farm (Bet Dagan, Israel) for their assistance with animal care. This research was financially supported by the Israeli Milk Board.
