How to use recombinant bovine somatotropin in crossbred Holstein × Gyr (3/4 and 7/8) cows?

The aim of this study was to evaluate the effects of dose, start time, and application interval of recombinant bovine somatotropin (rbST) on performance, health, and reproduction parameters of crossbred Holstein × Gyr dairy cows (3/4 and 7/8). A 2 × 2 × 2 factorial study was designed to test the effects of application dose (250 mg rbST or 500 mg rbST), start of application (40 or 63 days in milk (DIM)), and application interval (12 or 14 days); 180 crossbred Holstein × Gyr cows were enrolled. Treatments with 500 mg rbST resulted in increased milk production, 3.5% fat-corrected milk (FCM), and energy-corrected milk (ECM). In the factorial design analysis, greater FCM, ECM, and milk solid production (kg/day) were observed when 500 mg rbST was started at 63 DIM, while 250 mg rbST resulted in greater milk production (kg/day), FCM, ECM, and milk solids (kg/day) when administered at 12-day intervals. Administration of 500 mg rbST resulted in an increase of days open and number of services per pregnancy ( P = 0.01). Holstein × Gyr cows (3/4 and 7/8) presented a superior productive response to rbST when treated with 500 mg starting at 63


Introduction
Recombinant bovine somatotropin (rbST) is an important tool for improving animal productivity. However, the major knowledge about rbST results from research carried out on Holstein cows. In these animals, rbST promotes an increase in milk production of 10 to 15% and improves lactation persistency in response to homeorhetic changes (Bauman, 1992;Bauman, 1999, St.-Pierre et al., 2014. Treatment recommendations are based on subcutaneous application of rbST (500 mg) from the 9th week after parturition (63 days in milk (DIM)), at 14-day intervals, for cows that present adequate body condition score. This recommendation was postulated, in part, due to results indicating enhanced productive performance associated with improved economic response (Chilliard, 1988;Bauman and Vernon, 1993).
Cattle of different genetic compositions, such as Bos taurus × Bos indicus crosses, have been treated with rbST at dose and application intervals recommended for Holstein cows. Nevertheless, Oliveira et al. (2007) and Pereira et al. (2016) demonstrated that the lactation curve of crossbred (Holstein × Gyr) is different from that of Holstein cows, presenting smaller and earlier peak yield, approximately at 30 DIM (Pereira et al., 2016). This could directly influence the response to rbST by animals of different genetic compositions.
Few studies have evaluated crossbred cows' productive responses to treatments with rbST dose smaller than 500 mg (Cisse et al., 1996;Phipps et al., 1997;Fontes et al., 1997;Azizan and Phipps, 1999). Among these, only Fontes et al. (1997) evaluated the effects of treatment start earlier than the 9th week of lactation, reporting improved milk This article belongs to the Topical Collection: Dairy Science and Health in the Tropics yield when treatment was carried from 56 to 100 DIM. Glória et al. (2010) did not observe differences between application intervals of 12 or 14 days, for a 500-mg dose in Holstein × Gyr, but a large variation in genetic composition of these crossbred cows might have influenced results. Given the crossbred cows' lactation curve characteristics, an earlier start of rbST treatment with dosage and application interval adjustments could be justifiable. Thus, investigation of additional rbST application protocols, as well as their effects on performance and health parameters, is needed to establish effective treatment practices. The objectives of this study were to evaluate performance, health, and reproductive parameters of crossbred dairy cows (3/4 and 7/8 Holstein × Gyr) treated with rbST at different doses, start times, and application intervals.

Cows, treatments, and experimental management
All experimental procedures were approved by the Ethics Committee in Animal Experimentation of the Universidade Federal de Minas Gerais-CETEA/UFMG (230/2008).
One hundred and eighty (63 primiparous and 117 multiparous) Holstein × Gyr cows (3/4 and 7/8) were selected for the study based on body condition score (BCS: 2.75 to 3.5, according to the scale by Wildman et al., 1982), DIM (40 ± 3.5 and 63 ± 3.5 days), and age (mean of 2.5 years for primiparous and 5 years for multiparous). After calving, all cows were clinically evaluated and those that presented dystocia and mastitis were excluded from the study.
During the trial period, the cows were managed according to their production into primiparous, high yield, intermediate yield, and low yield, producing 19 to 23 kg of milk/day, 28 to 34 kg/day, 21 to 26 kg/day, and 15 kg/day, respectively. Animals were blocked by parity and birth (20 blocks) and homogeneously distributed according to genetic composition, parity, and production in the previous lactation in a 2 × 2 × 2 factorial design: application dose × treatment start × application interval, and control (CON) (Fig. 1). Treated animals received a subcutaneous injection of rbST (Boostin® 500 mg rbST or Hilac-250® 250 mg rbST, MSD Animal Health, São Paulo, Brazil), starting at 40 or 63 DIM, and repeated at regular intervals of 12 or 14 days up to 250 DIM. The total number of injections was 17 and 15 for those that received the first dose of rbST at 40 DIM every 12 and 14 days, respectively, while those starting at 63 DIM received 15 and 13 injections at 12-and 14-day intervals, respectively. All rbST injections were applied after morning milking. No applications of rbST or placebo were given to CON cows.

Feeding management
From May to September 2009, the animals were confined in paddocks where they received a total mixed ration twice a day. From October to April of 2010, the animals were kept in irrigated paddocks (Cynodon dactylon) and supplemented with concentrate twice a day.

Live body weight and body condition score
Body weight (BW) and BCS were measured every 30 days, from the start day of rbST treatment to 14 days after the last treatment. An electronic body weighing scale (EziWeigh7, Tru-Test®, Rio Grande do Sul, Brazil) was used to measure BW after morning milking, and BCS was assessed during milking by a single examiner according to Wildman et al. (1982).

Mammary gland health and reproduction
Cows were evaluated at each milking as to the occurrence of clinical mastitis by a team trained to visualize and identify clinical cases according to intensity (Pinzón-Sánchez et al., 2011). Each mammary quarter affected was considered as a clinical case of mastitis. Within mammary quarters, a clinical case was considered a new event if occurring at least 14 days after the end of the symptoms of the previous case.
Cows were artificially inseminated (AI) following visual heat detection by a trained employee. Heat detection was carried out twice a day (0800 and 1730 h). Heat and AI dates were registered daily for calculation of days open days, number of services per conception, and conception rate.

Nutritional composition analyses
Diet samples were collected monthly and stored in plastic bags in a cold room at − 20 °C. Samples were pre-dried in a forced air oven at 55 °C for 72 h for dry matter evaluation. After this procedure, samples were milled at 1 mm (Arthur H. Thomas, Philadelphia, EUA) and evaluated for crude protein (

Statistical analyses
Data were analyzed using SAS software (SAS Institute Inc., Cary, NC) (SAS, 1996). Data with repeated measures over time (milk production and composition, BW, BCS, SCC) were grouped in 30-day intervals, adjusted by DIM between 70 and 250 days in 7 evaluation points. The SCC data was transformed to Log10 for analyses.
Statistical analyses were performed in two steps. First, the effect of dose (0, 250, and 500 mg of rbST) was tested, where parametric variables with repeated measures over time (milk production and composition, lactation persistency, BW, and SCC) were analyzed using the mixed linear model (PROC MIXED from SAS), with the inclusion of the block (n = 20 blocks) as a random effect, and treatment (0, 250, 500 mg of rbST), DIM (70,100,130,160,190,220,and 250 DIM), and interactions as fixed effects. Milk production at 20 to 40 DIM and initial BW were used as covariates. All models were verified for normality and homoscedasticity of residuals and tested with the Shapiro-Wilk and Bartlett tests. Effects on BCS were analyzed using the Friedman test (NPAR1WAY).
Further analyses were performed with exclusion of CON treatment and the data were analyzed as a 2 × 2 × 2 factorial design on the mixed linear model (PROC MIXED from SAS). Block (n = 20 blocks) was included as a random effect, whereas dose (250 and 500 mg), treatment start (40 and 63 days), application interval (12 and 14 days), DIM (70,100,130,160,190,220,and 250), and their interactions were included as fixed effects. Clinical mastitis incidence was analyzed using a chi-square test corrected by Fisher's exact test (PROC FREQ). Means were considered different when P ≤ 0.05, and trends were considered when 0.05 < P ≤ 0.10.

Milk production
Average milk yield of trial period was 19.9, 19.7, and 21.3 kg/day for CON, rbST 250, and rbST 500 treatments, respectively (Table 2). Milk production was greater for rbST 500 at 100, 130, 160, and 190 DIM (P < 0.05), with a tendency to be greater also at 70 and 220 DIM (P = 0.08) ( Table 2), compared with CON. There was an increase of 1.4 kg/day (7.3%) in milk production for the rbST 500, with the maximum response (2.0 kg of milk per day; 8.7% increase compared with CON was observed at 100 DIM (fourth rbST application). Cows treated with rbST 250 did not increase milk yield compared with CON cows (Table 2). In addition, rbST 500-treated cows had greater response in FCM production at 100 and 130 DIM (P = 0.03 and P = 0.04, respectively) and greater ECM at 100 DIM (P = 0.02), with a tendency to be greater also at 160 DIM (P = 0.09).
Milk yield was greater for cows treated with 500 mg rbST compared with 250 mg, and for those receiving 250 mg at 12-day intervals compared with 250 mg at 14-day intervals, with effects of dose and dose × application interval on milk production (P = 0.01) (Fig. 2). Dose, dose × start time, and dose × application interval influenced FMC and ECM (P < 0.05) ( Table 3). When the first application of rbST 500 was performed at 63 DIM, FCM and ECM were greater.

Milk composition
Fat, lactose, and total solids (kg/day) were affected by dose (P ≤ 0.03), dose × application interval interaction (P ≤ 0.05), and tended to be affected by dose × start time interaction (P ≤ 0.09), while protein content (kg/day) was affected by dose (P = 0.01) and dose × application interval interaction (P = 0.04).
Therefore, rbST 250 promoted increased production of fat, protein, lactose, and total solids when administered at 12-day intervals, while rbST 500 tended to increase milk component content when started at 63 DIM, regardless of application interval.

Lactation persistency
The rate of milk production decline throughout lactation was equal for rbST 250 and control; however, it was greater for rbST 250 than for rbST 500 (− 0.061 vs. 0.050, respectively; P = 0.02). There was no effect of start time, application intervals, or interactions on lactation persistency. Milk production (

Live weight and body condition score
There was no difference in BW and BCS between treatments and in the factorial design analysis. Mean BCS during the trial was 3.23.

Somatic cell count and mastitis
There was an effect of dose × start time interaction. Clinical mastitis incidence was greater for cows treated with rbST 500 starting at 40 DIM when compared to rbST 250 starting at the same time (17 vs. 8 clinical mastitis cases; P = 0.05).

Reproduction
Open days (171.8 vs. 129.9; P = 0.01) and number of services per pregnancy (3.11 vs. 2.21; P = 0.01) were greater for rbST 500 (dose effect P = 0.02). There was no difference in first service conception rate.

Discussion
To our knowledge, this is the first study that evaluated the effects of rbST effects on productive performance, health, and reproductive parameters of crossbred Holstein × Gyr cows (3/4 and 7/8) combining different dosages (250 and 500 mg), start time (40 or 63 DIM), and application intervals (12 or 14 days). The greater milk production for rbST 500-treated cows and the lack of such response to rbST 250 compared with control suggest that the 250 mg rbST dose was not able to provide the minimum daily concentration necessary to induce increased production. Initial studies with daily administration of rbST observed a dose-dependent response where milk production increase was proportional to rbST increase until the formation of a plateau (Bauman et al., 1985;Hartnell et al., 1991).
Assuming that rbST is released uniformly and completely, the daily dose of rbST used in the present study would be 17.9, 20.8, 35.7, and 41.7 mg/day for treatments 250/14, 250/12, 500/14, and 500/12 (dose × application interval), respectively. The difference in the daily concentration of rbST between rbST 250 at 12-and 14-day intervals (approximately 3 mg/day) resulted in an increase of 2.3 kg of milk/cow/day, while the increase in rbST concentration from 20.8 mg/day (250/12) to 35.7 mg/day (500/14) resulted in an increment of 1.3 kg. This smaller increment and the absence of further increase with the daily dose of 41.7 mg/day (500/12) suggest the formation of a plateau from the dose of 35.7 mg/day for these cows. This response indicates that an optimal rbST dose for crossbred Holstein × Gyr cows (3/4 and 7/8) is 20.8 and 35.7 mg/day. Further investigation is needed to accurately determine the ideal daily dose for crossbred animals.
On the other hand, assuming a lack of uniformity in the release of somatotropin in slow-release formulations, it is possible that at given times, the daily rbST concentration was null or very low. Morais et al. (2017) reported an increase in milk production from the 2nd to the 11th days after application of rbST-vitamin E, when rbST was performed every 14 days, suggesting absence or lower concentration of rbST in the last 3 days of the application cycle. Thus, the lower milk production observed for the 250/14 treatment (dose × application interval) could be due to periods with null or very low rbST concentrations leading to lack of effects on productivity.
In the present study, the increase in milk production for rbST 500 (7.3% or 1.4 kg cow/day compared with CON) was smaller than reported elsewhere (10 to 15% increase, equivalent to 2 to 6 kg cow/day, on average) (Bauman, 1999;Chilliard et al., 2002;St.-Pierre et al., 2014). Most studies were conducted with Holstein cows, which present greater productive capacity compared to Gyr crossbred animals. Cows with greater genetic merit for milk production naturally have greater endogenous somatotropin concentration (Bauman, 1999), which positively regulates hepatic GH receptors' concentration (Gluckman et al., 1987). Given the results of the present study, this could explain the lack of productive response to rbST 250, and the smaller yield increase observed for rbST 500, since the enrolled crossbred cows are of lesser genetic merit and possibly have lesser availability of GH receptors. Such as observed in this study, Fontes et al. (1997) and Phipps et al. (1997) did not observe an increase in milk production by crossbred cows (Bos taurus × Bos indicus) when dosage changed from 250 to 500 mg rbST and from 334 to 500 mg rbST.
In addition to inducing greater milk production, rbST 500 resulted in greater FCM and ECM. Similar results were observed by Fike et al. (2002) that reported an increase in milk production and 4% fat-corrected milk of 9 and 12%, respectively, by Holstein cows under grazing conditions.
The start period of rbST treatment in Holstein cows is directly related to the successful response to somatotropin. During early lactation, cows are subjected to negative energy balance and have a lower number of GHR-1A receptors, which contributes to a reduction in circulating concentrations of IGF-1 and lower milk production responses (Radcliff et al., 2006). In this study, treatments were initiated after the milk production peak of crossbred Holstein × Gyr (3/4 and 7/8), which occurs around 30 DIM (Pereira et al., 2016). After this moment, the GHR-1A hepatic expression returns to prepartum values, and the animals are able to respond to treatment (Radcliff et al., 2006). Thus, for both start times tested in this study, the animals were in similar physiological periods.
There was no change in the levels of fat, protein, lactose, and total solids between treatments. This is possibly due the fact that the animals were in a positive nutrient balance, as there were also no differences in BW and BCS between treatments. The use of rbST does not change the nutritional composition of milk in positive nutrient balance cows (Tarazon-Herrera et al., 2000e Vicini et al., 2008. The synthesis of milk constituents, however, increases as a function of the production volume, without changing its contents, which proves the homeorrhetic effects of rbST (Bauman, 1999).
In a meta-analysis regarding the effects of rbST in cattle, St-Pierre et al. (2014) found no difference in BCS in treated animals similarly to results of this study. Morais et al. (2017) reported an increase in BCS of less than 0.5 (scale 1 to 5) in treated animals every 14 days, and this difference was considered clinically irrelevant.
In Holstein cows, the use of rbST is responsible for increasing milk production, in addition to providing an increase in lactation persistency (Bauman, 1992;Van Amburgh et al., 1997;Morais et al., 2017). Lactation persistency of crossbred cows in this study did not increase with rbST treatments, possibly due to the genetic composition of animals, since crossbred Holstein × Gyr has a lower and earlier lactation peak, lower milk production, and less lactation persistency than animals of the Holstein breed (Fontes et al., 1997;Glória et al., 2010;Pereira et al., 2016).
As in the present work, most studies did not report direct effects of rbST on SCC (Collier et al., 2001;St.-Pierre et al., 2014;Morais et al. 2017). In contrast, Bauman (1999) observed SCC increase in treated cows and attributed this result to the positive correlation between the SCC and milk production increase.
There was no difference in the incidence of clinical mastitis up to 250 DIM between treatments, similar to results reported by St-Pierre et al. (2014) in a meta-analysis. However, there was an interaction between dose × start time, where clinical mastitis occurrences were greater for rbST 500/40 DIM compared to rbST 250/40 (17 vs. 8 cases, respectively). The increase in clinical mastitis can be mainly attributed to two factors: an increase in milk production in response to the rbST 500 (White et al., 1994) and start applications performed in a period reported how of the greater risk of intramammary infections, up to 100 DIM (Oliveira et al., 2013). St-Pierre et al. (2014) reported that rbST did not affect days open and services per pregnancy but reduced the pregnancy proportion by 10.5% in multiparous cows. In this study, we observed negative effects on open days and number of services per pregnancy, possibly attributed to increased milk production in response to rbST 500. The increased production is associated with enhanced metabolic rate, which influences the circulating concentration of hormones such as estradiol and progesterone and can lead to suppression of estrus expression (Wiltbank et al., 2006). As the reproductive management was based solely on visual estrus detection, increased yield for rBST 500 cows could also have led to poorer estrus expression and detection. Another hypothesis would be associated with GH and IGF-1 serum concentration. Castigliego et al. (2009) reported a progressive increase in IGF-1 concentrations throughout lactation in rbST-treated animals (500 mg at 14-day intervals). Contrary to serum GH concentrations, which reduce from mid to end application cycles, IGF-1 concentration remains above basal concentrations throughout the period, increasing rapidly after each new application. Bilby et al. (2006) observed the existence of optimal somatotropin and IGF-1 concentrations for exertion of positive effects on reproduction, where an increased above set threshold could trigger negative reproductive performance responses.

Conclusions
The present study contributes with knowledge on optimal treatment start time, dose, and application interval of rbST in crossbred cattle. Holstein × Gyr cows (3/4 and 7/8) managed in a semi-confinement system presented greater productive performance and health when treated with 500 mg of rbST initiated at 63 DIM. The dose × application interval interaction indicates a possible optimal circulating rbST dose between 20.8 and 35.7 mg per day for greater responses. There was no effect of treatments on SCC and clinical mastitis incidence. Treatments with 500 mg of rbST resulted in increased open days and number of services per conception.