Two experiments were conducted during the period designated for FMD vaccination by the official campaign against FMD in Uruguay (i.e., 15 February to 15 March 2020). A total of 739 lactating Holstein cows were used in the experiments; these cows were located on two dairy farms that produce an average of 37 kg/d/cow in a free-stall housing system (Experiment 1) and 20 kg/d/cow in a pasture-based system (Experiment 2). All experimental procedures were approved by the Internal Animal Care Committee of Fundación IRAUy (protocol number 003/2019). In both experiments, within each location, the females were randomly assigned to two experimental groups to receive or not receive 2 ml of FMD vaccine (Bioaftogen series 945, Biogenesis Bagó, Buenos Aires, Argentina). According to the manufacturer, the vaccine consisted of an oil emulsion compound containing FMD virus types O1 Campos and A24 Cruzeiro replicated in BHK suspension cell culture, inactivated with binary ethylenimine, and purified with polyethylene glycol. Vaccination was administered as recommended by the official authorities by subcutaneous injection in the neck region with a 15 mm × 18 gauge needle/
2.1. Experiment 1
2.1.1. Animals and experimental groups
The study involved 593 lactating Holstein cows, consisting of 263 primiparous and 330 multiparous, with a body condition score (BCS) of 3.7 ± 0.1 (mean ± SEM, scale 1 to 5, from emaciated to obese, respectively; 21). The cows had an average milk yield of 37.2 ± 0.3 kg/d/cow (ranging from 21 to 60 kg/d) and 187.8 ± 2.7 days in milk (DIM, ranging from 34 to 492 days). These lactating animals were managed under the same conditions in a free-stall housing system across three locations within the same company. The cows were milked three times a day (i.e., at 10:00 h, which was considered the morning; 18:00 h, which was considered the afternoon; and 02:00 h, which was considered the evening milking) and received an average of 25.5 kg dry matter per cow of a total mixed ration with unrestricted access to water. On Day 0 of the experiment, the animals were randomly assigned to two experimental groups to receive (n = 268) or not receive (n = 325) the FMD vaccine. From Day -3 until Day 9, milk production was measured in terms of quantity (kg), conductivity (mS/cm), and milk flow (kg/min) (DemaTron 70 GEA Farm Technologies, Düsseldorf, Germany, endorsed by the International Committee for Animal Recording). The data were recorded (Dairy Plan C21, GEA) and analyzed via commercial software (DairyComp, Valley Agricultural Software, California, USA). Body temperature was determined from Day 0 to Day 3 postvaccination by measuring the vaginal temperature once a day (between 11:00 and 12:00) via digital thermometers (OMRON, Dalián, China) in a subset of 96 cows that received (n = 56) or did not receive (n = 40) the FMD vaccine. All the measurements were taken by the same operator. The environmental conditions, including humidity and air temperature, were recorded to calculate the temperature‒humidity index (THI) via the following formula: THI = (0.8 × T °C) + [(RH/100) × (T°C – 14.4)] + 46.4, where T = temperature and RH = relative humidity (22). The THI scores were as follows: 81 on Day 0, 78 on Day 1, 75 on Day 2 and 72 on Day 3.
2.1.2. Determinations
The daily milk weights from the morning, afternoon, and evening milking sessions were combined to determine the daily milk yield (kg/d). The morning milk accounted for 34.8 ± 0.1% of the daily production, the afternoon milk accounted for 31.9 ± 0.1%, and the evening milk accounted for 33.3 ± 0.1%. For data analysis, cows were classified according to daily milk production, which was calculated as the average for the three days preceding Day 0 of the experiment (37.2 ± 0.3 kg/d). For further analysis, cows were categorized into those producing above and below the average: ≥ 37.2 kg/d (mean 42.4 ± 0.3 kg/d, n = 276) and < 37.2 kg/d (mean 32.9 ± 0.2 kg/d, n = 317), respectively. Cows were also classified according to DIM, using the average DIM: < 188 DIM (mean 111.9 ± 1.4 days, n = 207) and ≥ 188 DIM (mean 228.5 ± 2.0 days, n = 386). Cows were classified as normothermic when their vaginal temperature was < 39.5 °C and hyperthermic when their vaginal temperature was ≥ 39.5 °C (23).
2.2. Experiment 2
2.2.1. Animals and experimental groups
The study involved 146 lactating Holstein cows, 45 primiparous and 101 multiparous, with a BCS of 3.6 ± 0.1 (mean ± SEM), a milk yield of 20.3 ± 0.3 kg/cow (ranging from 9.4 to -35.9) and 260.8 ± 8.5 DIM (ranging from 14 to -584 days). The cows were managed in a pasture-based system and milked twice a day, at 05:00 h (considered morning milking) and at 16:00 h (considered afternoon milking). During the experiment, the cows grazed on a sorghum field and were supplemented twice a day with 4 kg of a balance ration (18% vegetal protein) with unrestricted access to water. Only cows without clinical symptoms of mastitis were included in the experiment (24). The mean annual bulk milk somatic cell count (SCC) in this herd was 198.1 ± 8.7 × 1,000 cells/ml, with a protein content of 3.47 ± 0.02% and a fat content of 3.53 ± 0.05%. On Day 0 of the experiment, the animals were randomly assigned to two experimental groups to receive (n = 78) or not receive (n = 68) the FMD vaccine. On Day 0 and Day 3, milk yield was recorded at each milking time using milk meters (J. Delgado, Scuéllamos, Ciudad Real, España). On Days 0 and 3, milk yield was recorded at each milking time using milk meters (J. Delgado, Scuéllamos, Ciudad Real, Spain). During the afternoon milking on these days, individual 10 ml milk samples were collected from proportional-line samplers, preserved with bronopol (2-bromo-2-nitro-1,3-propanediol, CAS number 52-51-7), and transported to the laboratory (COLAVECO, Colonia, Uruguay) for analysis of milk components and SCC.
2.2.2. Milk composition
Milk samples taken on Day 0 and Day 3 were analyzed for fat content (g/100 ml, %), protein content (g/100 ml, %), lactose content (g/100 ml, %) and SCC (x 1,000 cells/ml). Additionally, on Day 3, milk urea nitrogen concentrations (MUN, mg/dl) were also analyzed. The SCC was performed with a somatic cell counteremploying aflow cytometerbased on the fluoro-optoelectroniccell counters technique (Delta Instruments Combiscope FTIR 600 Dairy Analyzer, Somascope 600, North Shore, New Zealand) according to ISO 13366-2/IDF 148-2:2006. The expanded uncertainty intralaboratory for the SCC was 0.002 log 10, and the interlaboratory uncertainty was 0.004 log 10. The milk components were analyzed via mid-infrared spectrometry (Delta Instruments Combiscope FTIR 600 Dairy Analyzer, Lactoscope 600, North Shore, New Zealand) according to ISO 9622:2013 – IDF 141. For fat content, the expanded uncertainty intralaboratory was 0.012 g/100 ml, whereas the interlaboratory uncertainty was 0.023 g/100 ml. For protein content, the expanded uncertainty intralaboratory uncertainty was 0.096 g/100 ml, and the interlaboratory uncertainty was 0.028 g/100 ml.
2.2.3. Determinations
Cows were classified according to DIM (classification was performed with respect to the average DIM) into < 261 DIM (mean: 189.8. ± 2.4 days and 22.6 ± 0.4 kg/d, n = 89) and ≥ 261 DIM (mean: 375.6 ± 8.2 and 19.0 ± 0.4 kg/d, n = 57). The relationship between milk fat and protein content was calculated (fat %/protein %).
2.2.4. Statistical analysis
In both experiments, the mean daily milk yield was analyzed via a generalized linear mixed model (GLMM) using InfoStat software (25). The model included treatment (vaccination vs. no vaccination), day of the experiment, parity status (primiparous vs. multiparous), DIM (< 189 DIM vs. ≥ 189 DIM for Experiment 1, and < 261 DIM vs. ≥ 261 DIM for Experiment 2) and their interaction as fixed factors. In Experiment 1, the analysis was performed separately for two different milk yield levels (i.e., ≥ 37.2 kg/d/cow and < 37.2). Animal identification (in both experiments, 1 and 2) and herd location (in Experiment 1) were included as random factors. In Experiment 1, body temperature was analyzed via GLMM, including the same variables previously described as fixed factors, while animal identification was included as a random factor. In Experiment 1, both Pearson's and partial correlation coefficients were used to describe the relationships between variables. Specifically, we analyzed the variation in milk production before vaccination (averaged over the three days preceding Day 0) and the average milk production on Days 1 and 2, alongside body temperature on Day 1. The logistic procedure was used to generate the regression model and determine the intercept and slope values according to maximum likelihood estimates for each significant continuous order effect. The probabilities were determined according to the following equation: probability ¼ (elogistic equation)/(1 þ elogistic equation). Logistic regression curves were constructed according to the coefficients provided by the interactive data analyses from InfoStat software. Polynomial regression was used to select the statistical models according to the significance of the regression coefficients (P < 0.05) and in relation to the coefficient of determination (R2). Regression analysis was used to determine the nature of the relationship (linear, quadratic, or cubic) between each measurement influenced by pregnancy or the occurrence of estrus. In Experiment 2, the SCC and milk composition were analyzed via GLMM using InfoStat software. The SCC data were analyzed using the natural logarithm of the SCC, and the SCCs were log-transformed to a somatic cell score (SCS = log2 (SCC/100,000 cells) + 3) (26). The statistical model for MUN included the following fixed factors: treatment, DIM (< 261 vs. ≥ 261 DIM), parity status and their interaction. The model for milk lactose, milk fat, milk protein and the relationship between milk fat and protein included treatment, DIM, parity status, milk production level, day of the experiment and their interaction as fixed factors. Animal identification was included as a random factor.
Data from both experiments are presented as the mean ± SEM, with significance set at P < 0.05 and tendency set at 0.05 < P < 0.10.