Materials and Methods

Cattle and study design—Seventy-eight Holstein cows with confirmed expected parturition dates were used. The study population originated from and was housed at the University of Missouri-Columbia Foremost Dairy. Cows had ceased lactation 0 to 6 days prior to enrollment. Upon initial enrollment, a coin flip was used to allocate the initial cow into its respective study group with binary allocation into the 2 study groups that followed. Cows assigned to the vaccinated group were given 2 mL of M bovis bacterin^a SC in the neck area on the right side. Simultaneously, a 10-mL blood sample was collected aseptically via the coccygeal vein. Cows assigned to the control group received no vaccination. A blood sample was collected from these cows in the same fashion. Three weeks later, a booster vaccination of the same product was given to the vaccinated group and a second blood sample was obtained in the same fashion. A blood sample was obtained from the control cows at the same time and in the same fashion. Cows were monitored for impending parturition by farm management personnel. At the time of calving, a third blood sample was obtained. All blood samples were centrifuged, and the serum sample was decanted into a 5-mL serum collection tube. Cow number, sample number, and collection date were used to identify samples. Samples were frozen at −10°C until serologic analysis was performed. At calving, a 10-mL composite sample of colostrum was obtained from each cow in an aseptic fashion and collected into a 10-mL tube. Colostrum samples were frozen at −10°C until serologic analysis was performed. A composite milk sample was collected in a similar way at 7 to 13 days of lactation.

Serologic, colostrum, and milk analysis—Analysis of serum, colostrum, and milk samples was performed by American Animal Health Laboratory in Grand Prairie, Tex. Serum samples were analyzed for M bovis –specific IgG₁ concentration. Colostrum and milk samples were analyzed for M bovis –specific IgG₁ concentrations. The technical staff was unaware of the study group from which each sample originated.

ELISA—An ELISA was developed and used for analysis of submitted samples. Mycoplasma bovis –coated plates prepared by American Animal Health Laboratory were used. Plates were blocked with gelatin. Bovine serum samples were diluted 1:100 with PBS-Tween 20 solution. One hundred microliters of sample was used. Serum samples collected from M bovis –infected cattle were used as positive controls at 2X serial dilution starting at 1:100. Fetal bovine serum at a 1:100 dilution was used as a negative control. Samples were incubated at 37°C for 1 to 2 hours. Samples were washed 3 times with dH₂0 or PBS-Tween 20 solution. One hundred microliters of anti–bovine IgG₁ for serum, milk, and colostrum analysis was added with horseradish peroxidase conjugate. Samples were again incubated for 1 to 2 hours at 37°C. Samples were washed 6 times with dH₂0 or PBS-Tween 20 solution. One hundred microliters of 3,3′,5,5′-tetramethylbenzidine substrate was added and left to develop for 10 minutes. One hundred microliters of 0.5M H₂SO₄ was added to stop the reaction. Optical density was read with a microplate reader with dual wavelength of 450 and 630 nm. To minimize ELISA plate-to-plate variability, S:P ratios were calculated for each sample by use of the following formula:

View Expanded

where OD₄₅₀ is optical density measured at a wavelength of 450 nm.

Statistical analysis—Anti–M bovis IgG₁ antibody responses are indicated as mean ± SEM S:P ratios. Variance ratio tests were performed for each comparison between groups. Significant differences in variance were adjusted by use of the Welch approximation. Mean serum, colostrum, and milk IgG₁ antibody response were compared between the 2 groups by use of the Student t test. Mean serum IgG₁ antibody response in groups were compared by use of paired t tests.³

Linear regression models that predicted maternal anti–M bovis IgG₁ S:P ratios at parturition as a function of maternal serum anti–M bovis IgG₁ S:P ratios 39 days prior to parturition were constructed. Linear regression models that predicted colostral anti–M bovis IgG₁ S:P ratios as a function of maternal serum anti–M bovis IgG₁ S:P ratios 39 days prior to parturition and at actual date of parturition were also constructed.⁴ Variables with associated values of P ≤ 0.05 were included. The regression coefficient, P value, and correlation coefficient were reported for each significant regression model.

Differences were considered significant at P < 0.05. All statistical analyses were performed by use of specialized software.^b

Results

Serologic results—The 2 groups consisted of 40 control cows and 38 vaccinated cows. Mean S:P ratio values between the control and vaccinated groups were similar upon enrollment in the study and at concurrent primary vaccination (0.2813 ± 0.012 and 0.2514 ± 0.021, respectively). Mean S:P ratio values were significantly different between the control and vaccinated groups at the time of second vaccination 3 weeks following enrollment (0.3228 ± 0.019 and 0.5535 ± 0.031, respectively). Mean S:P ratio values were significantly different between the control and vaccinated groups at the time of parturition as well (0.2321 ± 0.023 and 0.3090 ± 0.030, respectively; Figure 1). Mean ± SD nonlactating periods for the cows in the control and vaccinated groups were 53.51 ± 20 days and 55.00 ± 20.54 days, respectively (P = 0.75).

There was a significant difference in mean S:P ratio values between values measured at 39 versus 60 days prior to expected parturition and between parturition and 39 days prior to expected parturition in the control and vaccinated groups. There was also a significant difference between values measured at 60 days prior to expected parturition date and actual date of parturition in the control and vaccinated groups.

View Full Size

Mean serum S:P ratios for IgG₁ antibodies against Mycoplasma bovis in control (n = 40 [gray bars]) and vaccinated (38 [black bars]) cows at 60 and 39 days prior to expected parturition date. Day 0 represents actual date of parturition. ^aSignificant (P < 0.05) difference, compared with the control group. ^bSignificant (P < 0.05) difference, compared with prior measurement for the control group. ^cSignificant (P < 0.05) difference, compared with prior measurement for the vaccinated group. ^dSignificant (P < 0.05) difference, compared with value at −60 days prior to expected parturition for the control group. ^eSignificant (P < 0.05) difference, compared with value at −60 days prior to expected parturition for the vaccinated group.

Citation: American Journal of Veterinary Research 69, 7; 10.2460/ajvr.69.7.912

• Download Figure
• Download figure as PowerPoint slide

Colostrum—Colostrum samples from 28 control cows and 26 vaccinated cows were available for analysis. Colostrum samples from 24 of the cows available from the serology study were not available for analysis. These samples were not collected at the time of parturition or appropriate aliquots were not obtained from collected colostrum before the samples were frozen. Mean colostrum IgG₁ S:P ratio values were lower in the control group (Figure 2) . Mean S:P ratio values between the control and vaccinated groups were significantly (P = 0.011) different (0.3546 ± 0.050 and 0.5661 ± 0.087, respectively).

View Full Size

Mean colostrum S:P ratios for IgG₁ antibodies against Mycoplasma bovis between control (n = 28 [gray bars]) and vaccinated (26 [black bars]) cows and mean milk S:P ratios between control (n = 32 [gray bars]) and vaccinated (27 [black bars]) cows. See Figure 1 for key.

Citation: American Journal of Veterinary Research 69, 7; 10.2460/ajvr.69.7.912

• Download Figure
• Download figure as PowerPoint slide

Milk—Milk samples from 32 control cows and 27 vaccinated cows were available for analysis. Milk samples from 19 of the cows available from the serology study were not available for analysis. These samples were not collected at the appropriate time according to the study protocol of 7 to 13 days of lactation or these cows had been culled soon after parturition. Mean milk IgG₁ S:P ratio values were lower in the control group (Figure 2). However, mean S:P ratio values were not significantly (P = 0.322) different between the control and vaccinated groups (0.5585 ± 0.047 and 0.6511 ± 0.079, respectively). A significant difference in mean S:P ratio was found between colostrum and milk of cows in the control group (P = 0.013) but not in cows in the vaccinated group (P = 0.437).

Regression modeling—Predicting maternal anti–M bovis IgG₁ S:P ratios at parturition as a function of maternal serum anti–M bovis IgG₁ S:P ratios 39 days prior to parturition yielded a significant linear model, as did predicting colostral anti–M bovis IgG₁ S:P ratios as a function of maternal serum anti–M bovis IgG₁ S:P ratios 39 days prior to expected parturition (Table 1). Predicting colostral anti–M bovis IgG₁ S:P ratios as a function of maternal serum anti–M bovis IgG₁ S:P ratios at actual date of parturition did not yield a significant linear model.

Discussion

In the present study, on the basis of herd records, cases of Mycoplasma spp mastitis were not recorded and results of random bulk tank bacteriologic cultures for Mycoplasma spp were negative. To the authors' knowledge, cows enrolled in this controlled trial had never received a prior M bovis bacterin. As expected, all cows entering this trial had statistically similar anti–M bovis S:P ratios. At the time of booster vaccination, there was a substantially larger mean serum S:P ratio in the vaccinated group versus the control group, and this continued though parturition. There was a noticeable decrease in mean serum S:P ratios at time of calving. If this variable is used to gauge the overall immune status of the periparturient dams, these findings are consistent with other studies.^5–7 These alterations of immune status in dairy cows begin approximately 3 weeks prior to calving and continue until 3 weeks after calving.⁸ Serum IgG₁ concentrations typically decrease at time of parturition.^7,9 Although the present study only evaluated mean concentrations of anti–M bovis IgG₁, there was a difference between mean IgG₁ concentrations at calving versus enrollment 60 days prior to parturition in both groups, with vaccinated cows having higher concentrations at calving.

Predicting the concentration of anti–M bovis IgG₁ in the serum of cows at parturition can be accomplished by knowledge of serum concentrations approximately 5 to 6 days before parturition. High serum concentration before parturition is associated with high serum concentration at parturition, but the correlation is moderately weak. The same correlation cannot be made between serum IgG₁ concentration at parturition and colostrum concentration of IgG₁, which appeared to have no linear relationship.

Colostrogenesis begins about 15 days prior to parturition. This event is marked by active and selective transport of immunoglobulin, generally of the IgG₁ subclass, into the mammary gland.^10,11 Colostral quality and subsequent calf health can be influenced by vaccination status of the dam, exposure of the dam to infectious diseases, production of protective antibodies, and antibody transport into the mammary gland.⁸ In the present study, vaccinated cows produced colostrum higher in anti–M bovis IgG₁ concentration than did their nonvaccinated herdmates. Again, the purpose of this study was not to determine the effective concentration of colostrum IgG₁ needed to protect calves against M bovis. Calves have not been challenged with M bovis exposure to determine if any protection is provided via colostrum of vaccinated dams.

Analysis of milk, however, revealed no significant differences between the 2 study groups. With the purpose of vaccinating cows in late gestation for the sole benefit of promoting immunity in neonatal calves, milk anti–M bovis IgG₁ is of less benefit than colostral anti–M bovis IgG₁. Calf feeding practices in modern dairy production systems usually involve feeding a balanced milk replacer, rather than dump milk from lactating cows. However, milk anti–M bovis IgG₁ concentration is greater than colostrum anti–M bovis IgG₁ concentration. Higher IgG₁ concentrations are expected in colostrum than in milk. Further investigation into the colostrum and milk ELISA used in this study is needed to further validate their practical usage as rapid tests for determining whether adequate or protective concentrations of anti–M bovis IgG₁ are present in dams.