Comparison of bovine viral diarrhea virus-specific antibody responses of young beef calves vaccinated with either modified live virus or inactivated virus regimens

Abstract

Bovine viral diarrhea virus (BVDV) is an important pathogen causally associated with morbidity and mortality, and production losses in both suckling and weaned beef calves. Vaccination for protection against disease caused by BVDV is challenging because of the inhibitory effect of maternal antibodies; however, it is most convenient for beef producers to vaccinate calves before moving herds to summer pastures. We compared modified live and inactivated vaccines used for priming and boosting beef calves and found that neither type of vaccine results in strong antibody responses in the face of maternal antibodies. These data are generally consistent with previous studies and suggest that alternative protocols using early mucosal delivery followed by parenteral boosting should be examined to improve vaccine efficacy.

Bovine viral diarrhea virus (BVDV) is an important cause of morbidity and mortality in beef calves and the effects of this virus include loss of production post-weaning (1,2). Numerous inactivated (KV) and modified live vaccines (MLV) are commercially available and used for the control of BVDV genotypes 1 and 2 (3). Despite the availability of commercial vaccines, challenges in controlling the impact of these infections on newly weaned beef calves persist, as many calves are likely not primed appropriately. The most common time for vaccination of beef calves in western Canada is before movement to summer pasture when the calves are approximately 40 to 50 d old (4). This initial vaccination is typically administered by subcutaneous (SQ) injection in the face of maternal antibodies; this is especially important for BVDV. Maternal antibodies (MatAb) for BVDV types 1 and 2 have been shown to persist for 210 and 180 d, respectively (5), depending on the success of passive transfer. Previous work has shown that a circulating antibody response is observed in calves that are older than 70 d with an approximate BVDV type 1 geometric mean titer (GMT) of 256 upon administration of a SQ vaccine. However, if administered to 2-day-old calves, with an approximate GMT of 1000, no antibody response is found (6).

The current study compared the BVDV-specific antibody concentrations in 3 groups of calves administered either a KV or MLV vaccine at ~48 d (V2) [standard deviation (SD) ± 8] of age (an age that is typical for first vaccine administration of young calves in western Canada) and a subsequent dose of vaccine at weaning [V3) (~185 d (SD ± 8)] (Figure 1). The study was conducted with approval from the University of Saskatchewan, animal use protocol 20160003. Day-old beef calves (N = 75) born to multiparous beef cows that received pre-breeding vaccines (Bovi-Shield Gold FP 5; Zoetis Canada, Kirkland, Quebec) at a commercially operated research farm east of Lanigan, Saskatchewan were enrolled into a related study. The calves were randomly allocated into 3 groups of 25 (Figure 1) that either received saline intranasally (Group 1) or were administered a combination intranasal vaccine (V1) (Group 2, Group 3) (Inforce 3; Zoetis Canada), which contained bovine respiratory syncytial virus (BRSV), bovine herpes virus type 1 (BHV1), and bovine parainfluenza virus type 3 (BPIV3). Calves were enrolled into the study within 24 h of birth. Calves in groups 1 and 2 were next vaccinated at V2 and V3 (Figure 1) with an injectable combination modified-live vaccine [BRSV, BHV1, BPIV3, and bovine viral diarrhea virus (BVDV) types 1 and 2] (Bovi-Shield Gold FP 5; Zoetis Canada). Calves in group 3 were vaccinated at V2 and V3 (Figure 1) with an injectable inactivated vaccine with antigen to BRSV, BHV1, BPIV3 and BVDV types 1 and 2 (Triangle 5, Boehringer-Ingelheim Canada, Burlington, Ontario). All injectable vaccines were administered subcutaneously in the neck and all calves were abruptly weaned on the same day. The suckling calves were maintained on pasture as 1 group through the summer and during the post-weaning study period.

Figure 1.

Outline of treatment group vaccine regimens and timing. (V1 — Intranasal vaccine/saline; V2 — First SQ vaccination; V3 — Second SQ vaccination).

One calf in Group 2 died within 24 h of birth and a postmortem examination revealed no determinable cause of death. One calf in Group 1 was removed from the study midsummer because its dam developed severe lameness and the pair needed to be removed from the pasture. Finally, 9 calves from each of Group 2 and Group 3, and 2 calves from Group 1 were removed at weaning for a BRSV challenge study (7).

Virus neutralization assays, performed at a diagnostic laboratory (Prairie Diagnostics, Saskatoon, Saskatchewan), were used to measure BVDV type 1 and 2 antibody concentrations on the day of each parenteral vaccination, V2 and V3, and then at both V2 + 2 wk and V3 + 2 wk. Since the first serum collection occurred at V2, the day of first administration of BVDV vaccine when the calves were ~48 d old (SD ± 8), the antibodies measured at this time were those almost certainly derived from maternal colostrum. Antibody concentrations were measured in serum from blood collected by jugular venipuncture. Antibody cut-off values were determined by 3-fold sample dilutions of positive and negative controls on the first plate. The first dilution measured was 1 in 6.

Statistical analyses were performed using SPSS version 25 (IBM, Armonk, New York, USA). Normality of the data was tested using Shapiro Wilk’s test and data for which P ≥ 0.05 were considered to be normally distributed. Neither BVDV type 1 or 2 antibody concentrations were normally distributed. Antibody concentrations were natural log transformed and then analyzed using generalized estimating equations with an AR1 working correlation matrix. The main effects in the model were treatment, time, and a treatment and time interaction term. The results were then compared pairwise to estimate the differences between and within treatment groups at different time points. Differences with a P-value ≤ 0.05 were considered significant.

The V2 serum samples showed no significant differences in GMT among the groups, indicating that calves in Groups 1, 2, and 3 had similar concentrations of MatAb for BVDV type 1 and type 2. At V2 + 2 wk the within vaccine group, geometric mean antibody concentration comparisons showed that each of the 3 groups either had a significant decrease or no change in BVDV type 1 and 2 antibodies (Tables 1 and 2). However, all groups showed decreased antibody concentrations at V2 + 2 wk, and were not significantly different among vaccine groups. This may indicate that there was no systemic antibody response to the first parenteral vaccination (V2). Calves with MatAb concentrations ≤ 32 can have an antibody response to injectable vaccine, both MLV and KV, and the concentration of MatAb at time of vaccination will determine whether or not an antibody response occurs (8). In all groups, GMTs at V2 were greater than 32; therefore, the presence of this concentration of MatAb at V2 likely resulted in vaccine interference. A lack of significant antibody response in calves with high MatAb concentrations has been previously demonstrated for both MLV and KV vaccines, and is consistent with the current finding (9,10). Management of the dams of the calves in this study, with regard to pre-breeding vaccination, was similar to that of western Canadian herds, and therefore the concentration of MatAb in the study calves is most likely representative of commercial herds in this region (4). Therefore, these results are important to commercial herds in western Canada, as they indicate that a response to vaccination (for BVDV) is unlikely to occur in the presence of high MatAb.

Table 1.

Vaccine group change in geometric mean BVDV type 1 antibody concentration.

	Group 1 (95% CI)	Relative difference	95% CI	Group 2 (95% CI)	Relative difference	95% CI	Group 3 (95% CI)	Relative difference	95% CI
~48 days	1447a,* (920, 2277)			1489a,* (921, 2407)			963a,* (625, 1485)
~48 days + 2 weeks	938a,* (555, 1584)	0.68	0.46 1.02	900a,* (580, 1397)	0.60	0.41 0.90	791a,* (529, 1183)	0.82	0.56 1.22
Weaning (~185 days)	16a,† (11, 26)	0.02	0.01 0.03	19a,† (13, 28)	0.02	0.01 0.04	9b,† (6, 13)	0.01	0.01 0.02
Weaning + 2 weeks	25a,† (13, 51)	1.43	0.95 2.17	36a,b,‡ (20, 64)	1.80	1.11 2.92	63b,‡ (33, 121)	5.84	3.65 9.36

^a,bIndicates significant difference (P ≤ 0.05) within rows.

^*,†,‡Indicates significant difference (P ≤ 0.05) within columns.

CI — Confidence interval.

Table 2.

Vaccine group change in geometric mean BVDV type 2 antibody concentration.

	Group 1 (95% CI)	Relative difference	95% CI	Group 2 (95% CI)	Relative difference	95% CI	Group 3 (95% CI)	Relative difference	95% CI
~48 days	611a,* (395, 945)			563a,* (395, 802)			557a,* (372, 834)
~48 days + 2 weeks	356a,† (241, 525)	0.61	0.39 0.95	371a,* (255, 540)	0.66	0.42 1.03	442a,* (279, 699)	0.79	0.51 1.23
Weaning (~185 days)	36a,‡ (17, 82)	0.10	0.05 0.20	54a,† (30, 97)	0.15	0.07 0.29	6b,† (4, 9)	0.01	0.01 0.03
Weaning + 2 weeks	124a,* (54, 281)	3.50	2.20 5.58	237a,‡ (99, 567)	4.04	2.34 6.97	36b,‡ (17, 77)	5.54	3.26 9.41

^a,bIndicates significant difference (P ≤ 0.05) within rows.

^*,†,‡Indicates significant difference (P ≤ 0.05) within column.

CI — Confidence interval.

At V3, weaning, all 3 groups had a significant decrease in BVDV type 1 and 2 antibody concentration from the antibody measurement at V2 + 2 wk. On the day of weaning the calves were vaccinated with the same vaccine that they received at ~48 d of age. By V3 + 2 wk, antibody concentrations in all groups had significantly increased for BVDV type 2 (Table 2). However, the significant increase in GMT, after V3, does not necessarily mean that an anamnestic response occurred; the low magnitude of antibody concentration was more consistent with a primary response. This is contrary to previous studies that reported anamnestic responses to subsequent doses of vaccine even when a primary antibody response to the initial dose was not found (11). While antibody responses between independent studies are not directly comparable, previous work shows that priming vaccination often results in a lower antibody concentration than an anamnestic response to a second/booster vaccination. For example, a recent study comparing vaccine response in pre-weaned and weaned calves found that BVDV type 1 antibody concentrations were 1.0 units after administration of a priming dose compared to > 20 units after the second dose (12). It should also be noted that Group 3 calves herein had significantly lower BVDV type 2 antibody concentrations than calves in either Group 1 or Group 2. The significantly lower antibody concentrations of Group 3 calves indicate a response to the vaccine that may not be protective despite a significant increase in GMT 2 wk post-vaccination.

For BVDV type 1, only calves in Group 2 (MLV-vaccinated) and Group 3 (KV-vaccinated) had significant increases in antibody concentration at the V3 + 2 wk time point (Table 1). While the magnitude of the increase was small, it does indicate an antibody response. The KV vaccine response to BVDV type 1 is similar to that reported by Fulton et al (9) who showed that calves vaccinated twice with the same KV vaccine used in the current study, had significantly higher BVDV type 1 and 2 antibody concentrations than unvaccinated controls.

With regard to the BVDV type 1 antibody response to the V3 vaccination, it was previously demonstrated by Chamorro et al (13) that calves vaccinated with the same MLV as their dams had significantly lower GMT concentrations, post-viral exposure, than GMT concentrations in response to 3 other MLV vaccines. This could have been due to antigenic diversity among the vaccines that were administered. While the authors of that study observed the response to BVDV challenge rather than a subsequent vaccination, their findings agree with the results of the current study in that both studies showed small magnitude antibody responses. The small magnitude response suggests that the V3 vaccination did not illicit an anamnestic response for either vaccine type in the current study; the significant but small magnitude response is likely due to MatAb interference blocking a priming response at V2 vaccination. Future work should use MLV vaccines that are antigenically different from those administered to dams, because a larger magnitude antibody response may likely occur when calves are vaccinated with a different MLV than their dams (13).

Platt et al (14) used a combination MLV to compare differences in BVDV type 1 and 2 antibody response amongst 3 groups of calves vaccinated in the face of maternal antibodies and later challenged with BVDV type 2. Calves in the Platt et al (14) study were vaccinated at 1 to 2 wk, 4 to 5 wk, or 7 to 8 wk of age; the calves in the 4 to 5-week group had an antibody response to the vaccine. The calves in the 7- to 8-week group had a slower decline in antibody and significantly higher antibody concentration than the age-matched controls. Both the 4- to 5-week, and 7- to 8-week groups showed anamnestic responses to challenge 12 wk after vaccination and the 1- to 2-week group did not. When the concentration of MatAb for BVDV types 1 and 2 is considered, the calves in the 2 older groups of the Platt et al (14) study had lower antibody concentrations than the calves herein. Calves in the older groups in the Platt et al (14) study had BVDV type 1 antibody concentrations of 56 and 42 units, and BVDV type 2 concentrations of 91 and 34 units (transformed from published data). In contrast, antibody concentrations for the current study, at time of first subcutaneous vaccination, ranged between 963 and 1489 units, and 557 and 611 units for BVDV type 1 and type 2, respectively. Given the antibody response of the older groups in the Platt et al (14) study, it is unlikely that sufficient MatAb was present to inhibit the response to vaccination, compared to the current study. It is important to consider antibody concentration at time of vaccination rather than calf age (relative maturity of the immune system) because MatAb antibody concentration likely has a greater influence on vaccine response (9). Therefore, future work should examine and report vaccine response with reference to antibody concentration at the time of vaccination, rather than in reference to age of calf. However, it should be noted that the current study did not directly measure MatAb prior to first subcutaneous vaccination and it is assumed that the BVDV antibody present at V2 was due to colostrum antibody transfer; this is a weakness of the study.

The differences in post-weaning changes in type 1 and 2 antibody concentrations in the current study indicate a variable response to the vaccine administered at weaning, with a greater magnitude of antibody increase for BVDV type 2. Again, this could reflect the relative antigenic diversity among the isolates of the 2 genotypes used in the vaccines. The measurable, but relatively small magnitude of change in antibody concentration suggests an immune response occurred, and it was likely the result of a primary versus secondary or anamnestic response. Given the limited magnitude of response, it is likely that the V2 vaccination, at ~48 d of age, did not sufficiently prime the systemic B-cell or CD 4⁺ helper T-cell responses necessary for optimal antibody production. The basis for this disparity in response to the 2 genotypes is unresolved, but could have to do with the relative immunogenicity of the respective antigens.

A previous study by Woolums et al (6) compared antibody responses in calves primed at 2 or 70 d of age, with either a combination intranasal or injectable vaccine containing antigens for both BVDV types 1 and 2. In that study, calves vaccinated parenterally at 2 d of age, with a GMT antibody concentration of 1000, had significantly lower BVDV type 1-reactive antibody concentration after booster vaccination at weaning than calves vaccinated intranasally at 2 or 70 d of age with GMT antibody concentrations of 1000 and 256, respectively (estimated from data presented). The Woolums et al (6) results indicated that calves which received injectable vaccine in the presence of high MatAb had less effective systemic priming than the calves primed intranasally at either 2 or 70 d of age. However, the responses of calves vaccinated parenterally at 70 d of age were not significantly different from those of the intranasally vaccinated calves after the post-weaning booster. This is consistent with the concept that MatAb should be diminished sufficiently before parenteral vaccination for a systemic antibody response to occur (15). The poor BVDV type 1 antibody response in the current study indicates that the calves likely had too high concentrations of MatAb at ~48 d to respond effectively, and therefore had a similar response as the calves in a previous study (6) that were vaccinated parenterally at 2 d of age. Again, this shows that the concentration of MatAb is likely more important than the age of calf when performing studies on vaccine response in young calves.

The current study agrees with previous studies that show a lack in systemic priming when vaccine is administered parenterally in the face of maternal antibodies (6,15). While the current study did not use BVDV antigens intranasaly, due to unavailability of commercial intranasal BVDV products, the study by Woolums et al (6) showed significantly higher antibody response than the calves parenterally vaccinated at 2 d old; therefore, intranasal priming should be further studied. Future work should investigate whether a response to injectable vaccination in the face of maternal antibodies can be improved when calves are first primed by intranasal administration early in life. CVJ