In this study, we analyzed the PVL and the antibody level of naturally BLV-infected cattle from fifteen different herds. We found that the humoral response reflected the level of in vivo infection, and may therefore have useful epidemiological applications.
We worked with 15 commercial herds, which are typical large dairy herds of the country considering productive parameters and BLV prevalence (Table 1). This study complements an ongoing work, in which the same animals are being studied for the presence of polymorphisms in the complete genome that could be associated with the BLV PVL outcome.
In all farms, we found animals with different PVL status (Figure 4), which confirmed the concept that BLV PVL varies between naturally infected animals, as previously reported in two previous works, both based on one farm case [3, 4].
The results presented here provide a baseline to design an alternative control strategy based on the permanence of animals with low levels of infection in the herd, following the hypothesis that the level of proviral BLV in blood should play a major role in the success or failure of BLV transmission, as reported for experimental BLV infections [5, 10] and natural Human T Lymphotropic Virus-1 (HTLV-1) infections . The main goal of this program would be the rational control of intentional virus propagation, with the aim to obtain a low PVL in the whole herd, to finally diminish the risk of animal-to-animal transmission.
Under a practical point of view, the elimination of all infected animals with high proviral load is not be feasible on individual farms, since apart from mortality due to sporadic cases of lymphosarcoma, productive parameters are considered as normal in this group of cattle. In this context, the application of a selective segregation plan based on proviral load could be appropriate in a consortium-based strategy, where a group of farms work together toward the elimination of infection. In this case, animals with low proviral load should be recruited to form a low-transmission farm. In this scenario, the quantification using real-time PCR in whole natural herds to select animals that would be recruited to form the low-transmission farm would be extremely expensive. Hence, an alternative approach to select and recruit clean and low-infected animals becomes necessary. With this concept in mind, we analyzed the relationship between the individual level of infection and the serological profile with the aim to find an affordable indicator of PVL that could be used in local laboratories without the need of expensive equipment and reagents. We first analyzed whole-BLV and p24 antibodies in two herds, in which we quantified PVL in the total number of samples, with the aim to detect the best PVL predictor among the serological options. The ROC analysis showed that p24 antibodies allow predicting the PVL with a similar performance when titer or reactivity values are used for calculation (Figure 1). Even when whole-BLV antibodies were detected in a greater proportion of animals, we found no evident reason to work with these antibodies; as all these animals showed undetectable or low PVL. In our situation, this is important for two reasons. First, although the two ELISA antigens used were in-house, the p24-ELISA uses an Escherichia coli recombinant-derived antigen, which is less expensive and laborious to elaborate than the cell-derived antigen used in the whole-BLV ELISA. Secondly, there is no need to titrate the samples, since the strength of ELISA reactions seems to be as good as titer to predict the PVL (Figure 1). These two reasons make the serological analysis extremely cheap and straightforward. The prediction made by ROC analysis in these two farms was supported by the analysis of variance (Figure 2) and the correlation analysis, which showed that PVL is well reflected by the p24 antibody level.
Since the analysis of p24 antibodies showed a subpopulation of weak-p24 reactors in all herds (Table 1), we extended the analysis of PVL to this subgroup. We confirmed that a great proportion of animals of this subgroup showed undetectable or low PVL (Figure 3). This finding allows us to consider this subpopulation as a good candidate to be recruited as potentially low disseminators of infection. In a consortium-based approach, at least 20% of the animals of each farm should be considered as putative donors to form a BLV low-dissemination herd according to our analysis (Table 1).
Then, the control strategy would consist in the recruitment of animals with low or undetectable PVL using p24 serology as a low-cost method, instead of PVL quantification. This plan should be rationally designed considering that (1) all the animals that participate in the plan should be checked for p24 antibody reactivity, (2) negative and weak p24 reactors should be selected and recruited to form a new potentially low-transmission farm, (3) PVL should be checked up in the recruited group, in which animals with high PVL will still be present and should be eliminated, and (4) monitoring should be done in a regular basis to constantly “clean” high PVL individuals that might appear, even when PVL is thought to keep constant . With this strategy, the transmission rate should become lower and lower as the animals with high propagation potential are removed.
A risk assessment should be made to analyze the dynamics of infection under these conditions, especially considering the theoretical extinction time when compared with naturally infected farms, as previously discussed .
Although this kind of control program based on the in vivo level of infection has been proposed , there are no reports showing the application and/or success of this strategy. In this context, a field trial will be designed and set up to analyze the feasibility and risk of the strategy, as a rational alternative to control BLV infection.
Finally, whether or not the immune response only reflects or also controls PVL is still unknown. Further studies should be carried out to define the reasons of the individual PVL differences and whether they are caused by the host genetic background, the viral strain, the infectious dose, and/or are due to multifactorial reasons. In the context of HTLV infection, the PVL is also correlated with the level of circulating antibodies [14, 15] and the specific CD8+ lymphocyte response contributes to the control of the PVL and is associated with a lower risk of clinical disease progression . Similar phenomena could be occurring with BLV natural infection and studies regarding the BLV-specific cytotoxic response and its relationship with the PVL and the humoral antibody response would be appropriate.