Rabies is an infectious disease of viral etiology that causes acute encephalitis, with rapid and usually fatal evolution in all mammals (1). The etiological agent is the RNA virus belonging to the family Rhabdoviridae and genus Lyssavirus, and it is a zoonosis distributed worldwide that now affects mainly low-income countries (2). Rabies virus is transmitted by contact with infected saliva through bites or scratches. Dogs are the main reservoir of rabies and are responsible for most fatal cases in humans worldwide (2). However, in Latin America, with success obtained by the countries to control rabies mediated by dogs through the mass vaccination campaign (3, 4), bats have become the main reservoir of rabies over the last decade and are responsible for thousands of cases in livestock and for most of the remaining cases in humans (1, 4). Within bats, the common vampire bat Desmodus rotundus is the most important reservoir because of the high occurrence of spillover of rabies from this species of bat to other animals and humans, and vampire bat rabies (VBR) remains unpredictable and uncontrolled in several areas of the continent.

By feeding every night on livestock, VBR causes significant economic losses in Latin America, particularly to small-scale farmers (5). D. rotundus can also feed on the blood of human beings, resulting in occasional and unpredictable outbreaks of VBR in remote settings such as the Amazon region (6). The epidemiology of VBR circulation among D. rotundus that results in rabies cases among livestock is driven by several natural factors that are still poorly understood. Anthropogenic and ecological features favor the presence of bats (e.g., distribution of cattle herd and land occupation) (7), and factors favoring VBR among bats are also involved. However, understanding VBR circulation requires the capacity of measuring rabies exposure or infections within the bat population.

Most studies understanding rabies circulation among bats rely on official reports of livestock mortality, but this data is often biased by variable levels of under-reporting across the landscap (8, 9). Herbivores are accidental hosts of the rabies virus. Because they are a dead-end host, they only contribute as sentinels in the existence of the virus in the bat population (7). Despite its limitations, studies based on livestock mortality have shown that rabies can circulate in the landscape in different ways including wave-like spread into new areas, metapopulation dynamics, or endemically (9–11). However, the dynamics of the virus in endemic areas where the virus has been established for longer periods of time is less understood. Therefore, surveillance of rabies in bats, although logistically challenging, can bring unique insides into the circulation of the virus and our ability to predict and prevent future outbreaks in humans and livestock (9, 11).

The recognition of bats as reservoirs of the disease made epidemiological surveillance extend to these species. Rabies virus is rarely isolated from an infected bat because infected bats are often lower than 1% of the population (11). Because the presence of anti-rabies antibodies correlates with exposure to the virus, serological studies can contribute to understanding VBR circulation among bats. Several studies have shown the presence of antibodies to rabies in bats that do not die from the disease, as seen in other animals and also in humans (12, 13), implying that exposure to rabies in this species does not necessarily leads to mortality (14). For example, in the Botucatu region of Brazil, an endemic region of VBR, 45% of D. rotundus vampire bats had virus neutralizing antibodies (VNA) titers ranging from 0.10 to 0.20 IU/ml, 9.31% had between 0.20 and 0.3 IU/ml, and 11% had VNA levels >0.30 IU/ml (14). Similarly, D. rotundus bats from 16 colonies were tested in the endemic region of Paraiba in Brazil and all animals presented anti-rabies antibodies titers in some level, with titers higher than 0.5 IU/ml in 30.1% of bats (15). Delpietro et al. (16) reported the detection of 30% antibodies in bats after an outbreak of bovine rabies in Argentina. Before the outbreak, 2% of the bats had the antibodies, and during the outbreak, the percentage increased to 4%, suggesting that the presence of high proportion of seropositivity results from a high circulation of the VBR among bats at the time of an outbreak.

Here, we took advantage of recent outbreaks of VBR among livestock in the Sao Paulo region of Brazil to further test whether seroprevalence in D. rotundus reflects the incidence of rabies in bat populations. Field work was combined with captive experiments to follow seropositive animals and increase our understanding of within-host anti-rabies antibodies dynamics among D. rotundus.

Serological studies of VBR comparing populations of bats with different scenarios of rabies outbreaks in cattle are rare but crucial to understand within-bat dynamics of rabies. We found that vampire bats from the region of Botucatu, which had experienced a recent outbreak of rabies in cattle 30 days prior to sampling, showed a high proportion of seropositive individuals compared to two other sampling sites where the latest outbreak in cattle was reported at least 3 years prior to sampling. In Botucatu, 30 days after the outbreak, half of the bats had antibodies considered to be protective against rabies, which decreased in proportion 60 days after the outbreak in the region. Seropositive animals kept in captivity showed a variable decreased in titer after 4 months, suggesting a rapid loss of antibodies to non-protective levels. Our study suggests that rabies serology in bats can be an accurate indicator to identify if rabies had circulated in bats sampled.

The high proportion of antibodies considered as protective for rabies among bats sampled in Botucatu suggests the presence and circulation of the virus (8, 11) and corroborates cases reported among livestock in nearby farms. Although the second collection of bats 30 days later was much smaller (N = 5), there was a decrease on the proportion of seropositive animals. This decrease suggests that not all bats in a colony are exposed to the virus and that rabies does not necessarily persist within a colony after an outbreak (16).

Although outbreaks from cattle in Bofete and Anhembi were not reported for at least 3 years, antibodies were detected among bats in the first capture in both colonies, but not high enough to be considered protective (<0.5 IU/ml). This could reflect recent exposure to low levels of the virus that does not result in disease symptoms including exposure shortly after birth, subclinical or asymptomatic infections, sublethal infection, carrier state, or latent virus activated by stress (11, 21–23). In fact, Blackwood et al. (11) suggested that the probability of developing a lethal infection upon exposure to rabies is low in vampire bats (~10%), which enables viral persistence in the slowly reproducing bat colonies. This is also observed in other bats species (14, 24).

Low levels of antibodies within a bat colony without recent exposure could also result from long-term persistence of antibodies if rabies exposure happened several years in that colony (8, 25). However, the duration of rabies antibodies after exposure without re-exposure remains poorly understood for vampire bats. In this study, anti-rabies antibodies decreased to low levels 60 days after capture of seropositive individuals but fluctuations and sudden increases were also observed in some individuals. Antibodies fluctuation has been reported in free bats but explained by re-exposure to the virus, which was not possible in our laboratory settings (26). This result suggests possible rabies virus persistence in the bat tissue with periodic antigenic stimulation. This calls for future studies on the mechanisms behind anti-rabies antibody production and on the immunological and protective consequences of having low levels of antibodies in healthy bats.

The increase of anti-rabies antibodies titers on the second capture observed in Bofete suggests that rabies circulation can be detected in the bat population despite cases not been reported among livestock. This highlights the possibility of using bat serology surveillance to predict risk in that population and therefore affects livestock in the region too. Therefore, cost-effective analysis and more understanding on the serological behavior of bat population are needed to evaluate whether rabies serological surveillance among bats can be added as a tool to improve our understanding and prevention of rabies risk in both livestock and humans.

Overall, our study shows that anti-rabies antibodies among vampire bats could reflect the current risk of rabies outbreaks among nearby cattle. Furthermore, we observed a decrease in antibody titers 2 month after animals had high titers, although titers fluctuated in time and between individual bats. Future studies are required to assess whether the observed titer levels >0.5 IU/ml are indeed protective against rabies on bats and whether low-titers can also reflect rabies circulation among bats and subsequent risk to livestock and humans.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by CEUA FMVZ.

FC, LB, BR, CA, and JM Bats capture, blood collection and work execution. JB, JM, and LB Foreground research, writing and editing. IK, KS, SS, AR, GB, and LA Serology. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.