1. Introduction
Brucellosis in cetaceans was first described in 1994 [
1] and is caused by
Brucella ceti [
2]. Since then, infection and disease associated to
B. ceti have been increasingly recognized in many cetacean species worldwide [
3,
4], and brucellosis is considered an emerging threat for odontocetes and mysticetes [
5]. In the Mediterranean Sea, cetacean brucellosis was first diagnosed in 2013, and more cases have been reported since then [
6,
7,
8,
9,
10], but serological evidence of infection was already detected in dolphins stranded along the Mediterranean coast of Spain in 1997-1999 [
11], suggesting that Mediterranean dolphins were already exposed to the pathogen at that time. Disease due to
B. ceti in cetaceans can occur as neurobrucellosis, reproductive disease, spinal diskospondylitis or abscesses (frequently cutaneous and subcutaneous) [
4]. However, in many other instances, serological evidence of infection has been found in asymptomatic cetaceans, and
B. ceti has been isolated from tissues of apparently healthy cetaceans, suggesting that
Brucella-infected cetaceans may overcome initial infection and survive or become
Brucella carriers (see for review [
3,
4]). Health assessment of wild cetacean populations is done usually through necropsy of stranded dolphins and laboratory investigation. Since the first description of disease associated to
B. ceti in the Catalan Mediterranean coast [
8], we have intensified efforts to detect
Brucella in stranded cetaceans. In this paper we describe the cases of brucellosis detected in the period 2012 to 2023 in the North-western Mediterranean coast. Moreover, we perform an exploratory serological analysis using a commercial cELISA to compare antibody levels between infected and non-infected animals, age groups, sex, and cetacean morbillivirus (CeMV) status, with a special emphasis on striped dolphins. The general aims were to provide information about the epidemiology and pathogenesis of
B. ceti infection, and to evaluate its impact on cetacean species in the area.
2. Materials and Methods
2.1. Cases Investigated and Necropsy Procedure
A total of 131 cetaceans stranded along the Catalan coast during the period 2012-2023 were necropsied at the Universitat Autònoma de Barcelona, according to standard procedures [
12]. Necropsied species included striped dolphins (
Stenella coeruleoalba) (n=103), common bottlenose dolphins (
Tursiops truncatus) (n=13), Risso’s dolphins (
Grampus griseus) (n=10), common dolphins (
Delphinus delphis) (n=2), a fin whale (
Balaenoptera physalus), a pilot whale (
Globicephala melas) and a Cuvier’s beaked whale (
Ziphius cavirostris). Conservation code ranged from 1 to 3 (animals in overt autolysis were not transported for necropsy) following the scale of Kuiken & García-Hartmann 1991 [
13]. Retrieval of cerebrospinal fluid (CSF) was routinely attempted from the atlanto-occipital joint, or directly from the lateral ventricle after extraction of the brain. Frozen samples and swabs from multiple organs were collected for bacteriological and molecular investigations. A complete set of organs was sampled for histopathology, fixed in 10% neutral buffered formalin and routinely processed. Routine immunohistochemistry and RT-PCR for surveillance of CeMV were performed on lung, diaphragmatic lymph node and brain [
14]. After the determination of the cause of death, animals were classified in different groups: bycatch, neurobrucellosis, CeMV, PEM (polioencephalomalacia of unknown origin), mother-calf separation, infectious causes, sinusitis by
Crassicauda grampicola, erysipelas, other and unknown. “Infectious” cause of death comprised peritonitis, septicemia, protozoal infections, severe parasitism, mucormycosis, bacterial meningoencephalitis (negative
Brucella culture), and necrotizing enteritis. “Other” included animals dying from tension pneumothorax, meningoencephalitis of unknown origin, abortion, pulmonary angiomatosis and muscular degeneration.
2.2. Serology for Brucella
Cardiac blood retrieval was attempted by cardiac puncture at reception of each cadaver with a 20 ml syringe and a 16G, 1.7 x 133 mm catheter needle (AngiocathTM, BD, REF 382259) and portioned in vacutainer plain tubes. Blood was centrifuged at 2,500 x g, and serum was aliquoted in 1 ml cryotubes and frozen at -80 ºC until use.
For the present serological study, available serum samples from 2012 to 2022, excluding those from 2023, (n=66; 57 striped dolphins, 5 Risso’s dolphins, 2 common bottlenose dolphins, 1 common dolphin and 1 pilot whale), were defrosted (see Supplementary File 1 for all cetacean samples). A commercial blocking ELISA (INgezim Brucella Compac, INGENASA, Madrid, Spain) was used following the manufacturer’s instructions. This competitive ELISA test (Brucella cELISA) is a multispecies kit developed for detection of blocking antibodies against LPS of Brucella abortus in domestic ruminants and swine serum samples, using a peroxidase-conjugated secondary monoclonal antibody directed to the LPS of B. abortus. The dilution of the serum sample for cetaceans is not established for the test and, therefore, sera were diluted as recommended by the manufacturer for ovine and caprine sera (1:5), bovine and porcine sera (1:10), and additionally at 1:20, 1:40, 1:80, and 1:160. Available serum of culture-confirmed Brucella-infected cases (n=8) were further diluted to 1:320, 1:640, 1:1280, 1:2560, 1:5120, and 1:10240. Optical density (OD) values were measured at 450 nm within 5 minutes after the addition of stop solution using a spectrophotometer (FLx800, Bio-Tek Instruments, Winooski, VT). OD values were used to calculate the final results, read as a percentage of inhibition (PI) in comparison to positive and negative control sera included in the kit, with the formula PI=100x [1-(OD test sample/OD negative control), where OD= optical density. Following the procedures of the test, a sample was classified as positive if the PI in the well was ≥40%.
The possible association between the level of hemolysis and the
Brucella cELISA result was investigated in a non-published preliminary study. Briefly, the level of hemolysis was visually graded in the first dilution of samples when dispensed onto the ELISA plate into group 1 (non-hemolyzed or slightly hemolyzed serum), and group 2 (hemolyzed serum or hemolyzed blood). A comparison of positive and negative results in
Brucella cELISA (for 1:20 dilutions and 1:40 dilutions) with the hemolysis score was performed with the Wilcoxon Two-Sample Test (Epi Info Package,
www.cdc.gov/epiinfo/index.html).
2.3. Brucella Isolation
Brucella isolation, gold standard for brucellosis diagnosis, was attempted in dolphins with compatible lesions, a positive reaction to Bengal Rose Test (data not shown), or with CeMV infection. Samples used for
Brucella isolation were frozen swabs from the lateral cerebral ventricle, cerebral tissue around lateral ventricle, cerebrospinal fluid (CSF), spleen, mesenteric lymph node, or swabs from spondylytic lesions. Bacterial culture was performed as previously described [
8]. Briefly, tissue samples were superficially sterilized, homogenized in saline buffer and cultured in plates of both Farrell and CITA selective media. Isolates were identified as marine
Brucella using a Bruce-ladder PCR [
15]. Confirmation of
Brucella ceti as the isolated species was done by a multiplex PCR adapted from Bruce-ladder [
15].
2.4. Age Determination
The age of 57 striped dolphins for which serum was available was estimated using a Gompertz formula established for this species in the North Western Mediterranean Sea [
16]. Also, animals were divided in three age ranges (fetus, calves, juveniles and adults), considering adulthood above 6 years [
17]. The formula is not appliable to small animals under 120 cm (males) and 110 cm (females). For three calves of a few weeks of age (presence of fetal folds) age was estimated as 0.1 years. Age could not be estimated for dolphin N-604/17, which was found without a tail.
2.5. Data Analysis
Differences between antibody titers in striped dolphins, grouped by their cause of death and age, were compared in two-way contingency tables and using a Fisher Test. R software (version 4.4.0) was used to introduce data and elaborate the tables and StatCalc tool, from EpiInfo (version 7.2.6.0), to retrieve significance of the results. Significance was considered with p-value<0.05.
4. Discussion
The results show that brucellosis is regularly detected in small cetaceans in the western Mediterranean Sea, and it is the most relevant single bacterial cause of death in striped dolphins in the area, confirming previous similar findings from Mediterranean regions [
14,
18].
B. ceti was isolated in 10 out of 131 (7.6%) necropsied cetaceans, with this figure being the first estimated prevalence of the disease in dolphins from any Mediterranean coastal region. In 8 striped dolphins neurobrucellosis was considered the primary cause of death, whereas
B. ceti was considered a comorbidity in one striped dolphin with systemic CeMV infection and in one common bottlenose dolphin with
B. ceti diskospondylitis and a CNS mucormycosis [
8].
Understanding the pathogenesis of cetacean neurobrucellosis remains elusive, mostly due to the lack of knowledge about the types of immune responses elicited by
B. ceti in healthy and in sick cetaceans. It is generally accepted that Th1 responses are relevant in controlling replication of intracellular bacteria, and that a strong humoral (Th2-mediated) response seems to be unprotective, being antibodies a more useful indicator of infection than a proper defense mechanisms (for review, see [
19,
20]. Accordingly, a study with
B. melitensis in mice suggested that antibodies are not decisive in the control of infection [
21], although antibodies against LPS have proven to confer certain protection in
B. abortus [
19,
22]. As in other mammals, it may be hypothesized that cetaceans with a predominant Th1 response to
B. ceti survive the infection, and a shift to a Th2 response constitutes the hallmark of disease progression. However, to our knowledge, specific T-cell response against
Brucella sp. has never been measured in dolphins, and the role of serum antibodies against the bacteria is unknown. Several serologic tests have been used in cetacean species, either developed for terrestrial mammals or adapted to marine mammals [
11,
23,
24,
25,
26,
27,
28,
29,
30]. These studies have shown a high seroprevalence of
Brucella-infection ranging from 7.6% to 60%, both in cetaceans and pinnipeds, in many places of the world. However, the sensitivity (Se) and specificity (Sp) of these tests have not been established for cetaceans, and this knowledge is hindered by the lack of reference panel sera for cetaceans. Bearing in mind these limitations we applied a commercial cELISA for
Brucella diagnosis in cetaceans, not to establish Se and Sp values for this test, but rather to obtain a more defined picture of the prevalence of
B. ceti infection and disease, and to understand the practical utility of serologic results when applied to stranded cetaceans.
We observed a higher seropositivity in striped dolphins with neurobrucellosis than in the total of analyzed cetaceans and than in striped dolphins dying from other causes, considering seropositivity when PI≥40% at a serum dilution of 1:160. Causes of death in seropositive animals, however, were varied and included infectious and non-infectious causes, indicating a low specificity for the cELISA to predict neurobrucellosis. If these cetaceans with high titers represent subclinical forms of latent
Brucella infections inducing humoral immunity remains to be studied. In humans, whose pathogenesis of neurobrucellosis is often compared to cetaceans, neurobrucellosis is not the most common manifestation of disease [
31,
32] and occurs in subacute to chronic infections [
33,
34]. Thus, it is likely that the diagnosed cases presented in this work represent just the tip of the iceberg of an endemicity of
Brucella infection in the Mediterranean, as suggested previously [
14] and supported by the high seroprevalence reported in other studies around the world in many cetacean species (see [
4] for review).
The lack of specificity of serological tests is often attributed to cross-reactions with other LPS of Gram-negative bacteria. In terrestrial animals,
Campylobacter spp.,
Salmonella spp.,
Pasteurella spp.,
Yersinia enterocolytica [
11],
Francisella tularensis and
Vibrio cholerae [
35] have been associated to cross-reactions. Terrestrial bacteria, however, can not always be found in marine environments.
Salmonella spp. have been isolated from free-ranging cetaceans [
36]. In the seropositive dolphins of this paper,
Psychrobacter phenylpyruvicus (N-28/21),
Photobacterium damselae (N-557/17, N-232/18, N-42/18),
Pseudomonas spp. (N-557/17),
Vibrio sp. (N-232/18, N-329/18), and
Escherichia coli (N-329/18) have been isolated. The pathogenicity of some of these species is unclear and their role in cross-reaction is speculative.
The presence of high seropositivity in the cases of neurobrucellosis probably indicates that the humoral response is not effective against the pathogen. In our cases, Th1 response cytokines (IL-1 and TNFα) have been detected in
Brucella-induced meningitis by immunohistochemistry, which could suggest a role of this type of response in brucellosis [
37]. As speculated by the authors, it may be that in these cases, for unknown reasons, cellular response is ineffective. Polymorphisms in some molecules of the immune system may also play a role in individual susceptibility [
38,
39].
After analyzing the serological results, the authors hypothesized that the rise of antibodies, instead of conferring protection, contributes to disease progression in animals with neurobrucellosis. Nonetheless, the authors failed to find literature supporting that. Antibodies in humans with brucellosis can persist several months after the remission of symptoms, clinical relapse [
31,
32] and, in areas with endemicity, repeated infections [
31]. In rats inoculated with
B. abortus antigens, antibodies are present a minimum of 120 days [
40]. There is a report of a captive bottlenose dolphin with
Brucella osteomyelitis with sustained high antibody titers against
Brucella [
41]. Considering this, it is possible that antibodies would be indicative of chronicity rather than accountable for disease development.
Interestingly, the authors observed that juvenile striped dolphins were the age group with higher seroprevalence and proportion of neurobrucellosis. In the Italian coast, similar results have been reported, with 6/8 striped dolphins with brucellosis being juveniles [
18]. In our cases, this may be due to a larger contribution of anthropogenic interactions as cause of death for adults (21/64; 32.8%), diminishing the relative importance of neurobrucellosis. Immune immaturity as a predisposing cause for neurologic form of brucellosis in juveniles, however, can not be discarded. The results from this work contrast with some serologic in human patients, where they found a positive correlation between the age and seroprevalence of
Brucella antibodies [
42,
43].
A cluster of
Brucella-seropositive dolphins occurred in the period 2018-2021, in CNS-localized forms of CeMV (n=5) (see Supplementary File 1). The authors speculate that the previous immune suppression of the systemic phase prompted the increased replication and transmission of subclinical
Brucella, without causing death, and the persistence of
Brucella antibodies until the development of the chronic form of CeMV. Another hypothesis is that the immunosuppressive acute phase transiently limited antibody production. At least in the present cases, there is no direct correlation between the massive immunesuppression in acute CeMV infection and the development of neurobrucellosis. In other publications the two coinfections are more frequent, with 4/8 striped dolphins testing positive for both pathogens [
18].
Author Contributions
Conceptualization, M.D.; methodology, M.D. and L.M.; validation, M.D.; formal analysis, A.A.; investigation, M.D., S.W., L.P., L.G., A.P., and L.M.; resources, M.D.; data curation, L.M.; writing—original draft preparation, M.D., L.M. and M.C.; writing—review and editing, A.P., S.W., and M.C.; visualization, L.M.; supervision, M.D.; project administration, M.D.; funding acquisition, M.D. All authors have read and agreed to the published version of the manuscript.