Histopathological Differential Diagnosis of Meningoencephalitis in Cetaceans: Morbillivirus, Herpesvirus, Toxoplasma gondii, Brucella sp., and Nasitrema sp.

Infectious and inflammatory processes are among the most common causes of central nervous system involvement in stranded cetaceans. Meningitis and encephalitis are among the leading known natural causes of death in stranded cetaceans and may be caused by a wide range of pathogens. This study describes histopathological findings in post-mortem brain tissue specimens from stranded cetaceans associated with five relevant infectious agents: viruses [Cetacean Morbillivirus (CeMV) and Herpesvirus (HV); n = 29], bacteria (Brucella sp.; n = 7), protozoa (Toxoplasma gondii; n = 6), and helminths (Nasitrema sp.; n = 1). Aetiological diagnosis was established by molecular methods. Histopathologic evaluations of brain samples were performed in all the cases, and additional histochemical and/or immunohistochemical stains were carried out accordingly. Compared with those produced by other types of pathogens in our study, the characteristic features of viral meningoencephalitis (CeMV and HV) included the most severe and frequent presence of malacia, intranuclear, and/or intracytoplasmic inclusion bodies, neuronal necrosis and associated neuronophagia, syncytia and hemorrhages, predominantly in the cerebrum. The characteristic features of Brucella sp. meningoencephalitis included the most severe and frequent presence of meningitis, perivascular cuffing, cerebellitis, myelitis, polyradiculoneuritis, choroiditis, ventriculitis, vasculitis, and fibrinoid necrosis of vessels. The characteristic features of T. gondii meningoencephalitis included lymphocytic and granulomatous encephalitis, tissue cysts, microgliosis, and oedema. In the case of Nasitrema sp. infection, lesions are all that we describe since just one animal was available. The results of this study are expected to contribute, to a large extent, to a better understanding of brain-pathogen-associated lesions in cetaceans.

Infectious and inflammatory processes are among the most common causes of central nervous system involvement in stranded cetaceans. Meningitis and encephalitis are among the leading known natural causes of death in stranded cetaceans and may be caused by a wide range of pathogens. This study describes histopathological findings in post-mortem brain tissue specimens from stranded cetaceans associated with five relevant infectious agents: viruses [Cetacean Morbillivirus (CeMV) and Herpesvirus (HV); n = 29], bacteria (Brucella sp.; n = 7), protozoa (Toxoplasma gondii; n = 6), and helminths (Nasitrema sp.; n = 1). Aetiological diagnosis was established by molecular methods. Histopathologic evaluations of brain samples were performed in all the cases, and additional histochemical and/or immunohistochemical stains were carried out accordingly. Compared with those produced by other types of pathogens in our study, the characteristic features of viral meningoencephalitis (CeMV and HV) included the most severe and frequent presence of malacia, intranuclear, and/or intracytoplasmic inclusion bodies, neuronal necrosis and associated neuronophagia, syncytia and hemorrhages, predominantly in the cerebrum. The characteristic features of Brucella sp. meningoencephalitis included the most severe and frequent presence of meningitis, perivascular cuffing, cerebellitis, myelitis, polyradiculoneuritis, choroiditis, ventriculitis, vasculitis, and fibrinoid necrosis of vessels. The characteristic features of T. gondii meningoencephalitis included lymphocytic and granulomatous encephalitis, tissue cysts, microgliosis, and oedema. In the case of Nasitrema sp. infection, lesions are all that we describe since just one animal was available. The results of this study are expected to contribute, to a large extent, to a better understanding of brain-pathogen-associated lesions in cetaceans.

INTRODUCTION
Infectious and inflammatory processes are among the most common causes of central nervous system involvement in stranded cetaceans (1)(2)(3). Meningoencephalitis of known infectious etiology can be caused by either bacteria, fungi, viruses, or protozoans (4).
Despite the aforementioned CNS pathogen-associated lesions, meningitis, gliosis and perivascular cuffs are usually nonspecific and may accompany various pathologies. In this paper, we systematically describe and compare the distinctive histopathological features of meningoencephalitis in stranded cetaceans induced by CeMV, HV, Brucella sp., T. gondii and Nasitrema sp..

MATERIALS AND METHODS
All the cases included in the present study were diagnosed during routine pathological and cause-of-death analyses in stranded cetaceans at the Division of Histology and Animal Pathology of the Institute for Animal Health (IUSA), Veterinary School, Universidad de Las Palmas de Gran Canaria. A prospective study on stranded cetaceans on the coasts of the Canarian Archipelago, and occasionally of other geographic regions, has been systematically carried out since 1999. The stranded animals were examined and necropsied according to standard procedures (51,52). Stranding epidemiology (type, location and date) and life history data (species, age category, sex) were systematically recorded. During the necropsy, the body condition and the decomposition code of the carcass were also evaluated. Age categories were established based on total body length (53) and histologic gonadal examinations (54) in: neonate (animals with vibrissal hairs or vibrissal crypts, unhealed navel, fetal folds, and soft and folded dorsal fin and tail flukes), calf (animals with presence of milk in their stomach, or about the size of a nursing calf), juvenile (not sexually or physically mature animals), subadult (sexually but not physically mature animals), and adult (animals with mature gonads). Five codes of conservation condition were established (52): code 1 (extremely fresh carcass, just dead as an animal that has recently died or euthanized), code 2 (fresh carcass), code 3 (moderate decomposition), code 4 (advanced decomposition), and code 5 (mummified or skeletal remains). Four categories were established for body condition according to Joblon et al. (55): very poor (animals with extremely concave dorsal profile, visible costal reliefs, body fat low or absent, and fatty serous atrophy), poor (animals with concave dorsal profile, low body fat and the ribs can be noted by palpation, fair/moderate (animals with dorsal profile straight or slightly convex and moderate body fat), and good (animals with a dorsal convex profile and abundant body fat). During necropsy, formalin-fixed and fresh unfixed samples for histopathologic and virologic analyses, respectively, were prepared from selected tissues. CNS samples included cerebrum, cerebellum, brainstem and spinal cord. Before immersion in 4% formaldehyde solution in phosphate-buffered saline (PBS; pH 7.4) some longitudinal cuts (2)(3)(4) were made in both the cerebral and cerebellar hemispheres for a more rapid fixation of deep periventricular structures (56,57). The fixed tissue samples were trimmed, routinely processed, embedded in paraffin, sectioned at a thickness of 5 µm, and stained with haematoxylin and eosin (HE) for examination by light microscopy. The unfixed samples were stored frozen at −80 • C until processing for molecular virology testing.

Histopathologic CNS Analysis
Thirty-eight animals with a morphological diagnosis of CNS inflammation and an associated aetiological diagnosis were included in the present study. Necropsy reports, including histopathological diagnostic reports of CNS, as well as epidemiologic and biologic data, photographic material, and ancillary diagnostic techniques, were retrieved and further analyzed. A few specimens included in this study have also been previously published.
Brain cortex lesions were systematically recorded (since they were consistently represented in the sample set), including subjective evaluations of meningitis, perivascular cuffing, microgliosis, malacia and neuronal necrosis and neuronophagia as absent (-), minimal (+), mild (++), moderate (+ + +), and severe (+ + ++), while INCIBs, hemorrhages and oedema were evaluated as absent (-) or present (+). Lesions affecting other regions and the presence of CNS-associated lesions were also described when present. An aetiological diagnosis of CNS inflammation was made based upon molecular techniques. However, when indicated by histopathological observations, histochemical [periodic acid Schiff (PAS) and Grocott] and/or immunohistochemical (IHC) [anti-T. gondii, anti-Brucella sp. and anti-canine distemper virus (CDV)] techniques were performed (2,58). Appropriate positive and negative immunohistochemical controls (serial tissue sections in which primary antibodies were substituted by non-immune homologous serum) were included accordingly. Histological evidence of the involvement of other aetiological agents led to complementary analyses (microbiology).
For both T. gondii and Nasitrema spp., amplification was performed in 20 µl of a reaction mixture containing 4 µl of template DNA, 1X SsoAdvanced TM Universal SYBR R Green Supermix (BioRad Laboratories, Hercules, CA), 0.375 µM of each primer and 5.250 µl of H 2 O treated with diethyl pyrocarbonate (DEPC). The real-time PCR cycle conditions were as follows: initial denaturation at 98 • C for 3 min, template denaturation at 98 • C for 15 s, followed by 40 amplification cycles of template denaturation at 98 • C for 15 s, primer annealing at 60 • C for 30 s and primer extension at 65 • C for 5 s, with a final extension at 72 • C for 5 s. The thermal cycler was a CFX96 Touch TM Real-Time PCR Detection System. The melting curve analysis was generated immediately after the amplification protocol by heating from 55 to 95 • C in increments of 0.5 • C/5 s. In order to establish the sensitivity and quantification dynamic range of these novel real-time PCR techniques, each positive control was diluted into ten-fold serial dilutions up to 10 −6 . The standard curve was measured in triplicate. Two negative controls (for extraction and amplification) and an amplification-positive control were included in each protocol.
The PCR products from positive cases were purified using a Real Clean spin kit (REAL) and sequenced (Sanger method). A BLAST search (www.ncbi.nlm.nih.gov/blast/Blast.cgi) was conducted to confirm the identity of the PCR amplicons.

RESULTS
We analyzed 38 animals with a histopathological diagnosis of meningitis, encephalitis or meningoencephalitis and 9 10 , 65) 12 14 , 65) 19 27   . Adults (n = 16) were overrepresented compared with juveniles (n = 8), subadults (n = 8), and calves (n = 6). More animals were found dead (n = 22; 57.9%) than visually confirmed to be live-stranded (n = 13; 34.2%) or found floating offshore (n = 2; 5.3%). Animals presented different body conditions: good (n = 10; 26.3%), moderate (n = 14; 36.8%), poor (n = 10; 26.3%), and very poor (n = 3; 7.9%). Neither stranding type nor body condition information was available for one animal (case no. 33). Twenty-five animals presented a "very fresh" (code 1) or "fresh" (code 2) post-mortem preservation status, 11 a "moderate post-mortem autolysis" condition (code 3) and two an "advanced post-mortem autolysis" condition (code 4). CeMV was detected in 16 animals of five different species: striped dolphin (n = 8), short-finned pilot whale (n = 4), common dolphin (n = 2), bottlenose dolphin (n = 1), and Risso's dolphin (n = 1). Detailed CeMV-associated lesions in CNS are compiled in Table 1 and mainly consisted of minimal or mild non-suppurative meningitis (Figure 1A). It was moderate and severe in one animal, respectively. Animal presenting severe meningitis was also co-infected with Brucella sp. (case 6). Perivascular cuffing (consisting of lymphocytes and plasma cells) ranged from minimal to mild and moderate. Microgliosis was present in eleven animals, varying from minimal to mild. INCIBs were detected in two animals (cases 25 and 38). Malacia was only observed in four animals, ranging from minimal to mild. Neuronal necrosis and associated neuronophagia were detected in ten animals and ranged from minimal to mild to moderate ( Figure 1B); the latter animal presented a mucormycosis-like fungal co-infection (case 31) and scattered polymorphonuclear neutrophils were observed within vessels and intermixed with the inflammatory infiltrate. Hemorrhages and oedema were present in nine and four animals, respectively. Lesions in other CNS regions included lymphoplasmacytic ventriculitis in one animal, cerebellitis in three animals, myelitis and choroiditis in four animals, respectively (cases 6,9,35,38) and polyradiculoneuritis in one animal (case 6). Case 6 presented a CeMV and Brucella sp. co-infection, and cases 35 and 38 had CeMV and HV co-infection. Associated lesions included (pyo)granulomatous inflammation [specifically, granulomatous encephalitis (case 6), suppurative meningitis (case 9), and pyogranulomatous encephalitis (case 31, mucormycosis-like co-infection)], syncytia in four animals, vasculitis in three animals and neuronal degeneration and necrosis in one animal.
HV infection was detected in 13 animals of five different species: striped dolphin (n = 7), Atlantic spotted dolphin (n = 3), bottlenose dolphin (n = 1), common dolphin (n = 1), and Cuvier's beaked whale (n = 1). HV-associated lesions are described in Table 2 and included non-suppurative meningitis in 10 animals and ranged from minimal to mild to severe (severe perivascular cuffings were observed in case 37). The latter was a case of HV and Brucella sp. co-infection. Perivascular cuffing (mainly composed of lymphocytes and plasma cells) was detected in 11 animals and ranged from minimal to mild, moderate (case 10) and severe (cases 29 and 37). Case 10 was a CeMV coinfection, and case 37 was a Brucella sp. co-infection. Microgliosis (Figure 1C) of different severity was observed in 10 animals: minimal and mild in six and four animals, respectively. INCIBs ( Figure 1D) were present in six animals and were intranuclear in all of them. Malacia was minimal and mild in two and three animals, respectively. Neuronal necrosis and associated neuronophagia were detected in nine animals and ranged from minimal to mild. Hemorrhages and oedema were also present in six and two animals, respectively. Myelitis was present in two animals (cases 10 and 37). These two cases presented coinfections with CeMV and Brucella sp., respectively. Choroiditis was present in five animals (cases 29,32,34,35,38). Co-infection with CeMV was detected in cases 35 and 38 and with S. aureus in case 34. Syncytia were also present in a case co-infected with CeMV (case 38).   Brucella spp. was detected in seven animals of five different species: striped dolphin (n = 2), Atlantic spotted dolphin (n = 2), bottlenose dolphin (n = 1), common dolphin (n = 1), and shortfinned pilot whale (n = 1). The main Brucella spp.-associated lesions are compiled in Table 3 and included non-suppurative meningitis in all the animals, being severe in five animals, and, moderate in two animals (Figures 2A-C). Perivascular cuffings (mainly composed of lymphocytes and plasma cells) were detected in six animals and ranged from moderate in four animals to severe in two animals. Microgliosis was present in four animals, ranging from minimal to mild. Minimal malacia was present in two animals. Mild or moderate neuronal necrosis and associated neuronophagia were detected in one and two animals, respectively. Hemorrhages and/or oedema were not observed in this group of animals. Polyradiculoneuritis (Figure 2D) or neuritis (in five animals), ventriculitis (in one animal), choroiditis and cerebellitis (in three animals, respectively), and myelitis (in four animals) were the main lesions observed in other regions. Associated lesions included (pyo)granulomatous inflammation [specifically, pyogranulomatous meningitis (case 5), granulomatous encephalitis (case 6), and pyogranulomatous meningocerebellitis (case 22)], fibrinoid necrosis of vessels and vasculitis ( Figure 2D) (in two animals, respectively).
All the animals testing positive for T. gondii were Atlantic spotted dolphins (n = 6). All T. gondii-associated lesions were of similar severity (detailed description in Table 4). Minimal meningitis and perivascular cuffing ( Figure 3A) were present in five and six animals, respectively, and were largely composed of lymphocytes with few plasma cells or macrophages. Minimal microgliosis was observed in four animals. Minimal or mild malacia was observed in two and one animals, respectively. Mild or minimal neuronal necrosis and neuronophagia were detected in one and three animals, respectively. Hemorrhages were detected in one animal and oedema in two animals. Other lesions detected in this group of animals included choroiditis and cerebellitis in one and two animals, respectively. Associated lesions included granulomatous inflammation (specifically granulomatous encephalitis) ( Figure 3B) in all the animals and the presence of tissue cysts (confirmed by PAS staining and IHC) in four animals.
Nasitrema sp. was detected in a bottlenose dolphin (case 20). Nasitrema sp.-associated lesions included minimal non-suppurative meningitis and perivascular cuffing and moderate malacia. Oedema was also present. Lesions in other regions included meningomyelitis. Associated lesions were pyogranulomatous inflammation, vascular necrosis and vasculitis, and intralesional sections of characteristic golden-brown triangular eggs.
Number and percentages of animals displaying every type of lesion (including the severity of morphological lesion) grouped by associated pathogens are available in Table 5.

DISCUSSION
Meningitis and encephalitis are among the leading known natural causes of death in stranded cetaceans and may be caused by a wide range of pathogens. In some cases, brain is the only organ affected, thus, many infectious diseases may be overlooked if brain is not carefully investigated. This study describes histopathological findings in post-mortem brain tissue specimens from stranded cetaceans associated with five relevant infectious agents: viruses (CeMV and HV; n = 29), bacteria (Brucella sp.; n = 7), protozoan (T. gondii; n = 6) and helminths (Nasitrema sp.; n = 1). Aetiological diagnosis was established by molecular methods. Characteristic, but not pathognomonic, histological alterations associated with each infection are discussed below and compared. In the case of Nasitrema sp. infestation, lesions are just described since just one animal was available.
As in humans, viral meningoencephalitis is the most common type of meningoencephalitis in cetaceans. Most of the animals from our study (76.3%) presented viral meningoencephalitis. Sixteen animals with a molecular diagnosis of CeMV infection in the CNS, representing 42.1% of the total animals screened for this pathogen in our study, and 13 animals with a molecular diagnosis of HV infection in the CNS, representing 48.1% of the total animals screened for this pathogen in our study, are reported.
CeMVs are RNA viruses responsible for massive dieoffs worldwide and include three well-characterized strains (porpoise morbillivirus, DMV, and PWMV) and three less wellcharacterized strains detected in Hawaii and in the southern hemisphere (32). CeMV is a well-recognized neurotropic pathogen and localized brain lesions have been described in cetaceans that have cleared systemic CeMV infection, resembling subacute sclerosing panencephalitis (SSPE) and old dog encephalitis (ODE) (20,32,65). IHC labeling has been successfully used to diagnose CeMV preferentially in earlier stages of the disease. Herpes simplex virus encephalitis is the most common cause of sporadic fatal encephalitis in humans worldwide. In cetaceans, alphaherpesvirus (Herpesviridae family) infections have been less described than gammaherpesvirus infections, and associated lesions range from incidental and asymptomatic to necrotizing inflammation in different organ systems. Classical CNS lesions have been associated with the presence of alphaherpesviruses in some cases (33)(34)(35)(36). Immunohistochemistry has failed to consistently highlight HVinfected cells in brain samples in cetaceans (immunoreactivity has been proven in only two studies) (33,69).
In our study, CeMV-and HV-associated brain lesions were in concordance with previous descriptions (20,32). Comparing these results to those for lesions caused by other aetiological agents under study, we observed that mild malacia was more frequently detected in HV-positive animals, while it was mainly minimal in animals infected by T. gondii, Brucella sp., and CeMV. INCIBs were only detected in animals positive for HV or CeMV, being intranuclear in HV-infected animals. Neuronal necrosis and associated neuronophagia were predominantly minimal in animals positive for HV, T. gondii, and CeMV. Syncytia were more frequently detected in CeMV-positive animals. In one HVpositive animal, the presence of syncytia was reported (case 38, also co-infected by CeMV). Hemorrhages were present, with  Bacterial meningoencephalitis was the second most common cause of encephalitis in our study (18.4%). Seven animals with a molecular diagnosis of Brucella sp. infection in the CNS were found, representing 29.2% of the total animals screened for this pathogen in our study. Two Brucella sp.-PCR-positive animals were immunostained with the Brucella antibody (21,58). Brucella infection is reported for the first time in the short-finned pilot whale species (n = 1). The information presented here increases the number of confirmed Brucella sp.-positive cases within the Canarian archipelago from two previously reported cases to seven.
Brucellosis is a worldwide zoonosis characterized by its clinical polymorphism. In humans, neurobrucellosis (NB) is an uncommon complication of the infection (occurring in 0.5-25% of cases), in which meningeal involvement is the most common presentation (70)(71)(72). Human NB also includes encephalitis, myelitis, radiculoneuritis, brain or epidural abscesses, granuloma, and demyelinating and meningovascular syndromes (73). In cetacean brucellosis, the most frequent lesions involve the CNS; the cerebellum, brainstem, spinal cord, and medulla oblongata, with less frequent involvement of the cerebral cortex, are the most consistently affected regions (8,20). Brucella species antigens in phagocytic cells can be highlighted by IHC (20,58). In our study, histological analysis revealed Brucella sp.-associated brain lesions of different severity and frequency matching previous descriptions (8,20). Polyradiculoneuritis was only observed in CNS samples of animals infected by two different pathogens, with a higher prevalence in Brucella sp.-positive animals than CeMV-positive animals (case 6 was also co-infected by Brucella sp.). A predisposition to cranial nerve involvement in NB could be due to the pathogen's predilection for the base of the cranium (74). Choroiditis was a more common finding in animals from our study infected by Brucella sp. than in those infected by other pathogens, such as HV, CeMV, and T. gondii. Lymphoplasmacytic ventriculitis was more common in Brucella sp.-infected animals than in CeMVpositive animals. Vasculitis and fibrinoid necrosis of vessels were also more common among Brucella sp.-infected animals from our study than in animals infected with other pathogens. Vasculitis was also observed in a low percentage in CeMVpositive animals.
Protozoan meningoencephalitis was the third most common cause of encephalitis in our study (15.8%). Six animals with a molecular diagnosis of T. gondii infection in the CNS were seen, representing 37.5% of the total animals screened for this pathogen in our study. All the PCR-positive animals consistently immunostained for T. gondii antibody (1, 2). Toxoplasma gondii is a neurotropic protozoan globally distributed among mammalian hosts, including humans. Nonsuppurative meningoencephalitis due to T. gondii has been sporadically described in cetaceans (1,2,20,43). Tissue cysts and zoites are confirmed by IHC labeling (2,43). In our study, histological analysis evidenced T. gondii-associated brain lesions of different severity and frequency. These lesions are similar, except for the less prominent perivascular cuffing, to previous descriptions (20,43). Comparing these results to those for lesions caused by other aetiological agents under study, we observed that minimal microgliosis was a common feature caused by T. gondii and by the different pathogens under study: CeMV, HV, and Brucella sp. Oedema was more frequently present in animals positive for T. gondii than in animals positive for CeMV or HV. Granulomatous (in addition to lymphocytic) encephalitis was observed in all T. gondii-positive animals, higher than the rates in animals positive for CeMV and Brucella sp. Tissue cysts of T. gondii were frequently observed.
Helminthic meningoencephalitis was underrepresented in our study (2.6%), since just one animal was tested for Nasitrema sp., yielding positivity.
Nasitrema spp. are trematodes that normally inhabit the pterygoid sinuses and tympanic cavities of odontocetes (75). Non-suppurative meningoencephalitis has been associated with aberrant migration of this parasite. Sections of adult trematodes and eggs surrounded by multinucleated giant cells can be detected within parenchymal brain lesions. Neuritis of the eighth cranial nerve and otitis media can be occasionally present (20). Histological features of Nasitrema sp.-positive animal from our study included minimal meningoencephalitis, mild malacia, oedema, myelitis, pyogranulomatous encephalitis, vasculitis and fibrinoid necrosis of vessels.
In summary, compared with those produced by other pathogens in our study, the characteristic features of viral meningoencephalitis (CeMV and HV) included the most severe and frequent presence of malacia, INCIBs, neuronal necrosis and associated neuronophagia, syncytia and hemorrhages, predominantly in the cerebrum. The characteristic features of Brucella sp. meningoencephalitis includes the most severe and frequent presence of meningitis, perivascular cuffing, cerebellitis, myelitis, polyradiculoneuritis, choroiditis, ventriculitis, vasculitis, and fibrinoid necrosis of vessels. The characteristic features of T. gondii meningoencephalitis include lymphocytic and granulomatous encephalitis, tissue cysts, microgliosis, and oedema. However, histopathological findings in these cases can be influenced by superimposition by simultaneous or secondary infections. Specifically, three animals presented CeMV and HV brain co-infection; one case was co-infected by CeMV and Brucella sp.; and one was co-infected by HV and Brucella sp. Co-infection by HV and S. aureus and co-infection by CeMV and mucormycosis-like lesions were detected in one animal, respectively.

CONCLUSION
A multidisciplinary approach is needed for the early detection and surveillance of emerging and reemerging pathogens (76). Histopathological findings may suggest a list of differential diagnoses, but the use of additional laboratory techniques (microbiology, IHC and/or PCR) is essential to determine the specific infectious etiology. However, when these methods (specially molecular assays) are not accessible or fail to identify causative agents, histopathology is particularly useful as part of this multidisciplinary approach. The results of this study are expected to contribute, to a large extent, to a better understanding of brain-pathogen-associated lesions in cetaceans.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
Ethical review and approval was not required for the animal study because no experiments were performed on live animals because our work was based on dead stranded cetaceans, and the field studies did not involve endangered or protected species.

AUTHOR CONTRIBUTIONS
ES analyzed the data, drafted the manuscript, contributed to the gross, histological and molecular diagnosis of the cases, and designed new molecular assays. AF contributed to the gross and histological diagnosis of the cases and guided the ES during the drafting and publication process. IF-J contributed to the molecular analysis of the cases. DZ contributed to the histopathological and immunohistochemical diagnosis of the cases. JD-D contributed to the gross and histological diagnosis of the cases. RP-L, NC, FC, PD-S, and CS-S contributed to the gross and histological diagnosis of the cases. MA contributed to the gross diagnosis of the cases and guided the ES during the drafting process. All authors gave final approval of the version to be published.

FUNDING
This work was supported by a national project (refs. CGL2015-71498-P, PGC2018-101226-B508I00, and MAC/1.1b/149). This study was possible thanks to the Canary Islands Government, which funded the stranding network.