Cutaneous epitheliotropic lymphosarcoma in a captive white catfish (Ameiurus catus Linnaeus).

A wild caught white catfish (Ameiurus catus Linnaeus) developed multiple cutaneous masses. Cytology revealed neoplastic lymphocytes and microscopy confirmed dermal infiltration with epitheliotropism in the epidermis, oral mucosa, and cornea, without internal organ involvement. Transmission electron microscopy did not identify viral particles. Histopathology supported cutaneous epitheliotropic lymphosarcoma, a condition most commonly reported in mammals. This is the first reported case of cutaneous epitheliotropic lymphosarcoma in an ictalurid and one of the few published cases of this condition in any fish species.

White catfish (Ameiurus catus Linnaeus) are native to streams and lakes in eastern North America and are important in both recreational fishing and aquaculture (Hardman & Page, 2003). Multiple pale cutaneous masses and abrasions along the dorsal fin were noted on a 500 g, adult female white catfish at the North Carolina Aquarium at Roanoke Island (Manteo, NC). Approximately, 9 months prior, the catfish was collected as a part of a large mixed group of fish obtained from two sampling locations within the Pasquotank River Basin: Scuppernong River (Columbia, NC) and Mashoes Wildlife Access within Alligator River National Wildlife Refuge (East Lake, NC). Within 1 week, the lesions had progressed and the catfish was moved from the exhibit to an off‐exhibit holding system. Water quality parameters were within normal limits with no known contaminants. Observation revealed erosions on the dorsal and pectoral fins, a deep ulceration at the base of the first dorsal fin spine, an ulceration on the left side ventral to the dorsal fin, and multifocal, soft, rubbery cutaneous masses spanning from the head to the dorsal fin (Figure 1a). The catfish was anesthetized with buffered tricaine methanesulphonate (MS‐222) at 150 ppm for physical examination. Wet‐mount preparations revealed one (incidental) turbellarian from skin scrape and no external parasites from gill biopsy. Pending evaluation of fine needle aspirate cytology, treatment with florfenicol 30% was started at 30 mg/kg intramuscularly daily for 10 days for possible systemic bacterial infection secondary to skin ulceration and salinity was increased to 0.4 ppt to aid in osmoregulation. Diff‐Quik stained cytology of a cutaneous mass revealed pleomorphic, round neoplastic cells with dark basophilic cytoplasm, a high nuclear to cytoplasm ratio, and granular chromatin. With an absence of detectable bacteria, fungi, or neutrophils, the catfish was presumptively diagnosed with lymphosarcoma. Because of the severity and continued progression of the lesions and probable poor prognosis, the catfish was euthanized and necropsied approximately 4 weeks following the first observation of lesions. Euthanasia was performed via immersion overdose in buffered MS‐222 to effect followed by severing of the spinal cord and pithing of the brain stem to ensure death prior to necropsy.

FIGURE 1

White catfish, Ameiurus catus, with cutaneous epitheliotropic lymphosarcoma. (a) Multiple, tan‐white patches, and masses (arrows) are present over the dorsum with ulceration of the cranial dorsal fin. (b) Epidermis over a cutaneous mass with prominent epitheliotropism of neoplastic lymphoid cells that occasionally form intra‐epidermal Pautrier's microaggregates (arrows). Haematoxylin and eosin. Bar =20 micrometres. (c) Transmission electron photomicrograph of the epidermis with epitheliotropism. Round neoplastic lymphoid cells (arrows) expand the intercellular space between keratinocytes (arrowhead) and neoplastic cells have indented nuclei and prominent nucleoli. Bar =2 micrometres

At post‐mortem examination, the white catfish weighed 473 g and demonstrated clinically significant wasting (5% body weight loss over 3 weeks). The superficial abrasions noted at initial presentation had developed fibrosis. Multifocal, variably sized (<0.5 cm to 2 × 3 cm) cutaneous rubbery masses consistent with previous clinical findings extended from the dorsal rostrum to the dorsal fin and varied from tan to white. There was no evidence of systemic lymphosarcoma as all internal organs lacked gross abnormalities. Samples of cutaneous masses and organs were placed in 10% neutral buffered formalin for routine histology processing. Samples of four cutaneous masses were immediately placed in buffered McDowell and Trump's fixative (4% formaldehyde and 1% glutaraldehyde) for routine transmission electron microscopy (TEM) processing to search for a potential underlying retrovirus aetiology.

Histopathology revealed nonencapsulated dermal masses composed of sheets of monotypic, enlarged neoplastic lymphoid cells in scant new connective tissue with prominent infiltration of the epidermis (epitheliotropism) and occasional formation of Pautrier's microaggregates (Figure 1b). Round nuclei were sometimes indented or had convoluted nuclear envelopes. Mild‐to‐moderate anisokaryosis and anisocytosis were observed, and zero‐mitotic figures were counted per 10 microscope fields (2.37 mm2). Epitheliotropism was well established in the oral cavity and was also noted in the peripheral cornea where it extended from a skin mass. Neoplastic cells were also noted in the gill interstitium. Internal organs lacked significant microscopic findings, with no evidence of leukaemia. TEM revealed numerous enlarged, round, lymphoid neoplastic cells in the dermis and epidermis that had atypical morphology and contained moderate amounts of increased cytoplasm, nuclei with convoluted envelopes, highly cleaved nuclei, coarsely clumped heterochromatin, and enlarged nucleoli (Figure 1c). Small Golgi bodies, scant rough endoplasmic reticulum, and few oval and indented mitochondria were present, and mitochondria contained regular tubular cristae or sometimes cristolysis. Presumed degenerating mitochondria were electron dense and membrane bound (Arismendi‐Morillo, 2009). No viral particles were detected in numerous neoplastic cells reviewed in multiple ultrathin sections from different skin masses. Gross and microscopic findings were consistent with cutaneous epitheliotropic lymphosarcoma. Immunohistochemistry for T lymphocytes was attempted to confirm the neoplastic cell type; however, no immunoreactivity of any lymphocytes in normal positive control fish tissues could be demonstrated with commercially available antibodies. Additionally, external fish immunopathologists were contacted and they confirmed the lack of commercially available immunohistochemical markers for fish lymphocytes. In this case, the ultrastructural morphology of the neoplastic cells is consistent with that of lymphocytes and helps to confirm lymphosarcoma.

This case identifies cutaneous epitheliotropic lymphosarcoma in a white catfish with no evidence of viral particles on TEM and no known exposure to carcinogenic chemicals. Neoplasia is not uncommon in fish; however, incidence and prevalence of particular neoplasms in different fish populations have rarely been quantified (Vergneau‐Grosset et al., 2017). Reports of epitheliotropic lymphosarcoma in fish are very limited. A single presumed spontaneous case of cutaneous epitheliotropic lymphoma was reported in a wild‐type zebrafish (Danio rerio Hamilton) used in a two‐generation toxicity study, treated with tetrabromobisphenol A – a substance not known to be carcinogenic (Kuiper et al., 2009). An epizootic of an epitheliotropic lymphoblastic lymphoma of thymic origin presenting with neoplastic infiltration of the gills, integument of the opercular cavity, and integument around the nares and eyes was reported in a brood group of coho salmon (Oncorhynchus kisutch Walbaum) raised in a hatchery (Kieser et al., 1991). Infectious or chemical aetiology was suspected due to high disease prevalence within the group, but diagnostics performed failed to confirm either.

Although scarcely reported in fish, cutaneous epitheliotropic lymphoma has been reported in a variety of mammal species, most notably dogs and humans (Fontaine et al., 2009) and rarely in domestic cats (Fontaine et al., 2011). Single reports have been described in other mammals including a squirrel (Honnold et al., 2007), a coatimundi (Skorinsky et al., 2008), a baboon (Carias et al., 2019), an opossum (Higbie et al., 2015), and an alpaca (Hasbach & Stern, 2016). This uncommon condition is characterized by infiltration of neoplastic T lymphocytes with a specific tropism for the epidermis and adnexal structures. The Pautrier's microaggregates (formerly microabscesses) in the epidermis of the white catfish and involvement of the oral cavity are typical of cutaneous epitheliotropic lymphoma in dogs and humans but extension to the cornea is usually not a feature (Fontaine et al., 2009). The aetiology of cutaneous epitheliotropic lymphoma remains unknown (Rook, 2019). Although, there are reports in dogs and humans speculating possible associations with viral infections and chronic dermatitis, causal relationships have not been proven (Fontaine et al., 2009).

While reports of epitheliotropic lymphosarcoma in fish are very limited, there are many descriptions of other haematopoietic neoplasms in fish species. Many haematopoietic neoplasms that have been reported in fish are associated with retroviral or chemical aetiologies, rather than spontaneous origin. Retroviral‐induced esocid lymphosarcomas in muskellunge (Esox masquinongy Mitchell) and northern pike (Esox lucius Linnaeus) manifest as non‐epitheliotropic cutaneous round cell tumours. Immunohistochemistry and ultrastructural analysis have not yet determined a T lymphocytic or histiomonocytic origin (Coffee et al., 2013). In Chinook salmon (Oncorhynchus tshawytscha Walbaum), the tentatively named Salmon Leukaemia Virus causes plasmacytoid leukaemia, resulting in accumulations of neoplastic round cells in visceral organs, including the liver, spleen, pancreas, kidney, pericardium, and retrobulbar tissue (Coffee et al., 2013). In channel catfish (Ictalurus punctatus Rafinesque), two cases of visceral lymphosarcoma were reported following exposure to N‐methyl‐N’‐nitro‐N‐nitrosoguanidine, a known potent carcinogen (Chen et al., 1996). Lymphoma was reported in a black bullhead (Ameiurus melas Rafinesque) collected as part of a survey procedure in the Allouez Bay, Lake Superior, Wisconsin (Blazer & Schrank, 1995). Neoplastic cells were detected in liver, spleen, kidney, and gill tissues on histologic examination. Chemical aetiology was proposed based on collection location, but this was not confirmed.

Although many haematopoietic neoplasms in fish have been associated with retroviral or chemical aetiology, there are also numerous reports of haematopoietic tumours speculated to be of spontaneous origin. A single case of presumed spontaneous lymphosarcoma within the family Ictaluridae was reported in a channel catfish (I. punctatus) from a commercial production operation in Mississippi, with neoplastic cells noted in liver, intestine, gill, and kidney tissues on histologic examination (Bowser et al., 1985). In a broodstock population of Japanese medaka (Oryzias latipes Temminck & Schlegel) lymphomas were reported as the most common spontaneously occurring tumour in deceased or moribund fish (Okihiro & Hinton, 1989). Single cases of presumed spontaneous lymphosarcomas have also been reported in an African tilapia (Sarotherodon spiluris spiluris Günther) (Haller & Roberts, 1980), an Astyanax sp. (Nigrelli, 1947), and a rainbow trout (Oncorhynchus mykiss gairdneri Richardson) (Bernstein, 1984).

The case of cutaneous epitheliotropic lymphosarcoma in a white catfish described in this report was presumed to be of spontaneous origin given that there was no known exposure to potential viral or chemical aetiologies such as those described above. This is further supported by the fact that no other fish in the exhibit tank developed observable neoplasia over the 9 years from the introduction of the catfish to the present. However, viral or chemical aetiologies could not be ruled out as the fish was wild caught. This report adds to the limited body of literature surrounding the prevalence and presentation of cutaneous epitheliotropic lymphosarcoma in fishes.

CONFLICT OF INTEREST

The authors have no conflicts of interest to report.