First description of Mycobacterium heckeshornense infection in a feline immunodeficiency virus-positive cat

Abstract

A 13-year-old cat was presented to the veterinary clinic with poor condition, vomiting and a reduced appetite. A painful abdomen was diagnosed because of tension and defence reactions on palpation. Diagnostic laparotomy showed a thickening of the colon and caecal intestinal wall. Histopathological investigation of intestinal biopsies revealed focal severe granulomatous inflammation with numerous acid-fast bacilli in the tela submucosa. The complete blood count test showed a severe lymphopenia and anaemia, and the cat tested positive for feline immunodeficiency virus (FIV) antibodies by enzyme-linked immunosorbent assay. The cat was euthanased and necropsied. Multifocal granulomatous nodules were present in the intestines, liver and kidneys. The gastric lymph node was markedly enlarged and showed a caseous cut surface. Histopathology revealed a systemic mycobacteriosis affecting intestine, lymph nodes, liver and kidneys. The mycobacterial strain was cultured and determined by its unique 16S rRNA gene sequence as Mycobacterium heckeshornense. This is the first reported case of M heckeshornense in a cat. It was suspected that the disseminated mycobacteriosis was supported by the FIV infection.

Case Report

A 13-year-old castrated male domestic shorthair cat was presented to the veterinary clinic with vomiting, reduced appetite and a poor general condition with a body condition score (BCS) of 3/9 (body weight 6 kg). The cat had a history of skin lesions of unknown origin. At that time no skin lesions were evident. Clinical examination showed a tension of the abdomen and defence reactions on palpation. A first blood test in the veterinary clinic showed a leukopenia with a leukocyte count of 3.55 G/l.

Treatment with enrofloxacin (Baytril 5%; Bayer, 5 mg/kg SC), dexamethasone (Dexamethason; CP-Pharma, 0.25 mg/kg IM), butafosfan-cyanocobalamin (Catosal 10%; Bayer, 0.05 ml/kg SC), intravenous therapy (isotonic saline solution 0.9%; B Braun) and gastrointestinal diet (Hill’s a/d) did not improve clinical signs. Clinical differential diagnoses included a neoplastic process or a foreign body in the intestine. The radiological and ultrasound findings of the abdomen, and the cat’s continuing anorectic phase over several days, led to the decision for surgical intervention.

A diagnostic laparotomy revealed a mild thickening of the caecal and colon wall. The stomach, spleen, liver and small intestine were without gross pathological findings. To determine the cause of the thickening of the intestinal wall, two intestinal tissue samples of 6 mm in size, fixed in 4% neutral buffered formalin, were submitted for histopathological examination to the veterinary laboratory (Laboklin GmbH & Co KG, Germany). Haematoxylin and eosin staining (H&E) and Ziehl–Neelsen staining (ZN) were performed. The mucosa showed focal erosions and marked histiocytic infiltration. In the tela submucosa a focal severe granulomatous inflammation with multifocal neutrophils and mild fibrosis was present. With ZN staining, plentiful acid-fast bacilli were seen within the macrophages (Figure 1). The tunica muscularis and tela subserosa were without pathological findings.

Figure 1

Colon biopsy: marked granulomatous inflammation of mucosa and submucosa with intracellular acid-fast (red-stained) rod-like bacteria (Ziehl–Neelsen stain)

A blood test, done in the veterinary laboratory, returned negative results for feline leukaemia virus antigen by enzyme-linked immunosorbent assay (ELISA) and feline coronavirus antibody by indirect immunofluorescence antibody test. However, the blood test was positive for feline immunodeficiency virus antibody (FIV) by ELISA. Haematology revealed a decrease in erythrocytes (4.45 T/l; reference interval (RI) 5.0–10.0 T/l); haematocrit (0.23 l/l; RI 0.30–0.44 l/l); haemoglobin (70 g/l; RI 90–150 g/l); leukocytes (3.8 G/l, RI 6.0–11.0 G/l) and lymphocytes (7%; RI 15–38%); and an increase in segmented leukocytes (87%; RI 60–78%). It was also positive for anisocytosis. The differential blood count showed a decrease of lymphocytes (0.3 G/l; RI 1.0–4.0 G/l) and monocytes (0.0 G/l; RI 0.04–0.6 G/l).

The cat was euthanased because of the poor prognosis and was submitted for pathological investigation: the cat had a body weight of 6 kg, but showed a poor body condition with a BCS of 3/9. The gastric lymph node was 2.0 × 1.0 × 1.0 cm in size and had a smooth, caeseous, yellow-greenish cut surface (Figure 2). In the intestines, multifocal yellowish masses measuring approximately 2 mm² could be seen (Figure 3). Multiple yellow nodules (1–2 cm) were also present in the liver and kidneys (3 mm). Three smooth white nodules measuring 0.3–0.5 mm in diameter were found in the spleen.

Figure 2

Enlarged gastric lymph node with yellow caeseous cut surface

Figure 3

Multiple yellowish nodules (open arrows) in the intestine are visible from the serosal side and are prominent in the mucosa (inset). The site of colon biopsy is shown by the closed arrow

Tissue samples of the gastrointestinal tract, liver, spleen, kidneys, lung and heart were fixed in 4% neutral buffered formalin, embedded in paraffin and stained routinely by H&E and ZN. Histopathological findings showed nodular-to-diffuse granulomatous inflammation of the intestines, liver, spleen, kidneys and gastric lymph node. One mass in the spleen was identified as a small cell lymphoma. ZN staining of the regions with granulomatous inflammation identified multiple acid-fast, rod-like bacteria within the histiocytes in all organs. These findings were consistent with a generalised mycobacteriosis.

To differentiate the mycobacteria in this case, an unfixed tissue sample of the colon was sent to the national reference centre for mycobacteria at the Leibnitz Center for Medicine and Biosciences (Germany). There, identification of mycobacteria was attempted by direct sequence determination of the internal transcribed spacer (ITS) sequence initially. However, for the ITS sequence identified there existed no corresponding finding in the National Center for Biotechnology Information database. Furthermore, we attempted to cultivate the bacteria in vitro. The tissue sample was decontaminated using the N-acetyl-L-cysteine–sodium hydroxide method. ¹ Culture examination was done using mycobacteria growth indicator tube liquid media (Becton Dickinson Diagnostic Systems) and solid media (Middlebrook agar, Stonebrink media, Löwenstein Jensen media) at both 37°C and 31°C. Growth of mycobacteria occurred within 8 days on liquid media and 6 weeks on solid media. For identification of the colonies, amplification of ribosomal 16S RNA gene was performed. Four hundred and eighty-six base pairs of the amplified and sequenced fragment were compared using the Basic Local Alignment Search Tool (http://blast.ncbi.nlm.nih.gov/). Mycobacterium heckeshornense was identified as the causative agent with 100% sequence homology (accession number AF547934 representing the type strain).

It is possible that the lesions in the gastric lymph node were part of a complete ‘primary complex’ of the mycobacterial infection. During laparoscopy 6 weeks before necropsy, no nodules were seen in the inner organs. Several new nodules in the liver and spleen were identified during the post-mortem examination 6 weeks later. This may reflect the active process of the generalisation stage of the infection.

Mycobacterium heckeshornense was first isolated in an immunocompetent woman. ² Since then, only single case reports can be found in the literature. In humans the agent is described as a pathogen causing lung disease, lymphadenitis, tendosynovitis and peritonitis.^2–5 In animals only one case of M heckeshornense infection, detected in a slaughtered pig with mycobacterial lymphadenitis, has been reported. ⁶ This type of mycobacterium belongs to the group of slowly growing non-tuberculous mycobacteria and is phylogenetically closely related to Mycobacterium xenopi, which may result in some misidentifications of M heckeshornense as M xenopi. ² It grows at a temperature ranging from 37°C to 45°C, although 42°C seems to be the optimal temperature. ² In the present case cultural growth occurred within 8 days on liquid media at a temperature of 31°C and 37°C. The fast growth below the temperature optimum is likely because of the high number of mycobacteria in the tissue sample.

Several species of mycobacteria can cause diseases in cats.^7–10 Cutaneous mycobacteriosis appears to account for the majority of cases, mostly in areas often injured by fighting. ⁸ It is believed that mycobacteria infection is mainly caused by injury of the skin and contamination of cutaneous wounds. Entry through the gastrointestinal or respiratory tracts is less frequent. The cat in this case had a history of cutaneous lesions of unknown origin. At necropsy no cutaneous lesions were evident, so no histological examination could be done to verify the correlation. However, the enlarged gastric lymph node may indicate an oral infection with the primary complex in the gastrointestinal organs and secondary dissemination into the other organs. The cat was well known as a skilled mouse hunter. Though the isolation of the pathogen from rodents has not been shown by now, it can be speculated that the infection of the cat might have been due to the ingestion of a diseased rodent.

Although it is believed that the pathogenicity of M heckeshornense is higher than the closely related M xenopi, the immunodeficiency relating to the FIV infection in the present case seemed to be a predisposing factor for the disseminated mycobacteriosis.^2,11,12 Generalised mycobacterial infections are frequently described in correlation with a defect in the cellular immune response because of the intracellular life of the bacterium. ¹² Also, the lymphoma of this cat was suspected to be a consequence of the FIV infection. Lymphomas are found to be the most common neoplasms associated with a FIV infection. ¹³

Mycobacteriosis can be treated by using a combination of antibiotics. ⁸ Furthermore, surgical excision of granulomas would increase therapeutic success. However, treatment in the presented case would have had little prospect of success because of the disseminated form, irrespective of the lymphoma and FIV infection. Also, the risk of zoonotic transmission should be considered, although reliable data are not available.

Conclusions

This case report shows that M heckeshornense has to be considered as a pathogenic agent in the cat. Identification of the causative agent in mycobacteriosis is important because of the potential zoonotic risk. Good collaboration between veterinary clinics and different laboratories may favour the identification of uncommon pathogens like M heckeshornense.