Non-toxigenic Corynebacterium ulcerans sequence types 325 and 339 isolated from two dogs with ulcerative lesions in Italy

Abstract

Corynebacterium ulcerans, an emerging pathogen related to C. diphtheriae and C. pseudotuberculosis, is able to cause disease in both human and animal hosts. C. ulcerans may harbor acquired virulence factors such as dermonecrotic exotoxin phospholipase D (PLD) and the prophage-encoded diphtheria toxin (DT). Infections typically occur in persons reporting close contact with animals. In pets, C. ulcerans has been isolated from both asymptomatic carriers and clinically affected dogs and cats. We describe the isolation and characterization of C. ulcerans strains from 2 pet dogs with ulcerative lesions in Italy. The 2 isolates tested negative for both DT genes, but were PLD-producers and belonged to sequence types (STs) 325 and 339. These 2 cases highlight that C. ulcerans cutaneous infections might be underestimated in pets, given that many veterinary laboratories do not routinely consider and/or identify Corynebacterium species from cutaneous samples. Early detection and molecular typing of C. ulcerans is essential in order to implement effective treatment and to prevent diffusion and possible zoonotic transmission of certain STs.

Corynebacterium ulcerans, a potentially diphtheria toxin (DT)-producing species that is related to C. diphtheriae and C. pseudotuberculosis, is able to cause disease in both human and animal hosts. ¹⁵ C. ulcerans infections of humans can be fatal and have been associated with cases of classical respiratory and cutaneous diphtheria, pharyngitis, sinusitis, and extrapharyngeal disease. ¹⁵ Since the 1980s in Western Europe, cases of human respiratory diphtheria caused by C. ulcerans have progressively exceeded cases caused by C. diphtheriae. ⁹ Infections typically occur in persons reporting close contact with animals,^3,8,14,20 whereas person-to-person transmission is considered rare. ¹³ The presence of C. ulcerans has been reported in several domestic and wild animal species, including dogs^5,8,14,18 and cats.^{1,3,4,16,19,20} In companion animals, C. ulcerans has been isolated from both healthy asymptomatic carriers^5,10 and clinically affected animals with upper respiratory signs, bronchopneumonia, and/or skin lesions.^{4,14,16,18,19} The major C. ulcerans virulence factors are DT, carried by a lysogenic beta-corynephage and encoded by the classical C. diphtheriae tox gene or by a species-specific C. ulcerans tox gene, ¹⁷ and the dermonecrotic exotoxin phospholipase D (PLD), ⁷ although additional virulence factors have been identified. ¹⁵ A multilocus sequence typing (MLST) scheme for C. ulcerans typing has been developed and used to genotype isolates from different hosts for epidemiologic purposes, and to investigate whether specific sequence types (STs) possess higher pathogenic potential. ¹² We report herein the characterization of C. ulcerans isolates from 2 pet dogs in Italy with ulcerative lesions.

In September 2015, a 7-y-old male Jack Russell Terrier dog living indoors in the urban area of Milan (Lombardy, northern Italy) was presented to a veterinary clinic with a 6-mo history of scaly, multifocal alopecia on the muzzle, and swelling and ulcerative dermatitis of both pinnae (case 1). Cutaneous swabs from both pinnae were collected for cytology; a single swab was taken for microbiologic examination from the more exudative lesion on the left pinna.

In October 2015, an 11-y-old male Labrador Retriever dog living indoors in an urban area close to Milan (Lombardy, northern Italy) was presented to the same veterinary clinic with an ~1-mo history of xerosis, fissures, crusts of the left nostril, and unilateral nasal mucus discharge (case 2). Nasal swabs taken from the left nostril were collected for cytology and microbiologic examination.

In both cases, dog owners gave verbal informed consent for sampling and using data obtained from their dog and were informed that these data would remain anonymous. In both cases, cytologic examination was performed by directly rolling sterile swabs onto clean microscope slides and by microscopic observation of stained smears (Dip Quick, Dyaset, Portomaggiore, Italy).

For microbiologic examination, swabs were streaked on Columbia agar supplemented with 5% sheep blood (bioMérieux, Marcy l’Etoile, France), followed by incubation under aerobic conditions at 37°C for 24–48 h. The nasal swab from case 2 was also streaked on Sabouraud dextrose agar plates (Laboratorios CONDA, Spain), and incubated at 32°C for 72–96 h, with the plates also examined again after 1 wk. Suspected corynebacterium-like colonies were subcultured and identified as Corynebacterium spp. by means of standard techniques including colony morphology, Gram staining, catalase test, and API Coryne kit (bioMérieux). The CAMP (Christie, Atkins, Munch-Peterson) test phenomenon for Corynebacterium spp. isolates was evaluated by using Rhodococcus equi (ATCC 6939). The reverse CAMP test, for the detection of PLD activity, was performed using Staphylococcus aureus (ATCC 25923).

Identification at the species level was confirmed by PCR amplification and subsequent sequencing of a target region of the 16S ribosomal (r)DNA gene, ⁶ and of the RNA polymerase beta subunit (rpoB) gene. ¹¹ The C. ulcerans isolates were further investigated for the presence of the DT gene and of the species-specific C. ulcerans DT gene by PCR, ¹⁷ and were genotyped by using a MLST scheme, as described previously. ¹² The C. ulcerans isolates were also tested for their susceptibility to selected antimicrobials by a commercial broth microdilution method (Trek Diagnostic Systems, Westlake, OH).

In case 1, the cytologic preparation contained neutrophils and numerous rod-shaped bacteria. In case 2, neutrophils with intracellular coccal and rod-shaped bacteria were observed.

In both cases, pure cultures of white, chalk-like colonies (~0.5–1.0 mm diameter), suggestive of coryneform bacteria, were observed on Columbia blood agar after 24 h. No growth was observed on Sabouraud dextrose agar plates. Microscopic examination of Gram-stained culture smears revealed the presence of gram-positive coccobacilli of diphtheroid appearance (corynebacterium-like). Five corynebacterium-like colonies from each sample were selected and biochemically identified using the API Coryne kit. All of the selected colonies were identified as Corynebacterium spp., but were not unequivocally identified at species level (code 0111306 and 2111324 for colonies from cases 1 and 2, respectively). All of the selected colonies were catalase-positive, and gave a positive CAMP reaction with R. equi and a positive reverse CAMP phenomenon with S. aureus, the latter being indicative of PLD activity.

The 16S rDNA target sequence obtained from one isolate of each case showed a 485/486 bp homology with C. ulcerans (GenBank accession NR074467.1), and a 485/487 bp homology with C. pseudotuberculosis (NR102868.1). The rpoB target sequence allowed unambiguous identification (416/416 bp) with C. ulcerans (CP010818.1) for both isolates. The 2 isolates tested negative for both DT genes and belonged to ST325 (case 1) or ST339 (case 2).

Both isolates were considered susceptible to all of the tested antimicrobials for which Clinical and Laboratory Standards Institute human-derived clinical breakpoints for Corynebacterium spp. were available (penicillin, cefotaxime, tetracycline, ciprofloxacin, erythromycin, gentamicin, clindamycin, linezolid, rifampin, vancomycin), ² with the exception of clindamycin (minimum inhibitory concentration: 2 µg/mL = intermediate) for the isolate of case 2 (Supplementary Table 1). Indeed, when interpreted according to the European Committee on Antimicrobial Susceptibility Testing (http://www.eucast.org/) standard, this isolate was clindamycin-resistant (clinical breakpoint: R > 0.5 mg/L).

Following the results of microbiologic examination and pending those of antimicrobial susceptibility testing, the dog of case 1 was empirically treated with erythromycin (2% gel, q12h) and methylprednisolone (0.5 mg/kg body weight per os, q24h) for 2 wk. At the end of the treatment, dermatologic examination revealed significant improvement in the ear lesions. Three months later, all of the swabs collected from the pharynx, both ears, and the nasal cavities of the dog tested negative for C. ulcerans. In case 2, pending the microbiologic results, the dog was empirically treated with cephalexin for 7 d (22 mg/kg body weight per os, q12h), and the same therapy was continued for another 21 d following the microbiologic examination and antimicrobial susceptibility test results. Complete resolution of the nasal lesions was achieved 1 wk after the completion of antimicrobial treatment. Follow-up examination carried out 3 mo later confirmed the absence of nasal lesions. Also in this case, swabs collected from the pharynx, both ears, and the nasal cavities of the dog were negative for C. ulcerans.

It was not possible to collect specimens, in order to attempt C. ulcerans isolation, from the human members of the households where the dogs lived. However, all of the family members in case 1 (3 individuals: 70-y-old mother and father and 35-y-old daughter) and in case 2 (3 individuals: 53-y-old mother and father and 23-y-old son) were immunocompetent, did not complain of respiratory signs, and reported no occurrence of cutaneous lesions in the recent past. All of the household members had been vaccinated with diphtheria toxoid according to the relevant Italian regulation, which has prescribed compulsory vaccination for all newborns since 1939.

C. ulcerans is considered an emerging zoonotic pathogen that can have serious effects on human health, occasionally causing fatal infections.^9,15 Episodes of owners acquiring C. ulcerans infections from their pets have been clearly documented.^3,8,14,20 In both of our cases, the C. ulcerans isolates were tox-gene negative but produced PLD, the major virulence factor involved in caseous lymphadenitis in sheep, goats, horses, and wild ungulates. ⁷ The detection of C. ulcerans strains from healthy or diseased household pets, even if unable to produce DT toxin, should not be overlooked, given that C. ulcerans has been reported to cause severe disease in humans, such as lymphadenitis, dermatitis, subcutaneous abscesses, acute pharyngitis, and lower respiratory tract infections.^9,15 In our cases, the persons in contact with the infected dogs did not display respiratory signs or cutaneous lesions, but their C. ulcerans carrier status could not be evaluated to assess potential zoonotic transmission. Nonetheless, it is remarkable that the C. ulcerans isolate from case 1 belonged to ST325, a potentially zoonotic ST previously isolated from both symptomatic and asymptomatic humans and household pets in Europe.^12,20 The isolate from case 2 belonged to ST339, a ST325 triple-locus variant previously isolated only from the nares of an asymptomatic dog in Brazil. ⁵

C. ulcerans isolation and subsequent identification is often a challenge, frequently resulting in delayed specific therapy, patient isolation, and/or contact tracing. ¹ Moreover, biochemical differentiation between C. ulcerans and C. pseudotuberculosis can be problematic, and conventional tests are not able to discriminate between the 2 species. ⁷ In our cases, rpoB gene sequence analysis allowed accurate Corynebacterium species identification.

Antimicrobial treatment of infected companion animals, with the aim of both eliminating the infection and the subclinical carriage of a zoonotic pathogen, may be appropriate in cases such as those reported herein. Penicillin and erythromycin are considered the drugs of choice for treating diphtheria caused by toxigenic C. ulcerans in humans. ¹⁵ However, the demonstrated circulation of erythromycin-resistant C. ulcerans isolates highlights the importance of performing antimicrobial susceptibility testing prior to treatment and/or post-exposure prophylaxis. ¹⁵ In our cases, results of antimicrobial susceptibility testing, available shortly after the treatment onset, showed that the C. ulcerans isolates were susceptible to almost all of the drugs tested, including those already administered empirically. The topical application of erythromycin for 2 wk (case 1) and the systemic administration of cephalexin for 3 wk (case 2) prompted improvement or remission of clinical signs and led to negative microbiologic results. However, it must be underlined that antimicrobials should be administered to companion animals only after determining a correct etiologic diagnosis and valid antimicrobial susceptibility testing. The prudent use of antimicrobials, including the optimal selection of drugs, dosage, and duration of treatment, is critical in order to increase the probability of therapeutic success and to reduce the risk of development and spread of resistant pathogens, such as multidrug-resistant C. ulcerans.

Supplemental Material

DS1_JVDI_10.1177_1040638718764786 – Supplemental material for Non-toxigenic Corynebacterium ulcerans sequence types 325 and 339 isolated from two dogs with ulcerative lesions in Italy

Supplemental material, DS1_JVDI_10.1177_1040638718764786 for Non-toxigenic Corynebacterium ulcerans sequence types 325 and 339 isolated from two dogs with ulcerative lesions in Italy by Virginia Carfora, Fabia Scarampella, Manuela Iurescia, Valentina Donati, Fiorentino Stravino, Serena Lorenzetti, Erika Menichini, Alessia Franco, Andrea Caprioli, Antonio Battisti in Journal of Veterinary Diagnostic Investigation