Article Text


Toxigenic Corynebacterium diphtheriae isolated from a wound in a horse
  1. B. A. Leggett, BSc, MSc1,
  2. A. De Zoysa, PhD2,
  3. Y. E. Abbott, MISCT1,
  4. N. Leonard, MVB, PhD, MRCVS1,
  5. B. Markey, MVB, PhD, MRCVS1 and
  6. A. Efstratiou, PhD2
  1. 1 Veterinary Sciences Centre, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland
  2. 2 Respiratory and Systemic Infection Department, Health Protection Agency, Centre for Infections, 61 Colindale Avenue, London NW9 5EQ
  1. E-mail for correspondence: bernadette.leggett{at}

Statistics from

Corynebacterium species are a widely distributed group of bacteria, typically found in the environment in soil, and on the skin and mucous membranes of human beings and animals. Corynebacteria have also been documented to survive for long periods of time on objects that have been touched by infected individuals. Several Corynebacterium species are pathogenic for both human beings and animals, for example, Corynebacterium diphtheriae, Corynebacterium ulcerans and Corynebacterium pseudotuberculosis. C diphtheriae and C ulcerans possess the potential to produce a lethal toxin, which is encoded by the tox gene carried on a bacteriophage and causes either respiratory or cutaneous lesions (Efstratiou and others 2000, Anon 2008a).

The first record of diphtheria was made by Hippocrates in the 4th century BC. Epidemics of the disease swept through Europe and the USA in the 17th, 18th and 19th centuries (Efstratiou and others 2000, Anon 2008a). In 1929, Ramon developed the diphtheria toxoid vaccine (Efstratiou and others 2000, Anon 2008a). Due to widespread vaccination in the 1930s and 1940s, the incidence of diphtheria dramatically declined in developed countries (Anon 2008a). In Ireland, the last known case of a diphtheria-related death was in 1967 (Anon 2007), and in the UK two recorded deaths occurred in unimmunised children, in 2008 and 1997 (Anon 2008a). In addition, (Hogg and others 2009) reported a fatal case of diphtheria in an elderly woman in the UK, due to the possible zoonotic transmission of toxigenic Corynebacterium ulcerans from a companion animal. However, diphtheria still remains a problem in parts of eastern Europe, Asia, South America and Africa where vaccine uptake is generally low and where basic hygiene or healthcare may be lacking (Anon 2008b).

This short communication describes the isolation of a toxigenic C diphtheriae from a non-healing, pyogenic stake wound in a horse. The potential for horses to act as a reservoir of this human pathogen is discussed.

A 28-month-old male thoroughbred was presented at the University Veterinary Hospital, University College Dublin (UCD), with a discharging wound of eight months' duration. The injury had been acquired following a fall on an upturned stake. Despite initial antibiotic treatment and extensive local excision by the attending veterinary surgeon, a discharging sinus tract had developed. The animal was subsequently referred to UCD for specialist treatment and surgery. Upon examination, the horse was alert, responsive and in relatively good condition. The wound, which was situated on the right side of the thorax, was discharging copious amounts of pus. Samples were collected and sent to the bacteriology laboratory for culture and analysis. The presence of a foreign object was not detected by ultrasonography or radiography.

Surgery was performed to open the fistulous canal and clean and debride the area. During the procedure, two drains were placed in situ and the surgical openings were closed following lavage with 0.9 per cent normal saline (Baxter) and 10 per cent gentamicin solution for injection (Gentaject; Bayer).

Despite treatment, by 72 hours after surgery the wound was again discharging large amounts of purulent exudate. Daily flushing of the area with diluted iodine (1 per cent in aqueous solution; C+M) and 10 per cent enrofloxacin solution for injection (Baytril; Bayer) was instituted, which decreased the production of pus. It was decided that further investigative surgery was required. Exploration of the fistulous canal revealed the 13th rib to be fractured. The broken rib was removed and the area was again flushed with 0.9 per cent normal saline and 10 per cent enrofloxacin solution for injection. Postoperatively, the horse was treated with 7.5 mg/kg enrofloxacin (Baytril; Bayer), administered orally, and the wound was lavaged with 10 per cent enrofloxacin solution for injection. The wound healed fully without further incident.

Swabs of the pus taken when the horse was admitted were inoculated on to Columbia agar supplemented with 5 per cent sheep blood, Columbia blood agar supplemented with colistin-nalidixic acid and MacConkey's agar number 2 (Oxoid). The plates were incubated at 37°C, aerobically and anaerobically. After 24 hours, a β-haemolytic Streptococcus species was isolated, along with small, slow-growing colonies that were non-haemolytic, dull and dry in appearance. These colonies were friable and tended to break apart on contact with a metal loop. The streptococcal isolate was identified as Streptococcus zooepidemicus (profile 1551 2061 151) using the ID32 Strep API identification strip (bioMérieux). Antibiotic susceptibility tests were carried out for both isolates. On Gram staining, the dry colonies appeared to consist of pleomorphic Gram-positive diphtheroids, demonstrating the classic ‘Chinese letter’ appearance consistent with coryneform bacteria. In addition, the organism was catalase-positive (Begg 1994, Engler and others 1997). The isolate was confirmed as C diphtheriae var gravis (profile 1010326) using the API Coryne identification system (bioMérieux). Owing to concerns that the isolate might be toxigenic, it was submitted to the Health Protection Agency/World Health Organisation Streptococcus and Diphtheria Reference Unit, Respiratory and Systemic Infections Department, London, for toxin testing and molecular typing. Toxin production was detected using both the conventional and modified Elek test, which is the gold standard for detection of diphtheria toxin (Elek 1949, Briassoulis and others 2005, De Zoysa and others 2005). A test for cysteinase production was also performed using Tinsdale agar (Oxoid). The presence of black pinpoint colonies (tellurite reductase activity) surrounded by a brown halo (cysteinase enzyme activity) denoted the production of cysteinase. The Elek and cysteinase tests were both positive for the C diphtheriae isolate (Efstratiou and Maple 1994, Efstratiou and others 2000).

Molecular typing was undertaken on the isolate by ribotyping, which utilises a biotinylated cDNA probe derived from 16S and 23S rRNA of C diphtheriae type strain NCTC 11397 (6). The gene profile of the strain is shown in Fig 1. The profile of the strain was analysed using the software programme Bionumerics (Applied BioMaths) and was compared with 86 distinct ribotypes on the diphtheria surveillance network DIPNET ( ribotype database. The ribotype profile differed from all the profiles on the database. The closest match showed a similarity of 80 per cent to a ribotype (designated Breaza) that was seen among Romanian strains isolated in 1994 (a 95 per cent similarity cut-off value is used for delineating a ribotype) (De Zoysa and others 1995, 2005).


Bst EII rRNA gene profile of a toxigenic Corynebacterium diphtheriae strain isolated from a wound in a horse (Lane 1). Lane 2 λ Hind III digests as a size standard; sizes (kb) are indicated on the right

To the authors' knowledge, this is the first animal isolate of a toxigenic C diphtheriae in Ireland. Horses have been noted to be carriers of C diphtheriae (Funke and others 1997, Henricson and others 2000). The case reported here is of concern, as it may indicate the potential emergence by a zoonotic route of a rare disease. As diphtheria is a notifiable disease in Ireland, a public health investigation was instigated in response to these findings. Swabs were taken from the human contacts of the horse and were all negative. This may have been because the animal had not yet been broken in, and so had had little human contact. As Corynebacterium species are found in the environment, the authors speculate that the wooden stake that caused the original injury to the horse was the transmission vector for C diphtheriae (De Zoysa and others 1995, Efstratiou and others 2000, Anon 2008a). The ribotype profile obtained suggested a strain of C diphtheriae that could be unique to horse populations, as it was dissimilar to ribotypes previously observed in human isolates (Colman and others 1992, De Zoysa and others 1995, Efstratiou and others 2000).


The authors thank Nicolas Vos for his clinical notes concerning this case. They also acknowledge Florien Jenner for her assistance in interpreting the clinical details of the case.


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.