Haemolytic differential identification of Arcanobacterium haemolyticum isolated from a patient with diabetic foot ulcers

Introduction: Arcanobacterium haemolyticum (formerly known as Corynebacterium haemolyticum) is the causative agent of sore throat and also causes skin and soft tissue infections in diabetes patients. A. haemolyticum is a Gram-positive, catalase-negative, β-haemolytic bacillus. A. haemolyticum poses a diagnostic challenge in the hospital laboratory because most coryneform bacilli are considered as normal flora or contaminants, and it is therefore difficult to differentiate from β-haemolytic streptococci by colony characteristics. Case presentation: A. haemolyticum was isolated from a diabetic patient with foot ulcers and the isolate was identified by using a VITEK-2 system, CAMP inhibition test, reverse CAMP test and a 23S rRNA gene sequence analysis. The isolated A. haemolyticum inhibited haemolysis of Staphylococcus aureus in the CAMP test and enhanced haemolysis of Streptococcus agalactiae in the reverse CAMP test. The diabetic patient was treated with teicoplanin and imipenem, and the ulcers healed within 2 weeks. Conclusion: The present study suggests that a haemolytic differential method using the CAMP inhibition and reverse CAMP tests can be useful for differentiating A. haemolyticum from β-haemolytic streptococci.


Introduction
Arcanobacterium haemolyticum (formerly known as Corynebacterium haemolyticum) is the causative agent of skin infections and sore throat such as exudative pharyngitis and tonsillitis. It was first described by MacLean in 1946(MacLean et al., 1946Collins et al., 1982). A. haemolyticum is known to be part of the normal flora on the throat and skin, and mainly causes sore throat in young people. Occasionally, it causes skin and soft tissue infections, and in rare cases it can cause osteomyelitis, pneumonia, endocarditis and septicaemia (Waller et al., 1991;Mackenzie et al., 1995). A. haemolyticum is a b-haemolytic, Gram-positive, rod-shaped and catalase-negative bacillus, and can be characterized by its ability to inhibit b-haemolysis in a CAMP test. In the present study, we report a case of A. haemolyticum isolated from a patient with diabetic foot ulcers.

Case report
A 42-year-old male patient suffered from severe oedema, fever and pain from deep wounds on both feet that were not the result of trauma at the time of hospitalization. The deep wound infection on his left foot also had a foul odour (Fig. 1a, right panel). The patient had been diagnosed previously as having diabetes with high blood pressure and had been treated with hypoglycaemic and antihypertensive enhancers orally, as well as insulin intravenously. At the time of hospitalization, the patient's vital signs were a body temperature of 38.7 uC, blood pressure of 110/70 mmHg, a pulse of 70 beats min 21 and a respiratory rate of 20 beats min 21 . The patient's body temperature rose steadily to a peak of 39.9 uC. The number of leukocytes and the level of C-reactive protein in peripheral blood were determined to be 19 000 mm 23 and 202.42 mg l 21 , respectively. During hospitalization, samples from his foot lesions were collected nine times for bacterial culture and blood samples were collected for blood cultures. Blood cultures were conducted a total of four times with three sets each time. The patient was administered intravenously with ciprofloxacin following hospitalization in accordance with empirical therapy procedures.

Diagnosis
The samples from the foot lesions were cultured on 5 % sheep blood agar (SBA) and a MacConkey agar plate under conditions of 5 % CO 2 for 24 h. Very small grey colonies without observable b-haemolysis were found on the SBA plate (Fig. 1b, left panel), but no colonies were found on the MacConkey agar plate (data not shown). The isolates from the colonies were identified as Gram-positive rods by Gram staining and a catalase reaction was negative (data not shown). The colonies became more distinct and displayed a small degree of b-haemolysis after 48 h of culture (Fig. 1b, middle panel). Small smooth grey colonies with distinct b-haemolysis were observed after 72 h of culture ( Fig. 1b, right panel). The isolate was identified as A. haemolyticum (99 %) by the VITEK-2 automated microbiology system (BioMérieux) ( Table 1). Blood samples were incubated four times in the BACTEC 9240 blood culture system (BD) for 5 days, but all blood cultures were negative (data not shown).
A CAMP inhibition test was then performed with the isolated A. haemolyticum in order to confirm its identity. Staphylococcus aureus ATCC 25923 was streaked in a straight line across the centre of the SBA plate (Fig. 1a, indicated as 'a'). The isolated A. haemolyticum was streaked in a straight line perpendicular to the Staphylococcus aureus streak ( Fig.  2(b). The isolated A. haemolyticum haemolysis inhibition using Staphylococcus aureus ATCC 29213 was apparent within 24 h (Fig. 2b, yellow arrow in left panel) and increased after 48 h (Fig. 2b, yellow arrow in right panel). In contrast to the result in Fig. 2(a), haemolysis by Streptococcus agalactiae was not observed (Fig. 2b, blue arrow in both panels) due to enhanced haemolysis in Staphylococcus aureus ATCC 29213. An effective method to differentiate A. haemolyticum In the reverse CAMP test, the isolate A. haemolyticum was streaked in the upper left (Fig. 2c) perpendicular to the Streptococcus agalactiae streak. The waxing crescentshaped haemolysis on the contact surface was observed clearly after 24 h. The control A. haemolyticum ATCC 9345 was streaked in the lower right (Fig. 2c), perpendicular to the Streptococcus agalactiae streak. Similar to the isolate A. haemolyticum, the waning crescent-shaped haemolysis on the contact surface was observed clearly after 24 h. Finally, the isolated A. haemolyticum was confirmed by genotyping using 23S rRNA gene PCR and sequencing. The sense and antisense primers were 59-TAACGGTCCTA AGGTACCGA-39 and 59-GATAGGGACCGAACTGTCTC-39, respectively. The sequencing result revealed that the isolated A. haemolyticum shared 99.9 % identity with A. haemolyticum DSM 20595 (data not shown).

Treatment
The patient was administered intravenously with ciprofloxacin following hospitalization in accordance with empirical therapy procedures, but recovery was not observed in the early stages of treatment. There were no official guidelines for imipenem treatment available for the isolated A. haemolyticum by the Clinical and Laboratory Standards Institute although this pathogen was identified from the patient. Therefore, we used a similar Corynebacterium sp. as a reference. We performed an antibiotic susceptibility test with the isolated A. haemolyticum to teicoplanin and imipenem (in the carbapenem class of antibiotics) using an E-test. The MICs of teicoplanin and imipenem were 0.47 and 0.016 mg ml 21 , respectively. The patient was administered intravenously with imipenem twice per day and teicoplanin once every 2 days according to the results of the antibiotic susceptibility test. The vital signs of the patient returned to normal, and additional bacterial growth was not detected after 2 weeks of treatment with antibiotics.

Discussion
Although A. haemolyticum isolated from respiratory samples typically produces rough colonies without bhaemolysis, the isolate from the wound skin infection produced smooth colonies with haemolysis (Carlson et al., 1994). As expected, the isolated A. haemolyticum was identified as a b-haemolytic coryneform bacillus. Because coryneform bacilli are considered part of the normal flora or a contaminant, they are more difficult to identify (Meyer & Reboli, 2005 (Kim et al., 2004;Bae et al., 2010).
This report sought to redefine the role of the CAMP inhibition and reverse CAMP tests in identifying and differentiating A. haemolyticum from b-haemolytic streptococci using Staphylococcus aureus and Streptococcus agalactiae.
In the CAMP inhibition test, Staphylococcus aureus ATCC 29213 caused more haemolysis inhibition than Staphylococcus aureus ATCC 25923 (Fig. 2a, b). The reverse CAMP test with the isolated A. haemolyticum from the diabetic patient exhibited a waxing crescent-shaped haemolysis, and the reference A. haemolyticum ATCC 9345 exhibited a waning crescent shape haemolysis (Fig. 2c).
In conclusion, this report suggests that haemolytic differential identification using CAMP and reverse CAMP tests might be a more effective method to differentiate A. haemolyticum from other Gram-positive coryneform bacillus and b-haemolytic streptococci.