Mycobacterium marseillense is a member of the M. avium complex (1) that has caused infections with lymphatic or pulmonary involvement sporadically in humans (2–4). We report M. marseillense infection involving facial skin in an immunocompetent woman in eastern China.

In April 2018, a 59-year-old woman was referred to our institute (Institute of Dermatology, Chinese Academy of Medical Sciences and Peking Union Medical College, Nanjing, China) for a 4-year history of an erythematous plaque with ulceration located on the right cheek. The primary lesion was a small erythematic patch that gradually developed into an asymptomatic ulcerative plaque (i.e., the plaque had no heat, swelling, pain, or pruritus). She also reported occasional bloody, purulent nasal discharge over the course of 2 years. Two years before visiting our hospital, cutaneous tuberculosis was suspected, so she received treatment for tuberculosis (rifampin, isoniazid, ethambutol, pyrazinamide) for 10 months. No obvious improvement was observed with this treatment. Her medical history was otherwise unremarkable.

Figure

Figure. Skin lesions and computer tomography scans of woman with Mycobacterium marseillense skin infection, China, 2018, and genomic analysis of isolate. A, B) Facial skin lesion of woman with M. marseillense infection...

On physical examination, an infiltrated erythematous plaque with yellow scales and crusts on the right cheek was visible (Figure, panel A). Routine laboratory tests showed no remarkable findings. The results of autoantibody and HIV tests were negative, and immune subset cell counts were unremarkable. Histologic examination showed infiltration of a large number of lymphocytes, plasma cells, and neutrophils and some tissue cells in the dermis (Appendix 1 Figure 1). Computed tomography scan of the paranasal sinuses showed bilateral maxillary, right ethmoid, and frontal sinusitis (Figure, panel C). Culture and PCR for mycobacteria in nasal discharge yielded negative findings.

After 3 weeks of skin tissue culture at 32°C in Löwenstein–Jensen medium, we observed smooth, yolk-yellow bacterial colonies (Appendix 1 Figure 2). Ziehl-Neelsen staining confirmed the cultured organism was acid-fast bacilli. Sequence analysis indicated that the complete genetic sequence of 16S rRNA was 99.0%, hsp65 100%, and rpoB 99.8% homologous with M. marseillense strain FLAC0026. Phylogenetic analysis of the 16S rRNA sequence showed the isolate clustered with M. chimaera and M. intracellulare (Figure, panel D). Although the 16S rRNA gene sequence of the isolate was 100% similar to M. intracellulare subsp. yongonense 05-1390, the sequence similarities to hsp65 and rpoB were relatively low. Sequence analyses suggested M. marseillense infection.

Referring to the guidelines for pulmonary M. avium complex disease, we treated the patient with the antimicrobial drugs clarithromycin, rifampin, and ethambutol (5). Afterward, in vitro drug susceptibility testing showed the isolate was sensitive to clarithromycin, azithromycin, and amikacin; moderately sensitive to moxifloxacin; and resistant to ethambutol and rifampin. Therefore, 3 months after initiating treatment, we changed the regimen to clarithromycin, moxifloxacin, and amikacin, which she received for 2 months. The patient’s skin lesions healed gradually, and nasal symptoms disappeared, but a scar and erythema remained (Figure, panel B). Computed tomography scans of the paranasal sinuses showed the reduction of sinusitis (Figure, panel C). No recurrence was observed during 4 months of monitoring.

We characterized this isolate’s genome (GenBank accession no. VASI0000000) further to help determine the cause of its virulence and resistance (Appendix 1 Figure 3). Genetic analyses indicated the genome (≈5,706,022 bp) contained 5,343 predicted genes, 3 rRNAs, and 48 tRNAs and had a GC content of 67.73%. We annotated the genes functionally through multiple databases (Appendix 1 Table 1, Figure 4). Using the Virulence Factors of Pathogenic Bacteria database, we identified 137 potential virulence genes (identity >95.0%, E value <1 × 10^–5), such as type VII secretion system genes (e.g., esxH, esxC, esxH, and esxC) (6), in the isolate’s genome (Appendix 2). In Comprehensive Antibiotic Resistance Database searches, we detected the antimicrobial drug resistance genes mtrA, murA, and gyrA (identity >90.0%, E value <1 × 10^–5; Appendix 1 Table 2); mtrA modulates antimicrobial drug efflux, murA encodes the fosfomycin resistance protein, and gyrA encodes the fluoroquinolone resistance protein.

M. marseillense infections are rare in humans. Our case demonstrates that M. marseillense can cause infections in immunocompetent persons. For facial skin infection with M. marseillense, this and similar (7) reports indicate the need for vigilance of paranasal sinus infection. Although many potential virulence factors could be detected by genomic analysis, cases of infection and transmission with this bacterium are rarely reported, suggesting the presence of other influencing factors.

The drug resistance mechanisms of M. marseillense have not been completely elucidated. The drug susceptibility test results and treatment response we observed were generally consistent with those previously reported for cases of pulmonary infection, although sensitivity to rifampin and quinolones yielded various results (2–4). Drug susceptibility testing indicated that the isolate we obtained was resistant to ethambutol and rifampin. However, in genetic analyses, mutations associated with ethambutol and rifampin resistance were not detected. According to the Comprehensive Antibiotic Resistance Database, our isolate was resistant to fluoroquinolone, but drug susceptibility test results were inconsistent. Our results indicate that drug susceptibility testing should be performed for M. marseillense to guide antimicrobial drug treatment. If drug susceptibility results are absent, treatments including macrolides and amikacin appear to be reasonable.

Top