Streptococcus pyogenes Is Associated with Idiopathic Cutaneous Ulcers in Children on a Yaws-Endemic Island

Abstract

Exudative cutaneous ulcers (CU) in yaws-endemic areas are associated with Treponema pallidum subsp. pertenue (TP) and Haemophilus ducreyi (HD), but one-third of CU cases are idiopathic (IU). Using mass drug administration (MDA) of azithromycin, a yaws eradication campaign on Lihir Island in Papua New Guinea reduced but failed to eradicate yaws; IU rates remained constant throughout the campaign. To identify potential etiologies of IU, we obtained swabs of CU lesions (n = 279) and of the skin of asymptomatic controls (AC; n = 233) from the Lihir Island cohort and characterized their microbiomes using a metagenomics approach. CU bacterial communities were less diverse than those of the AC. Using real-time multiplex PCR with pathogen-specific primers, we separated CU specimens into HD-positive (HD+), TP+, HD+TP+, and IU groups. Each CU subgroup formed a distinct bacterial community, defined by the species detected and/or the relative abundances of species within each group. Streptococcus pyogenes was the most abundant organism in IU (22.65%) and was enriched in IU compared to other ulcer groups. Follow-up samples (n = 31) were obtained from nonhealed ulcers; the average relative abundance of S. pyogenes was 30.11% in not improved ulcers and 0.88% in improved ulcers, suggesting that S. pyogenes in the not improved ulcers may be azithromycin resistant. Catonella morbi was enriched in IU that lacked S. pyogenes. As some S. pyogenes and TP strains are macrolide resistant, penicillin may be the drug of choice for CU azithromycin treatment failures. Our study will aid in the design of diagnostic tests and selective therapies for CU. IMPORTANCE Cutaneous ulcers (CU) affect approximately 100,000 children in the tropics each year. While two-thirds of CU are caused by Treponema pallidum subspecies pertenue and Haemophilus ducreyi, the cause(s) of the remaining one-third is unknown. Given the failure of mass drug administration of azithromycin to eradicate CU, the World Health Organization recently proposed an integrated disease management strategy to control CU. Success of this strategy requires determining the unknown cause(s) of CU. By using 16S rRNA gene sequencing of swabs obtained from CU and the skin of asymptomatic children, we identified another possible cause of skin ulcers, Streptococcus pyogenes. Although S. pyogenes is known to cause impetigo and cellulitis, this is the first report implicating the organism as a causal agent of CU. Inclusion of S. pyogenes into the integrated disease management plan will improve diagnostic testing and treatment of this painful and debilitating disease of children and strengthen elimination efforts.

Yaws in Pygmy and Bantu children inhabiting the rural zones of Central Africa

“…pertenue to penicillin and azithromycin resistance rates are low) [ 16 ]. This may be due to the fact that some yaws-like lesions may have a more complex aetiology, and resistance to treatment can be explained by the presence of other etiological factors, including Haemophilus ducreyi [ 2 ], Streptococcus pyogenes [ 17 ] as well as pathogens responsible for cutaneous leishmaniosis [ 18 ], leprosy [ 18 ] or Buruli ulcer [ 18 , 19 ].…”

Two Streptococcus pyogenes emm types and several anaerobic bacterial species are associated with idiopathic cutaneous ulcers in children after community-based mass treatment with azithromycin

“…However, 16S rRNA gene sequencing can skew bacterial abundance, can only detect bacteria that retain sufficient homology to the degenerate PCR primers employed, and cannot detect other types of microorganisms [12]. As well, S. pyogenes was not detected in over half of the IU specimens [5]. Thus, although both studies suggested that bacterial pathogens were associated with IU, they were limited by their inability to detect a broad range of microorganisms and failed to identify IU associated microorganisms in many IU specimens.…”

“…Clinical management of CU and future eradication efforts would be greatly improved by identifying all causes of CU in endemic communities. Here, we applied shotgun metagenomic sequencing to detect a broader range of microorganisms in the same CU samples prospectively collected 36, 42, and 48 months post-MDA from the Lihir Island cohort that we had analyzed by 16S rRNA gene sequencing previously [5]. Our goals were to confirm that S. pyogenes is associated with IU and to identify other possible bacterial and non-bacterial agents of IU.…”

“…More recently, Haemophilus ducreyi (HD), the causative agent of chancroid, has been also identified as an etiological agent of CU [ 2 , 3 ]. However, no TP or HD DNA can be detected by PCR in ~30% of CU cases, and the etiology of these idiopathic ulcers (IU) has yet to be entirely defined [ 4 , 5 ].…”

“…The second study analyzed CU specimens prospectively collected specifically for PCR testing and microbiome analysis 36, 42, and 48 months post-MDA. Using 16S rRNA gene sequencing, a single amplicon sequence variant (ASV) of Streptococcus pyogenes was enriched in IU and in non-healing ulcers compared to TP, HD, and TP/HD ulcers and ulcers which showed signs of healing, respectively [ 5 ]. However, 16S rRNA gene sequencing can skew bacterial abundance, can only detect bacteria that retain sufficient homology to the degenerate PCR primers employed, and cannot detect other types of microorganisms [ 12 ].…”