Resistance to ceftriaxone and penicillin G among contemporary syphilis strains confirmed by natural in vitro mutagenesis

Abstract

Background: For many years, syphilis treatment was considered straightforward due to the universal susceptibility of Treponema pallidum subsp. pallidum (TPA) to penicillin antibiotics.

Methods: Penicillin-binding protein genes from a ceftriaxone treatment failure T. pallidum isolate were assessed, and the introduction of identified mutations into two laboratory strains via natural competence was aimed for, followed by in vitro analysis of antibiotic susceptibility of the recombinants.

Results: TPA from the ceftriaxone treatment failure case contained A1873G and G2122A mutations in the TP0705 gene. Introduction of the A1873G mutation into laboratory strains DAL-1 and SS14 resulted in partial resistance to ceftriaxone and penicillin G in vitro. Furthermore, in silico analyses revealed that the majority of contemporary TPA SS14-like strains harbors this mutation and are thus partially resistant to ceftriaxone and penicillin G.

Conclusions: This finding indicates that TPA strains accumulate mutations that increase their resistance to β-lactam antibiotics. Alternative approaches for controlling syphilis will be needed, including the development of the syphilis vaccine.

Fig. 1. Scheme of the natural in vitro mutagenesis of the TPA SS14 strain.

The mixture of TP0705 PCR products, containing A1873G and G2122A mutations present in TPA from a ceftriaxone failure case, was added to an in vitro culture of the SS14 strain. From day 7, ceftriaxone (2.5 ng/ml of culture medium) was used for the selection of recombinants. At day 140, the A1873G mutation was fixed in the SS14 culture. At day 212, this mutation was not detected in the control treponemal cultures without the PCR mixture and ceftriaxone. The treponemal culture without the PCR mixture was eliminated in the ceftriaxone-supplemented control. The arrows indicate position 1873 of the TP0705 gene. As G2122A recombinants were not found during the experiment, this position is not shown. The scheme contains real sequences from the experiment at relevant time points after mutagenesis.

Fig. 2. Susceptibility of the recombinant SS14 strain to ceftriaxone and penicillin G determined under in vitro conditions.

a Susceptibility of the recombinant SS14 strain to ceftriaxone. b Susceptibility of the recombinant SS14 strain to penicillin G. The introduction of the A1873G mutation resulted in decreased susceptibility to both tested antibiotics. The graphs show the numbers of polA genes detected in the in vitro culture (at day 7, using qPCR). All detected polA numbers were normalized to the numbers present in the corresponding controls without antibiotics. The red line represents the median ratio value, the boxplot shows IQR, and the colored dots represent individual ratio values. At least three biological replicates and two technical replicates were examined using qPCR and are shown. Motility data were obtained using dark-field microscopy. The Mann–Whitney test was used to calculate the statistical significance (*p < 0.05, **p < 0.01, and ***p < 0.001).

Fig. 3. In silico analysis of TPA sequences containing the 1873G variant at the TP0705 locus responsible for partial resistance to the tested antibiotics.

Data were taken from the PubMLST database and included 1408 isolates. Allele 10 is considered the reference allele since it was found in both reference TPA strains- Nichols and SS14. Allele 1 is the most common among contemporary SS14-like strains. *Among the rare 1873G alleles, alleles 4 and 6 (n = 2 isolates) were omitted from the analysis due to the absence of SS14/Nichols clade classification.