Abstract
Doxycycline post-exposure prophylaxis (doxy-PEP) for sexually transmitted bacterial infections reduces the risk of syphilis and chlamydia, but effectiveness against gonorrhea is variable, likely attributable to varying resistance rates. As doxy-PEP is incorporated into clinical practice, an urgent unanswered question is whether increased doxycycline use will drive tetracycline-class resistance in Neisseria gonorrhoeae. Here, we report an updated RT-PCR molecular diagnostic to detect the tetM gene that confers high-level tetracycline resistance in N. gonorrhoeae.
Text
Doxycycline post-exposure prophylaxis (doxy-PEP) for sexually transmitted bacterial infections reduces the risk of syphilis and chlamydia, while effectiveness against gonorrhea is variable (1-3), likely attributable to varying resistance rates. High-level resistance to tetracycline in Neisseria gonorrhoeae, the agent of gonorrhea, is estimated at 10% prevalence in the United States (4) and up to 70-100% in some populations (5-7). As doxy-PEP is incorporated into clinical practice, an urgent unanswered question is whether increased doxycycline use will drive tetracycline-class resistance in N. gonorrhoeae. However, current culture-based surveillance of N. gonorrhoeae antimicrobial susceptibility may not be sufficient to quickly detect changes in the prevalence of tetracycline resistance because surveillance programs collect and test a modest number of isolates (e.g. (8)). By contrast, molecular diagnostics for tetracycline resistance would enable the use of remnants from nucleic acid amplification tests (NAATs), the most widely used diagnostic, and can confidently detect changes in population-level resistance more quickly. An accurate molecular diagnostic for high-level tetracycline resistance thus has public health value (9).
High-level tetracycline resistance in N. gonorrhoeae is conferred by TetM, a plasmid-encoded ribosomal protection protein. Although primers for PCR-based detection of the tetM gene have been reported (10, 11), the existing reverse primer, originally designed from the Ureaplasma urealyticum tetM sequence, has poor predicted sensitivity for N. gonorrhoeae tetM sequences: only 992 of 2223 N. gonorrhoeae tetM sequences deposited in PubMLST (Supplemental File S1) (12) contain an exact match for this primer. We designed a new reverse primer (Supplemental Materials, Table S1) that recognizes all 2223 N. gonorrhoeae tetM sequences tested, and that, when used in combination with the reported forward primer from Ison et al., is predicted to successfully amplify 2221 of 2223 N. gonorrhoeae tetM sequences (or all 2223 sequences if tolerating a single nucleotide mismatch).
To test this primer set, we assembled a panel of 13 diverse clinical isolates comprising 6 tetM+ and 7 tetM-strains (Supplemental Materials, Figure S1, Table S2). Genomic DNA was extracted, and tetM primers were tested for specific and sensitive amplification against this panel (Supplemental Materials, Methods), using porA as a reference amplicon (13). As expected, tetM amplification segregates clearly with the presence or absence of tetM in all strains (Figure 1A). Accurate discrimination was maintained over a range of template concentrations (Figure 1B) to a limit of approximately 28 copies of the gonococcal genome per reaction, at which point the reaction approaches its sensitivity limit (Supplemental Materials, Figure S2).
RT-PCR-based detection of tetM in a diverse strain panel. Mean cycle threshold (Ct) of tetM amplification, baseline corrected by porA amplification, for (A) 2.5 ng genomic DNA purified from strains with tetM (n=6, left) and without tetM (n=7, right); and (B) over a variety of template concentrations for one representative strain each with tetM and without tetM. Amplification was monitored via SYBRGreen fluorescence on a QuantStudio™ 7 Flex Real-Time PCR system (see Supplemental Methods). Shown: mean ΔCt from three technical replicates per reaction; data representative of at least two independent experiments.
The sensitivity and specificity of this test for clinical samples, and importantly for GC-positive NAAT remnants, requires further characterization. Because tetM in N. gonorrhoeae originated from a streptococcal transposon (14), highly similar tetM sequences are found in other organisms, creating a risk of false positive tetM detection. Comparing tetM positivity rates in GC-positive and GC-negative NAAT specimens will be an important step in characterizing the expected false positive rate. The use of porA primers for normalization may also mitigate this issue. Although the data here were generated using dye-based (SYBR Green) RT-PCR detection, probe-based detection may also improve specificity.
Data Availability
All data produced in the present work are contained in the manuscript
Ethics statement
Clinical isolates were obtained from the CDC and FDA Antimicrobial Resistance Isolate Bank, NCIP panel (15) and from previously published studies (16, 17). The original collection of these isolates was deemed not human subjects research by the Center for Disease Control and Prevention’s Office of the Associate Director for Science for the Gonococcal Isolate Surveillance Program (GISP/eGISP) (16, 18) or by the institutional review boards at the University of California San Diego, University of Washington, and Harvard T.H. Chan School of Public Health (17).
Funding
This study was supported by R01AI132606 (to Y. H. G.), R01AI153521 (to Y. H. G.), contract 200-2016-91779 with the Centers for Disease Control and Prevention, and by the Pathogen Genomics Centers of Excellence Network. Wei Shen is supported by grants from the National Natural Science Foundation of China (32000474, 82341112), Chinese Scholarship Council scholarship (202308500105), and EMBL Visitor/Sabbatical Programme fellowship.
References
