Whole-genome sequencing of Chlamydia trachomatis isolates from persistently infected patients

Abstract

Background

Current understanding of the causes of treatment failure in Chlamydia trachomatis is poor and antimicrobial susceptibility data are lacking. We used genome sequencing to seek evidence of antimicrobial resistance in isolates sourced from patients who were persistently infected.

Methods

Genomic DNA was extracted from C. trachomatis isolates cultured in McCoy cell monolayers. Sequencing libraries were prepared using the SureSelect^XT Illumina paired-end protocol. Paired reads were mapped against a reference genome and single nucleotide variants (SNVs) were identified.

Results

Seven isolates from persistently infected patients and five isolates from successfully treated patients were sequenced. No previously reported SNVs associated with antimicrobial resistance were found. A unique SNV was identified in the gyrA gene of one treatment failure isolate but was located outside of the quinolone resistance determining region; this SNV has been previously reported in other members of the Chlamydiaceae family.

Conclusion

No genomic evidence was found to explain the differences in clinical outcome for our two groups of patients. A mutation unrelated to antimicrobial susceptibility was found in an isolate from a persistently infected patient. The cause of these persistent infections with C. trachomatis remains unclear.

Keywords

Chlamydia antimicrobial genome sequencing treatment failure

Introduction

In 2019, 229,411 diagnoses of Chlamydia trachomatis were reported to Public Health England (PHE).¹ Whilst infection with C. trachomatis is readily managed by antimicrobial therapy, anecdotal reports of treatment failure occur.^2–4 Understanding the causes of these treatment failures is poor and standardised antimicrobial susceptibility testing data are lacking. Where susceptibility data are available, there have been reports of multiple drug resistant strains recovered from treatment failures.⁵ Repeat infections after clinical diagnosis and treatment occur in up to 20% of C. trachomatis infections.⁶ Historically, genotyping of the ompA gene, which encodes a major outer membrane protein, has been used to differentiate treatment failure from re-infection.⁷ However, using ompA genovars to type C. trachomatis to track persistent or repeat infections is flawed. Recombination within this gene can mask the true evolutionary history of that strain⁸ and so longitudinal identification of a persistent infection becomes difficult without the use of whole-genome sequencing.

Antimicrobial susceptibility testing at PHE showed a significant difference in the doxycycline MICs obtained for C. trachomatis isolates sourced from patients with persistent infections to those who cleared infection with 1 g azithromycin (Table 1).⁹ The cause of this difference was unclear. Prior to 2018, the preferred first-line recommended therapy for treatment of uncomplicated C. trachomatis infection was a single oral dose of 1 g azithromycin.¹⁰ However, in response to growing concern about the effect of single-dose therapy on antimicrobial resistance in Mycoplasma genitalium, an update to the national guideline was issued by the British Association for Sexual Health and HIV (BASHH). The update preferentially recommended the alternate first-line therapy (100 mg doxycycline twice daily for 7 days) over an extended 2 g course of azithromycin (given over 3 days).

Table 1.

Isolate characteristics.

ID ()⁹	Site	ompA genotype	Treatment	Azithromycin MIC (mg/L)⁹	Doxycycline MIC (mg/L)⁹
Persistently infected
CC21 (Pt.6)	Ure	D	2 × 1 g azithromycin	≤0.125	0.125
CC26 (Pt.8)	Ure	E	2 × 1 g azithromycin	≤0.125	0.125
CC35 (Pt.2)	Ure	E	2 × 1 g azithromycin	≤0.125	>1
CC43 (Pt.3)	Cer	E	3 × 1 g azithromycin, 1 × 100 mg doxycycline 14/7 bd	0.125	0.125
CC79 (Pt.10)	Ure	E	2 × 1 g azithromycin, 1 × 100 mg doxycycline 7/7 bd	≤0.125	>1
CC81 (Pt.11)	Cer	E	1 × 1 g azithromycin, 1 × 500 mg erythromycin 7/7 qd	0.5	>1
CC83	Cer	E	1 × 1 g azithromycin, 1 × 100 mg doxycycline 7/7 bd, 2 × 1 g azithromycin + 500 mg azithromycin od 4/4	Not available	Not available
Successfully treated
Con18 (Ctrl.6)	Cer	D	1 × 1 g azithromycin	≤0.125	≤0.064
Con20 (Ctrl.7)	Cer	E	1 × 1 g azithromycin	≤0.125	1
Con27 (Ctrl.9)	Cer	E	1 × 1 g azithromycin	≤0.125	0.064
Con28 (Ctrl.10)	Cer	D	1 × 1 g azithromycin	≤0.125	0.064
Con31 (Ctrl.11)	Cer	D	1 × 1 g azithromycin	0.25	0.25

Note: Table detailing site of infection (Ure – urethral, Cer – cervical), ompA genotype, treatment regimen prescribed and minimum inhibitory concentration (MIC) of azithromycin and doxycycline (where available) for C. trachomatis isolates sourced from patients persistently infected and those successfully treated with 1 g azithromycin. MIC data are taken from Pitt et al. 2017, () in ID column indicate nomenclature in this reference. MIC data were unavailable for isolate CC83 as it was unrecoverable from archive.

To investigate the possibility of persistence of infection due to antibiotic resistance, we sequenced the genomes of C. trachomatis isolates referred to PHE from patients who were persistently infected despite treatment with recommended antimicrobials¹⁰ and were at low risk of re-infection. This was carried out using targeted enrichment of C. trachomatis DNA to allow sequencing of low level infections from patients exposed to multiple rounds of antimicrobial treatment, with greater sensitivity than other methods.¹¹

Methods

Isolates

Chlamydia trachomatis isolates from patients who were persistently infected, despite a minimum of two rounds of first-line recommended antimicrobials and a low risk of re-infection, were referred to PHE (between 2010 and 2014). Specific treatment regimens varied (Table 1), but all were in-line with the national treatment guidelines at the time.¹⁰ Risk of re-infection was assigned using self-declared sexual behaviour since initial diagnosis. Low risk of re-infection was assigned when the patient had no sexual contact, protected contact only or unprotected contact with a regular partner who had also tested positive and been treated or had not tested positive. Control isolates (collected in 2013) from patients who were successfully treated with a single dose of 1 g azithromycin were also sequenced.

Clinical and antimicrobial susceptibility data for some of these patients have been reported previously.^4,9

Tissue culture and DNA extraction

Isolates were cultured in McCoy cell monolayers in shell vials in Dulbecco’s Modified Eagle’s Medium (Gibco) supplemented with 10% gamma-irradiated foetal bovine serum (Gibco), 200 mm L-Glutamine (Sigma), 100 μg/mL Gentamicin (Gibco), 25 U mL-1 Nystatin (Sigma) and 100 μg/mL Vancomycin (Sigma). Cell lines were screened for the presence of contaminating Mycoplasma species using the MycoFluor^TM Mycoplasma Detection kit (Invitrogen). After infection, monolayers were incubated at 35°C, 5% CO₂ for 48 h before further passage. The MicroTrak Chlamydia trachomatis culture confirmation test (Trinity Biotech) was used to visualise inclusions prior to harvesting. Pooled shell vials were centrifuged at 2300 × g for 10 min, supernatant was aspirated, and the pellet re-suspended in 1 mL cold 1:10 PBS (plus glass beads). Samples were vortexed vigorously for 1 min and were then centrifuged at 284 × g for 5 min to remove cell debris. Supernatant containing C. trachomatis elementary bodies was aspirated and centrifuged at 13 684 × g for 5 min to pellet. Supernatant was aspirated and genomic DNA (gDNA) was extracted from the pellet using the Wizard Genomic DNA extraction kit (Promega).

Sequencing

gDNA was quantified (Qubit (Life technologies)) with carrier human genomic DNA (Promega) added to obtain a final concentration of 200 ng for shearing. Shearing was carried out using a Covaris E210 (6 × 60 s, duty cycle 10%, intensity 5 and 200 cycles per bust using frequency sweeping). The SureSelect^XT Illumina Paired-End Sequencing Library protocol was used for library preparation and sequencing was performed on a MiSeq (Illumina). RNA baits used to capture and enrich the C. trachomatis genome were designed previously.¹¹ SNV calling (point mutations, insertions (including acquired antimicrobial resistance genes, e.g. tet genes) and deletions) and phylogenetic tree reconstruction was carried out as previously described¹² except reads were mapped against the C. trachomatis D/UW-3/CX reference genome (GenBank accession number: NC_000117.1).

Ethics

Patients with persistent infections were referred to PHE as part of an enhanced surveillance programme; therefore, ethical approval was not required as PHE is able to handle these data under the Health Service (Control of Patient Information) regulation 2002, overseen by the Confidentiality Advisory Group. Control specimens were recruited through a sexual health clinic under ethics application reference 13/WM/0088.

Results

Sequencing

Seven isolates from persistently infected patients and five isolates from successfully treated patients were sequenced (EBI European Nucleotide Archive accession number: PRJEB45721). Phylogenetic reconstruction of the genomes showed that chromosomal and plasmid lineages generally separated on the ompA genotype (8 genotype E; 4 genotype D) (Figure 1). However, one genovar D isolate from a successfully treated patient (con28) was found in the genovar E lineage in both trees (Figure 1), likely due to a recombination event within the ompA sequence of that isolate. We found no evidence of previously reported point mutations,^13–21 acquired antimicrobial resistance genes or changes to efflux pumps and porin genes associated with antimicrobial resistance in any of our specimens. Nor did we find any major consensus variants across all treatment failure isolates only. A single, non-synonymous SNV was identified in gyrA in one treatment failure isolate (cc83) at position G1642T (amino acid change Alanine-548-Serine). This SNV was also confirmed by an independent PCR and sequencing of the product across that position. A BLASTP search against all non-redundant protein sequences on GenBank showed that this was a novel SNV not previously identified in C. trachomatis. Whilst this mutation was novel for C. trachomatis, it was found in other members of the Chlamydiaceae family, namely, Chlamydia suis and Chlamydophila felis.

Figure 1.

Maximum likelihood phylogenetic trees constructed from the 12 C. trachomatis chromosome (a) and plasmid (b) sequences in this study, with regions that have possibly undergone recombination removed. The tree is rooted against a chromosome sequence from LGV strain L2b/434/Bu (GenBank accession number: AM884176). ‘cc’ denotes an isolate from a patient who has failed treatment whilst ‘con’ denotes an isolate from a successfully treated patient.

Discussion

The aim of this study was to sequence the genomes of isolates from patients with persistent infections and patients who resolved infection after treatment with 1 g azithromycin to explore whether there was genetic evidence to explain the differences in clinical outcome for the patients. We did not identify any mutations, in isolates from either patient group, which had been reported previously as associated with in vitro antimicrobial resistance and/or treatment failure in C. trachomatis.^13–21 Further to this, no major consensus variants were identified across all treatment failure isolates that were not also identified in the isolates from the successfully treated patients. We did identify a non-synonymous SNV in gyrA, a gene known to be involved in fluoroquinolone resistance, of one treatment failure isolate. However, the locus where the mutation was identified was outside the quinolone resistance determining region of the gene so is unlikely to affect antimicrobial activity.

The DNA sequenced in this study was extracted from isolates which had undergone multiple passages in tissue culture. It is not known how subjecting the isolates to passaging affects the genome and whether culturing in the absence of antimicrobials selects out mutations that have a potential fitness cost to the organism in vitro. Direct sequencing from clinical specimens may resolve this. To complicate matters, heterotypic resistance has been described in C. trachomatis.^22,23 This is the result of adaptations by the bacteria to become less susceptible to antimicrobial therapy, for example, induction of slow-growing non-replicative forms in the presence of antimicrobials. These revert to replicative forms once the antibiotic pressure has been removed resulting in a relapse in infection. As stated previously, phenotypic antimicrobial susceptibility testing carried out for some of these isolates⁹ demonstrated reduced susceptibility of the isolates from persistently infected patients for doxycycline in comparison with the isolates from the successfully treated patients (Table 1). No genetic cause for this difference was detected in this study. Whilst cell lines were screened for the presence of contaminating organisms, generally as part of cell line maintenance and specifically for Mycoplasma species, due to the clinical origin of the specimens tested, it is not possible to completely rule out co-infection with another organism which may have influenced the susceptibility results obtained. However, it may be that the isolates from the persistently infected patients are exhibiting heterotypic resistance in vitro and this is the cause of the differing susceptibility results a theory also hypothesised recently by Shao et al.²⁴ This ‘resistance’ is not inherited and the genes involved in its induction are unknown at present.

To conclude, for our cohort, there was no evidence that repeat treatment and/or single-dose therapy for C. trachomatis infection resulted in antimicrobial resistance. Evidence of genomic recombination was found but was unrelated to susceptibility. If heterotypic resistance was the cause of the treatment failures, we would not necessarily expect to find genetic evidence of homotypic resistance. The causes of persistent C. trachomatis infection remain unclear.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Data repository

EBI European Nucleotide Archive (PRJEB45721).

ORCiD iDs

Rachel Pitt

Ronan Doyle

Paddy Horner

Neil Woodford

Michelle Cole

Judith Breuer

References

Mitchell

HAH

Sonubi

Kuyumdzhieva

, et al. Sexually transmitted infections and screening for chlamydia in England. London, UK: Public Health England, 2019.

Batteiger

Ofner

, et al. Repeated Chlamydia trachomatis genital infections in adolescent women. J Infect Dis 2010; 201: 42–51. DOI: 10.1086/648734.

Bhengraj

Vardhan

Srivastava

, et al. Decreased susceptibility to azithromycin and doxycycline in clinical isolates of Chlamydia trachomatis obtained from recurrently infected female patients in India. Chemotherapy 2010; 56: 371–377. DOI: 10.1159/000314998.

Pitt

Alexander

Horner

, et al. Presentation of clinically suspected persistent chlamydial infection: a case series. Int J STD AIDS 2013; 24: 469–475. DOI: 10.1177/0956462412472815.

Jones

Van Der Pol

Martin

, et al. Partial characterization of Chlamydia trachomatis isolates resistant to multiple antibiotics. J Infect Dis 1990; 162: 1309–1315.

Hosenfeld

Workowski

Berman

, et al. Repeat infection with Chlamydia and gonorrhea among females: a systematic review of the literature. Sex Transm Dis 2009; 36: 478–489. DOI: 10.1097/OLQ.0b013e3181a2a933.

Kapil

Press

Hwang

, et al. Investigating the epidemiology of repeat Chlamydia trachomatis detection after treatment by using C. trachomatis OmpA genotyping. J Clin Microbiol 2015; 53: 546–549. DOI: 10.1128/JCM.02483-14.

Harris

Clarke

Seth-Smith

, et al. Whole-genome analysis of diverse Chlamydia trachomatis strains identifies phylogenetic relationships masked by current clinical typing. Nat Genet 2012; 44: 413–419. DOI: 10.1038/ng.2214.

Pitt

Alexander

Ison

, et al. Phenotypic antimicrobial susceptibility testing of Chlamydia trachomatis isolates from patients with persistent or successfully treated infections. J Antimicrob Chemother 2017; 73: 680–686. DOI: 10.1093/jac/dkx454.

10.

Nwokolo

Dragovic

Patel

, et al. 2015 UK national guideline for the management of infection with Chlamydia trachomatis. Int J STD AIDS 2015; 27: 251–267. DOI: 10.1177/0956462415615443.

11.

Christiansen

Brown

Kundu

, et al. Whole-genome enrichment and sequencing of Chlamydia trachomatis directly from clinical samples. BMC Infect Dis 2014; 14: 591. DOI: 10.1186/s12879-014-0591-3.

12.

Doyle

Burgess

Williams

, et al. Direct whole-genome sequencing of sputum accurately identifies drug-resistant mycobacterium tuberculosis faster than MGIT culture sequencing. J Clin Microbiol 2018; 56.

13.

Sandoz

Eriksen

Jeffrey

, et al. Resistance to a novel antichlamydial compound is mediated through mutations in Chlamydia trachomatis secY. Antimicrob Agents Chemother 2012; 56: 4296–4302. DOI: 10.1128/AAC.00356.12

14.

Misyurina

Chipitsyna

Finashutina

, et al. Mutations in a 23S rRNA gene of Chlamydia trachomatis associated with resistance to macrolides. Antimicrob Agents Chemother 2004; 48: 1347–1349.

15.

Binet

Maurelli

. Frequency of development and associated physiological cost of azithromycin resistance in Chlamydia psittaci 6BC and C. trachomatis L2. Antimicrob Agents Chemother 2007; 51: 4267–4275. DOI: 10.1128/AAC.00962-07.

16.

Dessus-Babus

Bébéar

Charron

, et al. Sequencing of gyrase and topoisomerase IV quinolone-resistance-determining regions of Chlamydia trachomatis and characterization of quinolone-resistant mutants obtained In vitro. Antimicrob Agents Chemother 1998; 42: 2474–2481.

17.

Morrissey

Salman

Bakker

, et al. Serial passage of Chlamydia spp. In sub-inhibitory fluoroquinolone concentrations. J Antimicrob Chemother 2002; 49: 757–761.

18.

Rupp

Solbach

Gieffers

. Variation in the mutation frequency determining quinolone resistance in Chlamydia trachomatis serovars L2 and D. J Antimicrob Chemother 2008; 61: 91–94. DOI: 10.1093/jac/dkm447.

19.

Engström

Nguyen

Normark

, et al. Mutations in hemG mediate resistance to salicylidene acylhydrazides, demonstrating a novel link between protoporphyrinogen oxidase (HemG) and Chlamydia trachomatis infectivity. J Bacteriol 2013; 195: 4221–4230.

20.

O'Neill

Seth-Smith

HMB

Van Der Pol

, et al. Chlamydia trachomatis clinical isolates identified as tetracycline resistant do not exhibit resistance in vitro: whole-genome sequencing reveals a mutation in porB but no evidence for tetracycline resistance genes. Microbiology (Reading) 2013; 159: 748–756. DOI: 10.1128/jb.00506-1310.1099/mic.0.065391-0.

21.

Suchland

Bourillon

Denamur

, et al. Rifampin-resistant RNA polymerase mutants of Chlamydia trachomatis remain susceptible to the ansamycin rifalazil. Antimicrob Agents Chemother 2005; 49: 1120–1126. DOI: 49/3/112010.1128/AAC.49.3.1120-1126.

22.

Wang

Papp

Stamm

, et al. Evaluation of antimicrobial resistance and treatment failures for Chlamydia trachomatis: a meeting report. J Infect Dis 2005; 191: 917–923. DOI: 10.1086/428290.

23.

Horner

. The case for further treatment studies of uncomplicated genital Chlamydia trachomatis infection. Sex Transm Infect 2006; 82: 340–343. DOI: 10.1136/sti.2005.019158.

24.

Shao

You

Cao

, et al. High treatment failure rate is better explained by resistance gene detection than by minimum inhibitory concentration in patients with urogenital Chlamydia trachomatis infection. Int J Infect Dis 2020; 96: 121–127. DOI: 10.1016/j.ijid.2020.03.015.