Fluoroquinolone resistance in Neisseria meningitidis in Spain

Abstract

Objectives

To assess the ciprofloxacin resistance of Neisseria meningitidis isolated from 1999 through 2006 in Spain, susceptibility testing was conducted on 5300 isolates.

Methods

Ten isolates showed MICs of ciprofloxacin ranging between 0.06 and 0.25 mg/L, and they were characterized by analysis for mutations within the quinolone resistance determining regions (QRDRs) in the gyrA, gyrB, parC and parE genes. Mutations in the mtrR gene were also analysed.

Results and conclusions

Single mutations in gyrA appeared to be the main mechanism involved,Thr-91→Ile being the most frequent substitution seen. Two meningococci had four different gyrAsubstitutions. No mutations in the QRDRs of the parC and gyrB genes were detected, and three strains showed a His-495→Asn substitution in the parE gene. In addition, two different alterations in the mtrR gene affecting the expression of the MtrCDE efflux system were identified which may also contribute to the reduced susceptibility to quinolones seen in three strains.

ciprofloxacin, QRDRs, meningococci

Introduction

After widespread sulphonamide resistance in meningococcal strains, rifampicin has become the antimicrobial agent usually recommended for chemoprophylaxis. Fluoroquinolones are also used for this purpose particularly in adults, and their efficacy in eliminating nasopharyngeal carriage of Neisseria meningitidis has been frequently proved.¹ Only recently, the CLSI (formerly the NCCLS) established MIC interpretative standards of ciprofloxacin for meningococci,² so strains for which the MICs are ≤0.03, 0.06 and ≥0.12 mg/L are designated as susceptible, intermediate and resistant, respectively. The emergence of meningococcal strains with reduced susceptibility to ciprofloxacin has only been reported occasionally.³ This reduced susceptibility has been associated with point mutations in the quinolone resistance determining regions (QRDRs) of the target sites for the fluoroquinolones, and particularly with changes in the GyrA subunit of DNA gyrase. However, point mutations at that locus or in the other QRDRs (parC, gyrB and parE) were not detected in one meningococcal strain isolated in Argentina showing an MIC of ciprofloxacin of 0.12 mg/L. A deletion affecting the mtrR gene from the mtrRCDE gene complex, which encodes an energy-dependent efflux pump system, was suggested as responsible for the reduced susceptibility to ciprofloxacin in that case⁴ as has been already proposed for gonococci.⁵ The mtrCDE genes constitute a single transcriptional unit that is negatively regulated by the product of the adjacent but divergent mtrR gene, so that mutations in the mtrR gene or in its promoter can result in decreased susceptibility to several drugs.

Because the emergence and spread of fluoroquinolone resistance in other microorganisms has been associated with the increasing use of these drugs for the treatment of respiratory disease, the situation with meningococci colonizing the oropharynx might be similar. In fact, recently Shultz et al.⁶ have demonstrated that resistance to ciprofloxacin in N. meningitidis can be generated in vitro after exposure to the antibiotic.

In Spain, the study of resistance to ciprofloxacin was included in the routine antibiotic surveillance of the National Reference Laboratory for Meningococci in 2001 and one strain with reduced susceptibility to ciprofloxacin was soon described.³ In this study, the evolution of resistance to ciprofloxacin in meningococcal strains isolated in Spain from 1999 through 2006 is analysed, and the mechanisms for the decreased susceptibility, including both point mutations in the QRDRs but also changes in the mtrR gene, are characterized.

Materials and methods

Meningococcal strains

From 1999 through 2006, a total of 5300 isolates of N. meningitidis from patients but also asymptomatic carriers were received in the Reference Laboratory for Meningococci.

In all the strains, the serogroup and the serotype/serosubtype were determined as has been previously described.⁷

Antibiotic susceptibility testing

MICs of ciprofloxacin, rifampicin, penicillin, ceftriaxone and cefotaxime were determined in all the strains by the agar dilution method as has been previously described.²Streptococcus pneumoniae ATCC 49619 and Escherichia coli ATCC 25922 were used as quality control organisms.

MICs of ciprofloxacin ≥0.06 mg/L were confirmed using the Etest method (AB Biodisk, Solna, Sweden) according to the manufacturer's recommendations.

Sequencing of the QRDRs of gyrA, gyrB, parC and parE

The QRDRs of gyrA, gyrB, parC and parE were amplified by PCR from the DNA of those isolates with an MIC of ciprofloxacin ≥0.06 mg/L as described previously.^4,8 PCR products were purified (QIAquick PCR purification kit, QIAGEN) before sequencing. QRDRs from five ciprofloxacin-susceptible meningococci (MIC <0.06 mg/L) were also amplified and sequenced for comparison. The susceptible strains belonged to five different sequence types (STs).

The gyrA gene sequence was determined from nucleotides 160 to 438, corresponding to amino acids 54–146 of the GyrA protein. The gyrB gene sequence was determined from nucleotides 1310–1583 (amino acids 437–527 of the GyrB protein). The parC gene sequence was determined from nucleotides 166 to 420 (amino acids 56–140 of the ParC protein), and the parE gene sequence was determined from nucleotides 1117 to 1595 (amino acids 373–531 of the ParE protein).

Sequencing of the mtrR gene

Sequencing of the mtrR gene and the intergenic region between mtrC and mtrR was performed as previously described⁹ in all ciprofloxacin-resistant strains and also in five susceptible strains.

Multilocus sequence typing (MLST)

The nucleotide sequences of an ∼450 bp internal region from the abcZ, adk, aroE, fum, gdh, pdhC and pgm genes were amplified by PCR and sequenced, and the isolates were assigned to the corresponding ST through http://www.mlst.net.

Nucleotide sequence accession numbers

The sequences of QRDRs and the mtrR gene have been submitted to the GenBank database (http://www.ncbi.nlm.nih.gov/Genbank/index.html) and have been assigned the accession numbers: 17326 gyrA, EF117893; 17431 gyrA, EF093492; 17326 parE, EF117894; 13067 mtrR, EF117895; 17326 mtrR, EF117896.

Results

The analysis revealed 10 strains with ciprofloxacin MICs ≥0.06 mg/L (Table 2), ranging from 0.06 to 0.25 mg/L. All these isolates were susceptible to ceftriaxone and cefoxitin, and four exhibited decreased susceptibility to penicillin (Table 1). Additionally, two of these strains were resistant to rifampicin (MIC of 2 mg/L) and one was defined as intermediately resistant (MIC of 1 mg/L).

Table 1

MICs (mg/L) of ciprofloxacin and four additional antimicrobial drugs for meningococcal strains with reduced susceptibility to ciprofloxacin isolated in Spain

INa	Ciprofloxacin ADb	Ciprofloxacin Etestc	Rifampicin	Penicillin	Cefotaxime	Ceftriaxone
13067	0.25	0.19	2	0.03	0.0007	0.00015
13646	0.12	0.25	2	0.06	0.0015	0.0003
14959d	0.12	0.19	0.06	0.06	0.003	0.0007
15557	0.06	0.064	0.25	0.06	0.007	0.0015
16170	0.25	0.25	0.25	0.25	0.03	0.003
16172	0.25	0.19	0.06	0.25	0.03	0.003
17326	0.25	0.19	1	0.12	0.03	0.015
17346	0.12	0.125	0.06	0.015	0.0003	0.0003
17431	0.12	0.094	0.06	0.25	0.015	0.0007
17840	0.25	0.125	0.125	0.032	0.007	0.0007

^aIdentification number.

^bMICs values for ciprofloxacin using the agar dilution (AD) method.

^cMICs values for ciprofloxacin using the Etest method.

^dStrain already published.³

Eight different antigenic combinations were found (Table 2), with two pairs of strains appearing with the same combination (B:NT:P1.15 and W135:2b:P1.3) (Table 2). Those two B:NT:P1.15 strains appear epidemiologically related because the first one (16170) was isolated from the nasopharynx of a woman who was the mother of the 2-year-old patient from whom strain 16172 was isolated from blood. The other W135:2b:P1.3 meningococci were isolated in the same hospital from two different elderly male patients with respiratory disease.

Table 2

Characteristics of the N. meningitidis strains with reduced susceptibility to ciprofloxacin isolated in Spain (1999–2006)

INa	Year of isolation	Origin	Gender/age	Clinical diagnosis	Antigenic characterizationb	GyrA	GyrB	ParC	ParE	mtrR	STc	Clonal complex
13067	1999	spittle	M/98 years	—	W135:2b:P1.3	Thr-91→Ile	—	—	—	premature stop codon	22	ST22
13646	1999	spittle	M/80 years	—	W135:2b:P1.3	Thr-91→Ile	—	—	—	premature stop codon	22	ST22
14959d	2002	CSF	M/2 years	meningitis	B:NT:P1.5.10	Thr-91→Ile	—	—	—	—	5783	—
15557	2003	CSF	F/24 years	meningitis	C:2a:P1.5	Asp-95→Asn	—	—	—	—	11	ST11/ET37
16170	2003	nasopharynx	F/36 years	asymptomatic carrier	B:NT:P1.15	Thr-91→Ile	—	—	His-495→Asn	—	5784	—
16172	2003	blood	M/2 years	sepsis	B:NT:P1.15	Thr-91→Ile	—	—	His-495→Asn	—	5784	—
17326	2005	spittle	M/7 years	—	NG:4:NST	Asp-95→Asn Asn-103→Asp Ile-111→Val Val-120→Ile	—	—	His-495→Asn	deletion of 154 bp	1624	ST167
17346	2005	nasopharynx	M/17 years	meningitis + sepsis	Y:NT:NST	Thr-91→Ile	—	—	—	—	5785	—
17431	2005	CSF	M/11 months	—	B:1:NST	Thr-91→Ile Asn-103→Asp Ala-105→Ser Ala-136→Glu	—	—	—	—	1946	461
17840	2006	blood	—	—	Y:NT:P1.9	Thr-91→Ile	—	—	—	—	4789	ST5/subgroup III

^aIdentification number.

^bNT, non-typeable; NG, non-groupable; NST, non-subtypeable.

^cSequence type.

^dStrain already published.³

According to the MLST data, the 10 ciprofloxacin-resistant strains belong to several unrelated clones (Table 2).

Comparison of the GyrA, GyrB, ParC and ParE protein sequences in the 10 ciprofloxacin-resistant strains with those found in susceptible strains, showed that all 10 resistant strains contained mutations in the GyrA QRDR (Table 2).

The most common GyrA alteration (8/10 strains) was Thr-91→Ile, and two isolates showed four mutations at different positions in GyrA. No alterations in GyrB or in ParC were detected. A His-495→Asn change in ParE was found in three strains.

The mtrR gene showed no mutations in seven strains, while two different changes appeared in three isolates (Table 2): two strains (13067 and 13646) showed a premature stop codon at codon 183, and the other one (17326) showed the same deletion of 154 bp that was previously found in a strain isolated in Argentina.⁴

None of the ciprofloxacin-susceptible isolates had detectable mutations in the analysed genes.

Discussion

The emergence of meningococcal strains with reduced susceptibility to ciprofloxacin has been rarely reported³ and it has been associated with point mutations in the gyrA gene. In addition, changes affecting the promoter and coding region of the mtrR gene that acts as repressor of the mtrCDE gene complex have also been suggested to be responsible for the reduced susceptibility to ciprofloxacin in both meningococci⁴ and gonococci.⁵

The emergence of meningococcal strains showing decreased susceptibility to ciprofloxacin might have implications in chemoprophylaxis because ciprofloxacin is widely recommended in adults to eradicate meningococci from nasopharyngeal carriers.

In this study, the mechanisms for ciprofloxacin resistance in meningococcal strains appear to mainly involve mutations in the GyrA protein (Table 2). Changes did not appear in either GyrB or ParC, and three strains showed a particular change in ParE (His-495→Asn). This mutation, however, did not seem to contribute to increased MICs (Table 1), suggesting that it does not play any role in decreased susceptibility to quinolones. In fact, the exact role of point mutations in ParE has not been well established,^8,10,11 and this role need to be confirmed by transformation experiments in future studies. The most common substitution seen in GyrA was Thr-91→Ile (Table 2), which has already been described for meningococci both in vivo³ and in vitro.⁶ Eight strains presented with this change, it being the only one seen in six strains. One strain showed an Asp-95→Asn change in GyrA. This specific mutation has been previously described for meningococci⁴ and it has been described in N. gonorrhoeae.¹¹ It is possible that both mutations Thr-91→Ile and Asp-95→Asn in GyrA produce similar levels of resistance to ciprofloxacin in meningococci. The 91 position might represent a hot spot mutation in the gyrA gene because of its high frequency in so many different isolates.

Recently Shultz et al.,⁶ after the selection of mutant ciprofloxacin-resistant strains, found a first step with a single GyrA mutation, followed by double GyrA changes, and finally a single ParC substitution. By contrast, in this study, two strains showed four different GyrA mutations without ParC changes (Table 2). Both strains showed slightly different MICs of ciprofloxacin and they were not different from the MICs for strains with a single gyrA mutation. Whether additional GyrA mutations play a particular role in the development of reduced susceptibility to ciprofloxacin is the objective of current research.

Regarding the analysis of the mtrR gene, one strain (Table 2) showed the same deletion that also appeared in a meningococcal strain with resistance to ciprofloxacin previously isolated in Argentina.⁴ The contribution of mtrR changes to the increase in the resistance to antimicrobial agents has already been proposed for N. gonorrhoeae.¹² However, it has been suggested that the MtrCDE efflux system in meningococci may not be subject to the MtrR regulatory scheme.¹³ Further studies are necessary to elucidate the contribution of the deletion found in this study to the resistance to ciprofloxacin as well as other antimicrobial agents. Two other strains (Table 2) showed a mutation truncating MtrR. Once again, the role of this finding in ciprofloxacin resistance is not clear and the MIC of ciprofloxacin was not much higher than in other strains. Decreased ciprofloxacin susceptibility was seen in different and unrelated clones (Table 2), but was rare, suggesting that strains acquiring this type of resistance are not spreading perhaps because of a biological cost as has been suggested for resistance to rifampicin.³

Although the emergence of meningococcal strains showing decreased susceptibility to ciprofloxacin seems to be slowly evolving, continuous surveillance is necessary to monitor its emergence and spread and to guide proper public health interventions in preventing drug-resistant meningococci.

Funding

R. E. and R. A. were recipients of pre-doctoral fellowships from the Institute of Health Carlos III (ISCIII 04/0021 and ISCIII03/0037, respectively). This work was partially supported by a grant FIS PI060297 and by Ministerio de Sanidad y Consumo, Instituto de Salud Carlos III, Spanish Network for Research in Infectious Diseases (REIPI RD06/0008).

Transparency declarations

None to declare.