A Long-Term Study of the Diversity of OXA-48-Like Carbapenemase-Producing Bacterial Strains in Infected Patients and Carriers

Abstract

Purpose:

Since their emergence at the beginning of the century, OXA-48 carbapenemases have spread in the community and in hospitals. To assess the diversity of OXA-48-producing bacterial strains and plasmids in the hospital setting, we studied the strains isolated from patients in three hospitals in the Paris area.

Materials and Methods:

All possible OXA-48-like strains were included in the study. OXA-48-like and extended-spectrum beta-lactamase-encoding genes were identified, and fingerprinting analysis was performed for all Escherichia coli and Klebsiella pneumoniae strains. The backbones and close genetic environments of bla_OXA-48 were assessed by amplifying genes that were specific to the pOXA-48a plasmid and PCR, encompassing the junctions between bla_OXA-48 and its direct genetic environment.

Results:

Overall, 68 strains from 30 patients were studied. These strains belonged to seven different enterobacterial species. OXA-48, OXA-204, and OXA-401 were identified in 62, 3, and 3 isolates, respectively. Additional broad-spectrum beta-lactamases were identified in 34% (23/68) of the strains. The strain diversity was high between and within patients. Identical patterns were observed only within individual patients or among epidemiologically related patients. Plasmid mapping was performed in the 62 OXA-48-producing strains and the 3 OXA-405-producing strains, resulting in the identification of 5 different patterns.

Conclusion:

Because of their ability to transfer between strains, OXA-48 carbapenemases have a high risk of dissemination and may become endemic in France.

Introduction

Recently, carbapenemase-producing strains have emerged and rapidly disseminated across the world, exhibiting some regional specificities.¹ Since its first description in 2004 in Turkey,² the OXA-48 enzyme has been found in almost all of the surrounding countries as well as in Europe³ and the United States.^4,5 The dissemination of OXA-48-producing strains has become endemic in many areas, particularly in the Mediterranean region.^6,7 In addition, other alleles closely related to OXA-48, described as OXA-48-like enzymes, have emerged simultaneously: OXA-181, OXA-204, and OXA-232 are the most common enzymes of this type.⁸

To date, OXA-48 has been described almost exclusively in plasmids closely related to pOXA-48a, an ∼62 kb plasmid belonging to the L/M incompatibility group.⁹ These plasmids are characterized by the insertion of the Tn1999 transposon inside the tir gene. Interestingly, the tir gene, which encodes a transfer inhibition protein, is rendered nonfunctional by this insertion. As a consequence, plasmid transfer may be enhanced, a feature that could explain the rapid spread of bla_OXA-48 across enterobacterial species.¹⁰ In addition to plasmid dissemination, the spread of some clones has also contributed significantly to the OXA-48 pandemic, as previously described.^3,11

There have been few long-term assessments of the diversity of OXA-48-like strains and plasmids within and between patients. To better understand the reasons for the successful spread of OXA-48, it may be useful to investigate the ability of plasmids to transfer among enterobacterial strains within or across species with specific genetic features. Similarly, comparing the genetic backbones of strains from epidemiologically unrelated patients may also provide information about the role of clonal dissemination in the OXA-48 pandemic outside of an epidemic situation. Here, we report on the diversity of OXA-48-like plasmids and strains isolated from clinical samples and/or the feces of inpatients at admission and during their stay in three French hospitals in the Paris area from 2011 to 2014.

Materials and Methods

Strain selection

In this study, data on the strains with an OXA-48-like resistance mechanism from 2011 to 2014 were extracted from the computer systems of three French hospitals. Strains were suspected to produce an OXA-48-like enzyme when the susceptibility to ertapenem was reduced and/or high-level resistance to temocillin was observed. This was observed whether from strains isolated from clinical specimens or from strains isolated from fecal samples performed in the framework of screening multidrug resistant bacteria (stool culture or rectal swabs). Confirmation was obtained whether by the use of local PCR or sending to the French National Reference Centre for antibiotic resistance. The strains were not de-duplicated to allow an evaluation of the diversity between and within carriers. For the same reason, epidemiologically related strains, that is, those isolated among the contacts of a common index case, were also included. In all cases, the geographical origin of the carriers was reported. Patients repatriated or coming to France for medical care were systematically screened for multidrug resistant bacteria carriage, as recommended by the French High Public Health Committee (HCSP, France). In that case, when the screening was positive for OXA-48, it was assumed that the OXA-48 strain had been acquired abroad. In all the other cases (including French residents of foreign backgrounds), the OXA-48 strain was considered as acquired in France. Identification of the strains was performed with the VITEK2 system (bio-Meraux, Marcy l'Etoile, France) until March 2014 and with the LT-2 Maldi-Tof system (Andromas, Paris, France) thereafter. All strains were analyzed for antibiotic susceptibility by using the agar diffusion assay, and extended-spectrum beta-lactamase (ESBL) production was assessed by the double-disk synergy test according to the recommendations of the French Microbiology Society (www.sfm-microbiologie.org). In each confirmed carrier, all different isolates of OXA-48-like producing strains were isolated and stored at −80°C after routine analysis in a brain-heart infusion medium with 10% glycerol, in all the three institutions. Isolates were considered different according to their macroscopic aspect or when different resistance phenotypes were observed on the antibiogram. In that last case, colonies were reisolated and studied until obtaining pure resistance phenotypes and before freezing. At the time of our study, all the available strains were thawed at the same time for subsequent analysis.

DNA extraction, molecular characterization of ESBL and OXA-48-like genes, plasmid backbone characterization, and fingerprinting analysis

Total DNA extraction was performed by using an InstaGene Matrix kit (BioRad, Hercules, CA). OXA-48-like and ESBL-encoding genes were identified by using specific primers for the bla_OXA-48, bla_CTX-M, bla_TEM, and bla_SHV genes as previously described.¹² All PCR products were further sequenced on both strands and screened against the GenBank database to precisely characterize the detected alleles. The repA, traU, and parA genes, which encode proteins involved in replication, transfer, and partitioning, and the relL/M gene, which encodes the relaxase of the pOXA-48a plasmid, were amplified by PCR to characterize the backbone of the OXA-48 plasmids.⁹ PCR encompassing the junctions between bla_OXA-48 and tir and between tir and mucAB was performed to assess the close genetic environment of bla_OXA-48 (Table 1). Finally, fingerprinting analyses were performed for all Escherichia coli and Klebsiella pneumoniae strains by using repetitive extragenic palindromic PCR (rep-PCR) and enterobacterial repetitive intergenic consensus sequence PCR (ERIC-PCR) with the rep-1R, rep-2T, and ERIC-2 primers as previously described.¹³ Pattern profiles were visually judged and considered different if they showed a single band of diverse size, regardless of the intensity of the bands, according to the ESGEM guidelines.¹⁴

Table 1.

Primers Designed in This Study

Region/gene amplified	Primers	Sequence (5′–3′)	Size (bp)
bla_OXA-48-Δtir	OXA-48-Δtir F	GGGCGATCAAGCTATTGGGA	1,629
	OXA-48-Δtir R	GGCAAACAGCAAAGACGTTA
Δtir-mucAB	Δtir-mucAB F	GCATTTTCAGGCCAATACCA	1,588
	Δtir-mucAB R	GAACGTAACCACACCCCATA
rel L/M	Rel L/M F	TACATGGCAGACAAGCCGTC	1,716
	Rel L/M R	AAACTCGCCGTCACCATCAA

Resistance transfer assays

Representative strains of each clone lacking at least one molecular marker of the pandemic pOXA-48a plasmid were selected and submitted to resistance transfer assays. Conjugations were carried out in brain-heart broth, and E. coli J53^rif was used as the recipient strain. Mating broths were incubated for 6 and 18 hours, and transconjugants were selected on Drigalski agar plates supplemented with rifampin (300 μg/ml) and temocillin (100 μg/ml).

Results

Strains and patients

In all, 68 isolates from 45 samples corresponding to 30 patients were included in this study (Table 2). Among the patients, 13 were originally from France, 10 were from North Africa (Morocco, n = 4; Algeria, n = 3 and Egypt, n = 3), 3 were from the Eastern Mediterranean (Lebanon, n = 1; Turkey, n = 1 and Koweit, n = 1), and 3 were from other areas (French West Indies, n = 1; Poland, n = 1 and Portugal, n = 1). The geographical origin of one patient was not available. In almost all of the cases, the country of acquisition was identical to the country of origin of the patient. However, three patients originating from abroad (one from Morocco, one from Turkey, and one from Portugal) acquired their strains in France during their hospitalization. Conversely, one patient originating from France was repatriated from a Senegalese hospital, where he had acquired an OXA-48 strain. Most of the samples were rectal swab samples (n = 25), stool cultures (n = 5), or urine samples (n = 5). The other samples included tracheal or bronchial aspirations, a cervical sample, a breast sample, and a sputum sample (n = 3, 1, 2 and 1, respectively). The nature of the remaining three samples was not available (Table 2). Twenty-nine E. coli, 25 K. pneumoniae, 5 Citrobacter freundii, and 4 Enterobacter cloacae strains were isolated from 18, 14, 4, and 3 patients, respectively (Table 3A, B and C). Three Serratia marcescens, one Citrobacter koseri, and one Klebsiella oxytoca were isolated from one patient each. Twenty-two patients were carrying only 1 OXA-48-like-producing enterobacterial species: E. coli in 11 cases, K. pneumoniae in 7 cases, E. cloacae in 2 cases, and C. freundii and C. koseri in one case each. In four cases, both E. coli and K. pneumoniae were isolated from the same patient. In the four last cases, E. coli and/or K. pneumoniae were found in association with another enterobacterial species. Interestingly, as many as 18 strains were isolated from Patient No. 7, not only because he was a long-term carrier but also because the diversity of the species and/or resistance phenotypes were particularly high. A few carriers were part of a contact screening list, which included Patient Nos. 1, 3, 6, and 9 in one group, No. 4 and No. 11 in a second group, and No. 24 and No. 25 in a last group.

Table 2.

Characteristics of the Included Patients and Nature of the Studied Samples

No. of the patient ^a	Patient's country of origin	Country of acquisition	Type of sample
1 (a)	France	France	2 RS
2	Algeria	Algeria	1 RS
3 (a)	France	France	1 RS
4 (b)	France	France	1 RS
5	France	France	1 CW
6 (a)	Algeria	Algeria	1 TA
7	Morocco	France	5 RS, 2 BR, 1 BA
8	Algeria	Algeria	1 RS
9 (a)	France	France	1 TA
10	Lebanon	Lebanon	1 U, 2 RS
11 (b)	France	France	1 SC, 1 U
12	Turkey	France	1 U
13	Morocco	Morocco	2 RS, 2 SC
14	Egypt	Egypt	1 RS
15	Morocco	Morocco	1 SC
16	French West Indies	French West Indies	1 SC
17	Kuweit	Kuweit	1 RS
18	Portugal	France	2 RS
19	France	France	1 RS
20	Egypt	Egypt	1 na
21	na	na	1 na
22	France	France	1 SP
23	France	France	1 na
24 (c)	Egypt	Egypt	1 RS
25 (c)	France	France	1 RS
26	France	Senegal	1 RS
27	Poland	Poland	1 RS
28	Morocco	Morocco	1 U
29	France	France	1 U
30	France	France	1 RS

Epidemiologically related patients are flagged with a lowercase letter in brackets.

BA, bronchial aspiration; BR, breast sample; CW, cervical wound; na, not available; RS, rectal swab; SC, stool culture; SP, sputum; TA, tracheal aspiration; U, urine sample.

Table 3.

Characteristics of the Studied Escherichia coli (A), Klebsiella pneumoniae (B), and Other Enterobacterial (C) Isolates

(A)
			Co-resistance		Mapping of the OXA-48 plasmid ^c
No. of the patient ^a	E. coli clonal pattern (origin of the sample)^b	Type of OXA-48-like carbapenemase	ESBL/cephalosporinase	Other	rep	traU	parA	bla_OXA-48-ΔTir	Δtir-mucAB	rel L/M
2	a (RS)	48	0	K sxt tet	+	+	+	+	+	+
3 (a)	b (RS)	48	CTX-M-15	fos	+	+	+	+	+	+
4 (b)	c (RS)	48	0	0	+	+	+	+	+	+
7	d (RS)	48	0	0	+	+	+	+	+	+
	e (RS)	48	0	0	+	+	+	+	+	+
	f (RS)	48	0	0	+	+	+	+	+	+
8	g (RS)	48	CTX-M-15	K fq sxt tet	+	+	+	+	+	+
10	A (U)	48	CTX-M-15	fq sxt tet	0	0	+	+	+	0
	h (RS)	48	CTX-M-15	KTA fq sxt tet	+	+	+	+	+	+
	A (RS)	48	CTX-M-15	fq sxt	0	0	+	+	+	0
12	i (U)	48	0	fq tet	+	+	+	+	+	+
13	B (RS)	48	0	0	+	+	+	+	+	+
	B (SC)	48	0	0	+	+	+	+	+	+
	B (SC)	48	0	sxt	+	+	+	+	+	+
14	j (RS)	48	CTX-M-14	TG fq sxt	0	0	+	+	+	0
15	k (SC)	48	0	K sxt	+	+	+	+	+	+
15	l (SC)	48	0	sxt tet	+	+	+	+	+	+
17	m (RS)	48	CTX-M-14	TG fq sxt	0	0	0	+	+	0
18	n (RS)	204	SHV-27	KTG fq tet	0	0	0	0	0	0
	C (RS)	204	CMY-4	KTG tet	0	0	0	0	0	0
	C (RS)	204	CMY-4	KTG fq sxt tet	0	0	0	0	0	0
19	o (RS)	48	CTX-M-14	K fq sxt tet	+	+	+	+	+	+
20	p (RS)	48	CTX-M-9	KTG na sxt	0	0	+	+	+	0
21	q (RS)	48	0	KTG fq sxt tet	0	0	0	0	0	0
22	D (SP)	48	0	tet	+	+	+	+	+	+
22	D (SP)	48	CTX-M-15	tet	+	+	+	+	+	+
24 (c)	E (RS)	48	CTX-M-14	G na sxt	0	0	+	+	+	0
25 (c)	E (RS)	48	CTX-M-14	KTG na sxt	0	0	+	+	+	0
25 (c)	r (RS)	48	0	fq sxt tet	+	+	+	+	+	+

(B)
			Co-resistance		Mapping of the OXA-48 plasmid ^c
No. of the patient ^a	K. pneumoniae clonal pattern (origin of the sample)^b	Type of OXA-48-like carbapenemase	ESBL/cephalosporinase	Other	rep	traU	parA	blaOXA-48-ΔTir	Δtir-mucAB	rel L/M
1 (a)	A (RS)	48	0	0	+	+	+	+	+	+
1 (a)	A (RS)	48	0	0	+	+	+	+	+	+
3 (a)	a (RS)	48	0	0	+	+	+	+	+	+
5	B (CW)	48	0	0	+	+	+	+	+	+
6 (a)	A (TA)	48	0	0	+	+	+	+	+	+
7	c (RS)	48	CTX-M-15	KTA fq sxt	+	+	+	+	+	+
	B (BR)	48	0	KTG fq sxt	+	+	+	+	+	+
	B (BR)	48	0	KTG fq	+	+	+	+	+	+
	C (RS)	48	0	tet ft	+	+	+	+	+	+
	C (RS)	48	0	fq tet ft	+	+	+	+	+	+
	d (RS)	48	0	tet ft	+	+	+	+	+	+
	C (RS)	48	CTX-M-15	KTA fq sxt ft	+	+	+	+	+	+
	e (BA)	48	0	tet ft	+	+	+	+	+	+
	f (BA)	48	0	KTG fq tet ft	+	+	+	+	+	+
9 (a)	A (TA)	48	0	0	+	+	+	+	+	+
10	g (U)	48	0	0	+	+	+	+	+	+
10	h (RS)	48	0	0	+	+	+	+	+	+
11 (b)	D (SC)	48	CTX-M-15	KTGA fq sxt tet	+	+	+	+	+	+
11 (b)	D (U)	48	CTX-M-15	fq tet	+	+	+	+	+	+
12	i (U)	48	0	tet	+	+	+	+	+	+
13	j (RS)	48	0	0	+	+	+	+	+	+
22	k (SP)	48	0	0	+	+	+	+	+	+
23	l (NA)	48	CTX-M-15	KTG fq sxt	+	+	+	+	+	+
27	m (RS)	48	CTX-M-15	KTG fq sxt	+	+	+	+	+	+
28	n (U)	48	CTX-M-15	KTG fq sxt tet	+	+	+	+	+	+

(C)
			Co-resistance		Mapping of the OXA-48 plasmid ^c
No. of the patient ^a	Species identification (origin of the sample) ^b	Type of OXA-48-like carbapenemase	ESBL/cephalosporinase	Other	rep	traU	parA	blaOXA-48-ΔTir	Δtir-mucAB	rel L/M
7	Serratia marcescens (BR)	405	0	fq	+	+	+	+	+	0
	S. marcescens (BA)	405	0	fq	+	+	+	+	+	0
	S. marcescens (RS)	405	0	fq	+	+	+	+	+	0
	Citrobacter freundii (RS)	48	0	0	+	+	+	+	+	+
	C. freundii (RS)	48	0	fq	+	+	+	+	+	+
	Klebsiella oxytoca (RS)	48	0	0	+	+	+	+	+	+
13	Enterobacter cloacae (RS)	48	0	0	+	+	+	+	+	+
13	E. cloacae (SC)	48	0	0	+	+	+	+	+	+
15	C. freundii (SC)	48	0	0	+	+	+	+	+	+
16	Citrobacter koseri (SC)	48	0	0	+	+	+	+	+	+
26	E. cloacae (RS)	48	CTX-M-15	KTG fq sxt tet chlo ft	+	+	+	+	+	+
27	C. freundii (SC)	48	0	0	+	+	+	+	+	+
29	C. freundii (U)	48	0	0	+	+	+	+	+	+
30	E. cloacae (RS)	48	0	0	+	+	+	+	+	+

Epidemiologically related patients are flagged with a lowercase letter in brackets.

Capital letters indicate strains with the same profile in rep-PCR and ERIC-PCR for each species; lowercase letters indicate strains with single rep-PCR and ERIC-PCR profile.

+, positive; 0, negative.

A, amikacin; BA, bronchial aspiration; BR, breast sample; chlo, chloramphenicol; CW, cervical wound; ERIC-PCR, enterobacterial repetitive intergenic consensus sequence PCR; ESBL, extended-spectrum beta-lactamase; fos, fosfomycin; fq, fluoroquinolons; ft, furan; G, gentamicin; K, kanamycin; na, nalidixic acid; NA, not available; RS, rectal swab; SC, stool culture; SP, sputum; sxt, trimethoprim/sulfamethoxazol; T, tobramycin; TA, tracheal aspiration; tet, tetracyclin; U, urine sample.

Distribution of species and strain clonality

As previously described, a large diversity of species was observed, and we identified 7 different enterobacterial species in our collection of 68 isolates. Moreover, the assessment of clonality for E. coli and K. pneumoniae strains showed a great diversity. Indeed, 23 and 18 separate patterns were observed in the 29 E. coli and 25 K. pneumoniae strains, respectively (Fig. 1 and Table 3A and B). Thus, diversity was high between patients as well as within each patient, and identical patterns were observed only within a single patient or in epidemiologically related patients.

FIG. 1.

ERIC-PCR comparison of the Escherichia coli (top) and the Klebsiella pneumoniae (bottom) isolates. Patient's inclusion numbers are indicated for each pattern. Underlined numbers correspond to patterns found for more than one isolate. *Indicates the DNA molecular size marker (1 kb). ERIC-PCR, enterobacterial repetitive intergenic consensus sequence PCR.

Plasmid mapping and resistance transfer assays

Plasmid mapping was performed in the 62 OXA-48-producing and 3 OXA-405-producing strains and resulted in the identification of 5 different plasmid patterns (Table 3). The first pattern, which consisted of positivity for all of the targeted fragments, was shared by 87% (54/62) of the OXA-48 strains and corresponded most strongly to the pandemic pOXA-48a plasmid. Six of the remaining OXA-48 strains were positive only for parA and the junctions bla_OXA-48-Δtir and Δtir-mucAB. Interestingly, Patient No. 10 was carrying both patterns. All the three OXA-405 strains were positive for all of the fragments except rel L/M. Finally, only two OXA-48 strains had single patterns: the E. coli strain isolated from Patient No. 17, which was positive only for the junctions bla_OXA-48-Δtir and Δtir-mucAB, and the E. coli strain isolated from Patient No. 21, which was negative for all of the tested genes (Table 3A).

All representative strains of each clone lacking at least one molecular marker of the pandemic pOXA-48a plasmid failed to transfer any plasmid conferring resistance to carbapenems to the recipient strain.

Discussion

Although the dissemination of resistance genes has historically been associated with the spread of certain clones, it is increasingly mediated by mobile genetic elements such as plasmids. The major role of plasmids in the dissemination of CTX-M-type ESBL has already been demonstrated in the community and in hospitals.¹⁵ The high transfer rate of the IncL/M OXA-48-like plasmid, which has been previously demonstrated,¹⁰ and the evident fit between this plasmid and its various hosts are likely responsible for their rapid and global diffusion in the human community. Similarly, we have shown that the diversity may be high even within a single individual. Indeed, among the 30 carriers, 8 were carrying more than one species, and E. coli and K. pneumoniae were the species co-detected most frequently. Moreover, when the number of available strains made it possible, the carriage of multiple strains belonging to the same species was demonstrated.

Although it is well established that the OXA-48 epidemic is mainly driven by a few mobile genetic supports⁵ and that the transfer rate of the pOXA-48 plasmid is particularly high,¹⁰ the diversity of strains both at hospital admission and during hospitalization over time has been poorly documented. In this regard, our study clearly confirms the multiclonal nature of the pandemic since all the identified profiles were distinct from one patient to another, with the exception of epidemiologically related strains.

Another goal of our study was to explore the diversity of the OXA-48-like enzymes and to determine the nature of the plasmids involved in our patients. As expected, the vast majority of carbapenemases were identified as OXA-48. Among these samples, nearly all were tested positive in the PCR screen, which targeted genes either the backbone of the plasmid pOXA-48a (genes rep, traU, parA, and relL/M) or the genetic environment of the carbapenemase bla_OXA-48 (junctions bla_OXA-48-Δtir and Δtir-mucAB). Although complete sequencing of all the plasmids was not performed, the presence of these genes strongly suggested the presence of the pandemic plasmid pOXA-48a in these strains. However, some variations have also been observed. There were six E. coli strains isolated from four patients in which the genes rep, traU, and relL/M were absent. Such variants have already been identified in other studies and were in all cases associated with the ST38 E. coli clone in the United Kingdom, France, and the Czech Republic.^3,16,17 In accordance with the possible chromosomal location of bla_OXA-48 in these strains, the transfer assays failed despite several attempts. Although sequence typing was not performed in our study, the rep-PCR and ERIC-PCR patterns clearly indicated that the strains were different from each other, which excludes their belonging to a single clone. In our study, the very unusual OXA-405-producing S. marcescens strain lacked expression of the relaxase RelL/M. In fact, a recent publication showed that this particular strain results from the deletion of a portion of the plasmid pOXA-48a.¹⁸ Finally, two E. coli strains displayed single original profiles that should be more extensively studied.

Until recently, the acquisition of OXA-48 such as carbapenemases occurred mainly outside of Western Europe, either in hospitals^19,20 or in the community.^21,22 Bacteria expressing this enzyme are commonly detected in patients repatriated from North African or Eastern Mediterranean areas.²³ In our study, we showed that in half of the cases, OXA-48-like producing bacteria were acquired in France by patients who were mostly of French origin. Imported cases represented the other half of our cohort, represented mainly by patients from North African or East Mediterranean areas, who did acquire their MDR strains in their respective countries of origin. These results are in accordance with the growing number of carbapenemase outbreaks in Western countries that are not associated with foreign migration. This trend has also been observed in Belgian hospitals, where the incidence of carbapenemase-producing bacteria, including OXA-48-like-producing bacteria, is increasing among hospitalized patients.²⁴ The extraordinary ability of OXA-48 plasmids to transfer from one enterobacterial strain to another probably plays a major role in their rapid dissemination across the world. Similarly, the mobility associated with these plasmids results in the frequent emergence of strains harboring not only the OXA-48 carbapenemase but also many other broad-spectrum beta-lactamases, such as ESBL or cephalosporinases, as demonstrated in our study.

Although our study is one of the first to address the diversity of the OXA-48-like-producing strain carriage, it has several limitations. First, we have retrospectively explored a collection of strains, the diversity of which was based on the morphology of the colonies and the observed resistance phenotypes. A systematic and prospective assessment of the genetic background of the strains would have provided a more extensive assessment of the diversity. However, our sampling method had the advantage of targeting the more significant isolates. Another limitation to our study was that the follow-up of the patients was limited by the samples already available because this work was based on strains isolated from samples obtained for medical purposes. For this reason, the number of studied strains, which varied between 1 and 18 per patient, may have been underestimated in many carriers. A systematic follow-up of the cohort would probably have provided more information, as observed in Patient No. 7, whose carriage was continuously evolving over time.

Conclusion

Our study confirms the high potential for dissemination of OXA-48-bearing plasmids among enterobacteria. This transfer can occur both within and between species. Very high transfer rates have been demonstrated in vitro. Moreover, the polyclonal nature of the epidemic related to OXA-48-like enzymes has already been documented in epidemiological studies carried out in Western European hospitals.^11,25 In accordance with these data, our study illustrates the diversity of OXA-48-like bacteria in patients at the time of hospital admission and their diversification in hospitalized patients over time. Our results also show how quickly these enzymes have emerged worldwide and how they are currently becoming endemic in Western countries. These results imply that the practice of screening the most at-risk patients is of limited efficiency.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Nordmann

, Naas

, and Poirel

. 2011. Global spread of carbapenemase-producing enterobacteriaceae. Emerg. Infect. Dis., 17:1791–1798.

Poirel

, Heritier

, Tolun

, and Nordmann

. 2004. Emergence of oxacillinase-mediated resistance to imipenem in Klebsiella pneumoniae. Antimicrob. Agents Chemother., 48:15–22.

Potron

, Poirel

, Rondinaud

, and Nordmann

. 2013. Intercontinental spread of OXA-48 beta-lactamase-producing enterobacteriaceae over a 11-year period, 2001 to 2011. Euro Surveill., 18:31.

Lyman

, Walters

, Lonsway

, Rasheed

, Limbago

, and Kallen

. 2015. Notes from the field: carbapenem-resistant enterobacteriaceae producing OXA-48-like carbapenemases—United States, 2010–2015. MMWR Morb. Mortal. Wkly. Rep., 64:1315–1316.

Mathers

A.J.

, Hazen

K.C.

, Carroll

, Yeh

A.J.

, Cox

H.L.

, Bonomo

R.A.

, and Sifri

C.D.

. 2013. First clinical cases of oxa-48-producing carbapenem-resistant Klebsiella pneumoniae in the United States: the “menace” arrives in the new world. J. Clin. Microbiol., 51:680–683.

Beyrouthy

, Robin

, Dabboussi

, Mallat

, Hamze

, and Bonnet

. 2014. Carbapenemase and virulence factors of enterobacteriaceae in North Lebanon between 2008 and 2012: evolution via endemic spread of OXA-48. J. Antimicrob. Chemother., 69:2699–2705.

Djahmi

, Dunyach-Remy

, Pantel

, Dekhil

, Sotto

, and Lavigne

J.P.

. 2014. Epidemiology of carbapenemase-producing enterobacteriaceae and Acinetobacter baumannii in Mediterranean countries. Biomed. Res. Int., 2014:305784.

Evans

B.A.

, and Amyes

S.G.

. 2014. OXA beta-lactamases. Clin. Microbiol. Rev., 27:241–263.

Poirel

, Bonnin

R.A.

, and Nordmann

. 2012. Genetic features of the widespread plasmid coding for the carbapenemase OXA-48. Antimicrob. Agents Chemother., 56:559–562.

10.

Potron

, Poirel

, and Nordmann

. 2014. Derepressed transfer properties leading to the efficient spread of the plasmid encoding carbapenemase OXA-48. Antimicrob. Agents Chemother., 58:467–471.

11.

Arana

D.M.

, Saez

, Garcia-Hierro

, Bautista

, Fernández-Romero

, Ángel de la Cal

, Alós

J.I.

, and Oteo

. 2015. Concurrent interspecies and clonal dissemination of OXA-48 carbapenemase. Clin. Microbiol. Infect. 21:148.e141–e144.

12.

Dallenne

, Da Costa

, Decré

, Favier

, and Arlet

. 2010. Development of a set of multiplex PCR assays for the detection of genes encoding important beta-lactamases in enterobacteriaceae. J. Antimicrob. Chemother., 65:490–495.

13.

Eckert

, Gautier

, Saladin-Allard

, Hidri

, Verdet

, Ould-Hocine

, Barnaud

, Delisle

, Rossier

, Lambert

, Philippon

, and Arlet

. 2004. Dissemination of CTX-M-type beta-lactamases among clinical isolates of enterobacteriaceae in Paris, France. Antimicrob. Agents Chemother., 48:1249–1255.

14.

van Belkum

, Tassios

P.T.

, Dijkshoorn

, Haeggman

, Cookson

, Fry

N.K.

, Fussing

, Green

, Feil

, Gerner-Smidt

, Brisse

, Struelens

; European Society of Clinical Microbiology and Infectious Diseases (ESCMID) Study Group on Epidemiological Markers (ESGEM). 2007. Guidelines for the validation and application of typing methods for use in bacterial epidemiology. Clin. Microbiol. Infect., 13 Suppl 3:1–46.

15.

Woerther

P.L.

, Angebault

, Jacquier

, Hugede

H.C.

, Janssens

A.C.

, Sayadi

, El Mniai

, Armand-Lefèvre

, Ruppé

, Barbier

, Raskine

, Page

A.L.

, de Rekeneire

, and Andremont

. 2011. Massive increase, spread, and exchange of extended spectrum beta-lactamase-encoding genes among intestinal Enterobacteriaceae in hospitalized children with severe acute malnutrition in Niger. Clin. Infect. Dis., 53:677–685.

16.

Dimou

, Dhanji

, Pike

, Livermore

D.M.

, and Woodford

. 2012. Characterization of Enterobacteriaceae producing OXA-48-like carbapenemases in the UK. J. Antimicrob. Chemother., 67:1660–1665.

17.

Skalova

, Chudejova

, Rotova

, Medvecky

, Studentova

, Chudackova

, Lavicka

, Bergerova

, Jakubu

, Zemlickova

, Papagiannitsis

C.C.

, and Hrabak

. 2016. Molecular characterization of OXA-48-like-producing enterobacteriaceae in the Czech Republic: evidence for horizontal transfer of pOXA-48-like plasmids. Antimicrob. Agents Chemother., 61:e02648-16.

18.

Dortet

, Oueslati

, Jeannot

, Tande

, Naas

, and Nordmann

. 2015. Genetic and biochemical characterization of OXA-405, an OXA-48-type extended-spectrum beta-lactamase without significant carbapenemase activity. Antimicrob. Agents Chemother., 59:3823–3828.

19.

Dautzenberg

M.J.

, Ossewaarde

J.M.

, de Greeff

S.C.

, Troelstra

, and Bonten

M.J.

. 2016. Risk factors for the acquisition of OXA-48-producing enterobacteriaceae in a hospital outbreak setting: a matched case-control study. J. Antimicrob. Chemother., 71:2273–2279.

20.

Girlich

, Bouihat

, Poirel

, Benouda

, and Nordmann

. 2014. High rate of faecal carriage of extended-spectrum beta-lactamase and OXA-48 carbapenemase-producing Enterobacteriaceae at a university hospital in Morocco. Clin. Microbiol. Infect., 20:350–354.

21.

Halat

D.H.

, Moubareck

C.A.

, and Sarkis

D.K.

. 2017. Heterogeneity of carbapenem resistance mechanisms among Gram-negative pathogens in Lebanon: results of the first cross-sectional countrywide study. Microb. Drug Resist. [Epub ahead of print]; DOI: 10.1089/mdr.2016.0077.

22.

Ruppé

, Armand-Lefevre

, Estellat

, El-Mniai

, Boussadia

, Consigny

P.H.

, Girard

P.M.

, Vittecoq

, Bouchaud

, Pialoux

, Esposito-Farèse

, Coignard

, Lucet

J.C.

, Andremont

, and Matheron

. 2014. Acquisition of carbapenemase-producing enterobacteriaceae by healthy travellers to India, France, February 2012 to March 2013. Euro Surveill., 19:14.

23.

Decré

, Birgand

, Geneste

, Maury

, Petit

J.C.

, Barbut

, and Arlet

. 2010. Possible importation and subsequent cross-transmission of OXA-48-producing Klebsiella pneumoniae, France, 2010. Euro Surveill., 15:46.

24.

De Laveleye

, Huang

T.D.

, Bogaerts

, Berhin

, Bauraing

, Sacré

, Noel

, and Glupczynski

. 2017. Increasing incidence of carbapenemase-producing Escherichia coli and Klebsiella pneumoniae in Belgian hospitals. Eur. J. Clin. Microbiol. Infect. Dis., 36:139–146.

25.

Gottig

, Gruber

T.M.

, Stecher

, Wichelhaus

T.A.

, and Kempf

V.A.

. 2015. In vivo horizontal gene transfer of the carbapenemase OXA-48 during a nosocomial outbreak. Clin. Infect. Dis., 60:1808–1815.