Dissemination of Multidrug-Resistant Shigella flexneri and Shigella sonnei with Class 1,Class 2,and Atypical Class 1 Integrons in China

Abstract

Background:

Emergence of multidrug-resistant Shigella, a major causative agent of bacterial dysentery, has generated many concerns not only in China but also worldwide. However, the prevalence of Shigella resistance caused by integron in the nonpopular season of diarrhea is not clear.

Materials and Methods:

Thirty-one Shigella flexneri and 22 Shigella sonnei samples collected in December 2010 from 10 cities of China were characterized for antimicrobial susceptibility, gene cassettes, widespread of integrons, and pulsed-field gel electrophoresis (PFGE) profile.

Results:

Multidrug resistance (MDR) was detected in 29 (93.5%) S. flexneri and 20 (90.9%) S. sonnei isolates. Class 1 integrons were detected in 25 (80.6%) S. flexneri and in 13 (59.1%) S. sonnei isolates; class 2 integrons were detected in 26 (83.9%) S. flexneri and in 19 (86.4%) S. sonnei isolates. Interestingly, the atypical class 1 integrons were mostly detected in S. flexneri (45.2%) isolates, whereas in only 1 (4.5%) S. sonnei isolate. DNA sequencing revealed two novel cassette arrays, dfrA5 and aacA4-cmlA, of class 1 integrons in S. flexneri, and dfrA17-aadA5 in S. sonnei isolates. The cassette arrays, dfrA1-sat1-aadA1 of class 2 integron and bla_oxa-30-aadA1 of atypical class 1 integron, were also identified. PFGE profiles demonstrated A6 subtype of S. flexneri strains prevalent in Shanghai, Changchun, Jinan, and Changsha; and F6 subtype of S. sonnei prevalent in Jinan, Changchun, and Shanghai.

Conclusion:

The dissemination of MDR Shigella strains with integrons makes it an increasing public health problem in China. Increased surveillance and the development of adequate prevention strategies are warranted.

Introduction

Shigella spp. are gram-negative intracellular pathogens that are responsible for 5–10% of diarrheal diseases in many areas.¹ There are four species of Shigella, including Shigella dysenteriae, Shigella flexneri, Shigella boydii, and Shigella sonnei, all of which could cause diarrhea. Approximately 1.1 million deaths are recorded annually, and 164.7 million people are affected by Shigellosis worldwide.² In China, Shigella is the most common enteropathogenic bacteria causing acute diarrhea, and the detection peak is in summer and autumn.³ S. sonnei is predominant in developed countries (70%), whereas S. flexneri is largely found in developing economies (60%).^2,4 Indeed, in 2000, 0.8–1.7 million cases of shigellosis were reported in China, of which the dominant serogroup was S. flexneri.⁵ So, it is urgent to reveal the prevalence of Shigella spp.

The current treatment of Shigella infection faces challenges in the world. Presently, the first-line antibiotics to treating Shigella infection include fluoroquinolones, third-generation cephalosporin; however, the emergence of multidrug-resistant Shigella has been a concern worldwide.^1,6–8 In our previous study, we found that 36.8% of S. flexneri isolates were resistant to norfloxacin in Jiangsu Province.⁹ The emerging of fluoroquinolone resistant may be related to many mechanisms, including multiple mutations in the genes encoding DNA gyrase (gyrA and gyrB) and topoisomerase IV (parC and parE), plasmid-mediated quinolone resistance (PMQR), and efflux pump mediators mechanisms.¹⁰ What's more, integrons and gene cassettes have recently been added to the list of gene transfer elements among gram-negative isolates.^11,12

We have previously reported class 1 and class 2 integrons in Shigella in the epidemic season of diarrhea from Nanjing, China.¹³ Pulsed-field gel electrophoresis (PFGE) analysis had proved that related groups of isolates may have circulating in recent years in many European countries.^14,15 In this study, we aimed to determine the molecular typing pattern, prevalence of integrons, and cassette arrays in Shigella strains from several crucial monitoring medical institutions of China in the nonpopular season of diarrhea.

Materials and Methods

Collection of bacterial isolates

Shigella isolates were collected from monitoring hospitals of 10 deferent cities in China (Table 1) from nonseasonal epidemic in December 2010. The isolates in group 1 were collected from nine general hospitals of nine cities from outside Jiangsu Province (Fig. 1), whereas in group 2, the isolates were collected from the Nanjing Center for Disease Control and Prevention in Nanjing in Jiangsu Province. All Shigella isolates, which were identified first by biochemical and serological tests in the hospital, were collected indiscriminately in December. The further identification and the confirmations of the 53 strains collected were performed using API identification system (bioMerieux, Marcy-l'Etoile, France) and slide agglutination test with type- and group-specific Shigella antisera (Lanzhou Institute of Biological Products, Lanzhou, China), respectively.

FIG. 1.

Distribution of isolates in Chinese cities. Shanghai (10 Shigella flexneri, 13 Shigella sonnei), Changchun (4 S. flexneri, 1 S. sonnei), Changsha (1 S. flexneri), Chongqing (1 S. sonnei), Shenzhen (1 S. flexneri), Jinan (4 S. flexneri, 5 S. sonnei), Beijing (1 S. flexneri), Lanzhou (1 S. sonnei), Wuhan (1 S. sonnei), and Nanjing (10 S. flexneri). Inset box shows South China Sea.

Table 1.

Distribution of Isolates Collected from Hospitals in China

	Hospitals	Level	Number of beds	Isolates (53)
	Hospitals	Level	Number of beds	Shigella flexneri (31)	Shigella sonnei (22)	Shigella dysenteriae (0)	Shigella boydii (0)	Total
Group 1 (outside Jiangsu Province)	Z	University-Affiliated	1,700	10	13	0	0	23
	SE	University-Affiliated	1,740	4	1	0	0	5
	X	University-Affiliated	3,500	1	0	0	0	1
	S	University-Affiliated	2,600	0	1	0	0	1
	SH	University-Affiliated	1,180	1	0	0	0	1
	J	University-Affiliated	1,080	4	5	0	0	9
	B	University-Affiliated	970	1	0	0	0	1
	L	University-Affiliated	2,400	0	1	0	0	1
	R	University-Affiliated	1,450	0	1	0	0	1
Group 2 (in Jiangsu Province)	N			10	0	0	0	10

B, Beijing Children's Hospital; J, Jinan Central Hospital Affiliated to Shandong University; L, Lanzhou University Second Hospital; N, Nanjing Center for Disease Control and Prevention; R, Renmin Hospital of Wuhan University; S, Southwest Hospital of Third Military Medical University; SE, The Second Hospital of Jilin University; SH, Shenzhen People's Hospital; X, Xiangya Hospital Central-South University; Z, Zhongshan Hospital Fudan University.

Antimicrobial susceptibility testing

In accordance with the Clinical and Laboratory Standards Institute (CLSI) guideline, we investigated, using disk diffusion, the resistance to the following 17 antimicrobial drugs: ceftazidime (CAZ, 30 μg), aztreonam (ATM, 30 μg), ampicillin (AMP, 10 μg), amoxicillin (AML, 10 μg), piperacillin (PRL, 100 μg), cefepime (FEP, 30 μg), amoxicillin/clavulanic acid (AMC, 30 μg), imipenem (IMP, 10 μg), meropenem (MEM, 10 μg), chloromycetin (CHL, 30 μg), tetracycline (TET, 30 μg), gentamicin (GEN, 120 μg), cefotaxime (CTX, 30 μg), ciprofloxacin (CIP, 5 μg), minocycline (MH, 30 μg), levofloxacin (LEV, 5 μg), and trimethoprim/sulfamethoxazole (SXT, 25 μg). Zones of inhibition were analyzed as per the criteria set by the CLSI.¹⁶ As a quality control strain, Escherichia coli ATCC 25922 was used.

PFGE typing

As per the method described on PulseNet (www.cdc.gov/pulsenet), profiling of complete genome of the Shigella isolates was carried out by PFGE. S. flexneri isolates genomic DNA was digested with the restriction endonuclease NotI (TaKaRa, Dalian, China), followed by gel electrophoresis on the CHEF Mapper system (Bio-Rad, Nanjing, China) at 6 V/cm for 17 h using a protocol that involved initial and final switch times of 5 and 35 sec, respectively. S. sonnei genomic DNA was digested with the restriction endonuclease XbaI (TaKaRa), followed by gel electrophoresis at 6 V/cm for 17 h. Conditions of electrophoresis included a 2.2-sec initial and a 54.2-sec final switch time, and Salmonella enterica serovar Braenderup H9812 served as a size marker.

Using Dice correlation coefficient, similarity indices among the isolates were determined with 0.5% optimization and 1.0% tolerance position. By using BioNumerics Fingerprinting program version 4.0 (Applied Maths, Kortrijk, Belgium), comparisons of the macrorestriction patterns were performed. A dendrogram was constructed using the average linkages by the unweighted-pair group method. The strains owning more than 85% of the same band would be concluded to the same strain; however, if more than 50% of the bands are different, they would be considered to be different strain, which was epidemiologically unrelated.

Resistance genes detecting

DNA was extracted from all the Shigella isolates according to a previous protocol.¹² Using primers shown in Table 2, resistance genes were detected in the isolates by PCR method. PCR amplification was performed simultaneously with an initial denaturation step of 5 min at 94°C, followed by 30 cycles of amplification with a three-step profile consisting of denaturation for 45 sec at 94°C, annealing for 45 sec at respective temperature, and extension for 60 sec at 72°C, and a final 5 min-extension step at 72°C. To ensure the accuracy of gene amplification, the amplification products of each gene were sequenced on an ABI 3730 automatic sequencer, and the sequences were compared with the National Center for Biotechnology Information database.

Table 2.

The Primers of the Resistant Genes

Primer	Sequence (5′–3′)	Expected size, kb	Annealing temperature, °C	Reference
TEM	F: ATGAGTATTCAACATTTCCG	0.86	50	¹⁷
TEM	R: CCAATGCTTAATCAGTGAGG	0.86	50	¹⁷
CTX-M	F: TTTGCGATGTGCAGTACCAGTAA	0.71	51	¹⁸
CTX-M	R: CGATATCGTTGGTGGTGCCATA	0.71	51	¹⁸
SHV	F: ATGCGTTATATTCGCCTGTG	0.86	58	¹⁹
SHV	R: AGCGTTGCCAGTGCTCGATC	0.86	58	¹⁹
KPC	F: AGGACTTTGGCGGCTCCAT	0.75	55	²⁰
KPC	R: TCCCTCGAGCGCGAGTCTA	0.75	55	²⁰
VEB	F: ACGGTAATTTAACCAGATAGG	0.97	46	¹⁹
VEB	R: ACCCGCCATTGCCTATGAGCC	0.97	46	¹⁹
PER	F: ATGAATGTCATTATAAAAGC	0.93	42	¹⁹
PER	R: AATTTGGGCTTAGGGCAGAA	0.93	42	¹⁹
qnrA	F: ATTTCTCACGCCAGGATTG	0.51	53	²¹
qnrA	R: GATCGGCAAAGGTTAGGTCA	0.51	53	²¹
qnrB	F: GATCGTGAA AGCCAGAAAGG	0.49	53	²¹
qnrB	R: ACGATGCCTGGTAGTTGTCC	0.49	53	²¹
qnrS	F: ACGACATTCGTCAACTGCAA	0.47	53	²¹
qnrS	R: TAAATTGGCACCCTGTAGGC	0.47	53	²¹

F, forward; R, reverse.

Integron detection

Degenerate PCR primers hep35 and hep36 shown in Table 3 were used to screen for class 1, class 2, and class 3 integrons in all isolates. For the classification of each integron, restriction fragment length polymorphism (RFLP) analysis was performed on the integrase PCR products after digestion by HinfI restriction enzyme.²² All the Shigella isolates were also used to detect the atypical class 1 integron using primers int1F and IS1R (Table 3), as described previously.²³

Table 3.

Oligonucleotides Used for Polymerase Chain Reaction Amplification of Integron

Primer	Sequence	Location	Product size, bp	Reference
hep35	5′-TGCGGGTYAARGATBTKGATTT-3′	intI1, intI2, and intI3	490	¹⁵
hep36	5′-CARCACATGCGTRTARAT-3′	intI1, intI2, and intI3
5′CS1	5′-GGCATCCAAGCAGCAAG-3′	5′CS of class 1 integron	Variable	¹²
3′CS1	5′-AAGCAGACTTGACCTGA-3′	3′CS of class 1 integron
5′CS2	5′-CGGGATCCCGGACGGCATGCACGATTTGTA-3′	5′CS of class 2 integron	Variable	¹²
3′CS2	5′-GATGCCATCGCAAGTACGAG-3′	3′CS of class 2 integron
int1F	5′-CGTAGAAGAACAGCAAGG-3′	intI1 of atypical class 1 integron	2,453	²³
IS1R	5′-AGTGAGAGCAGAGATAGC-3′	IS1 of atypical class 1 integron

Cassette region amplification

5′CS1 and 3′CS1 PCR primers (Table 3) were used to amplify the cassette regions of class 1 integrons, whereas 5′CS2 and 3′CS2 PCR primers were used to amplify the cassette regions of class 2 integrons, as described previously.¹³

RFLP and DNA sequencing analysis of cassette regions

The restriction enzyme HinfI was used to digest the PCR amplicons of the integron cassette regions. Amplicons that showed similar RFLP profiles were considered to have similar gene cassettes. Each restriction profile was represented with one or two cassette amplicons, which were chosen for characterization of the DNA sequences.

Statistical analysis

p-Values were calculated using the chi-square tests to assess the difference in the drug resistance between S. flexneri and S. sonnei strains (Table 4). A p-value <0.05 was considered statistically significant.

Table 4.

Resistant and Susceptible Rates to Shigella flexneri and Shigella sonnei from China

Antimicrobial agents	S. flexneri (n = 31)			S. sonnei (n = 22)			p^a
Antimicrobial agents	R%	I%	S%	R%	I%	S%	p^a
Aztreonam	35.5	3.2	61.3	9.1	9.1	81.8	0.067
Ampicillin	87.1	0.0	12.9	86.4	4.5	9.1	0.638
Amoxicillin	87.0	6.5	6.5	91.0	4.5	4.5	1.000
Piperacillin	71.0	12.9	16.1	90.9	0.0	9.1	0.131
Cefotaxime	35.5	0.0	64.5	18.2	0.0	81.8	0.223
Ceftazidime	3.2	0.0	96.8	4.5	0.0	95.5	1.000
Cefepime	29.0	3.2	67.8	9.1	9.1	81.8	0.165
Amoxicillin/clavulanic acid	74.2	12.9	12.9	59.1	18.2	22.7	0.584
Imipenem	0.0	0.0	100.0	0.0	0.0	100.0	1.000
Meropenem	0.0	0.0	100.0	0.0	0.0	100.0	1.000
Chloromycetin	83.9	0.0	16.1	0.0	0.0	100.0	<0.001
Tetracycline	100.0	0.0	0.0	95.5	0.0	4.5	0.415
Gentamicin120^b	6.5	0.0	93.5	18.2	36.4	45.4	<0.001
Ciprofloxacin	29.0	54.8	16.2	0.0	13.6	86.4	<0.001
Minocycline	41.9	35.5	22.6	4.5	31.8	63.7	0.001
Levofloxacin	29.0	0.0	71.0	0.0	0.0	100.0	0.007
Trimethoprim/sulfamethoxazole	61.3	0.0	38.7	95.5	0.0	4.5	0.008

The difference of proportion of isolates resistant to antibiotics.

Due to high drug resistance in the region, the research committee decided to replace Gentamicin with Gentamicin120.

I%, intermediate rates; R%, resistant rates; S%, susceptible rates.

Results

Antimicrobial resistance of Shigella isolates

Twenty-nine (93.5%) S. flexneri and 20 (90.9%) S. sonnei isolates showed multidrug resistance (MDR), that is, they exhibited resistance to 3 or more classes of drugs. For S. flexneri, the higher resistant rates were to tetracycline (100.0%), ampicillin (87.1%), amoxicillin (87.0%), chloromycetin (83.9%), amoxicillin/clavulanic acid (74.2%), piperacillin (71.0%), and trimethoprim/sulfamethoxazole (61.3%). For S. sonnei, the higher resistant rates were to tetracycline (95.5%), trimethoprim/sulfamethoxazole (95.5%), amoxicillin (91.0%), piperacillin (90.9%), and ampicillin (86.4%).

We also compared susceptibility data from S. flexneri and S. sonnei isolates. S. flexneri isolates showed significantly higher resistance to four antimicrobial agents than S. sonnei isolates, including ciprofloxacin (p < 0.001), levofloxacin (p = 0.007), chloromycetin (p < 0.001), and minocycline (p = 0.001). While S. sonnei isolates showed significantly higher resistance to gentamicin (p < 0.001) and trimethoprim/sulfamethoxazole (p = 0.008) than S. flexneri isolates (Table 4).

PFGE pattern

The 31 S. flexneri and 22 S. sonnei isolates were subjected to PFGE analysis, and all strains were typeable by this method. For S. flexneri, a total of 5 clusters (A, B, C, D, and E) and 19 PFGE profiles were obtained based on drawn dendrogram (Fig. 2). Cluster A was the main type (80.6%, 25/31) among S. flexneri isolates. For S. sonnei isolates, a total of 10 PFGE profiles were obtained based on drawn dendrogram, they all belong to cluster F (Fig. 3). Cluster F6, F2, and F5 were the three main PFGE subtypes among S. sonnei isolates.

FIG. 2.

S. flexneri dendrograms of PFGE fingerprints after NotI digestion. PFGE, pulsed-field gel electrophoresis.

FIG. 3.

S. sonnei dendrograms of PFGE fingerprints after XbaI digestion.

Distribution of resistance genes

Of the 31 S. flexneri isolates, 7 (22.6%) were positive for bla_CTX-M-14, 6 (19.4%) for bla_CTX-M-55, and 1 (3.2%) for bla_CTX-M-15; of 22 S. sonnei isolates, 5 (22.7%) were positive for bla_CTX-M-14 and 1 (4.6%) for bla_CTX-M-15. Four (12.9%) S. flexneri and 22 (100.0%) S. sonnei isolates carried non-Extended Spectrum Beta-Lactamases (ESBL) gene bla_TEM-1, and only 1 S. flexneri isolate carried bla_SHV-1 gene. Other ESBLs genes and bla_KPC were not found. The qnrA gene was detected in 19.4% (6/31) of S. flexneri and 59.1% (13/22) of S. sonnei strains, and the qnrB gene was detected in 12.9% (4/31) and 50.0% (11/22), respectively, and the qnrS gene was not found in all Shigella isolates (Table 5).

Table 5.

The Carrying Status of Resistant Genes in This Study

Resistant genes	S. flexneri (n = 31)	S. sonnei (n = 22)
Resistant genes	Positive isolates (%)	Positive isolates (%)
CTX-M
CTX-M-14	7 (22.6)	5 (22.7)
CTX-M-15	1 (3.2)	1 (4.6)
CTX-M-55	6 (19.4)	0 (0.0)
TEM-1	4 (12.9)	22 (100.0)
SHV-1	1 (3.2)	0 (0.0)
VEB	0 (0.0)	0 (0.0)
PER	0 (0.0)	0 (0.0)
KPC	0 (0.0)	0 (0.0)
qnrA	6 (19.4)	13 (59.1)
qnrB	4 (12.9)	11 (50.0)
qnrS	0 (0.0)	0 (0.0)

Distribution of integrons

Forty-nine (92.5%) of the 53 Shigella isolates contained at least one kind of integron. For S. flexneri, class 1 and class 2 integrons were detected in 80.6% (25/31) and 83.9% (26/31) strains, respectively, whereas for S. sonnei, they were detected in 59.1% (13/22) and 86.4% (19/22) strains, respectively. No class 3 integrons were found. The atypical class 1 integrons were mostly detected in S. flexneri, 45.2% (14/31) isolates showed positive. Only 1 (4.5%) of the 22 S. sonnei isolates showed atypical class 1 integron positive.

We also analyzed the integron pattern in S. flexneri and S. sonnei isolates. Among S. flexneri isolates, the most prevalent integron pattern was “class 1+class 2+atypical class 1 integron” (41.9%, 13/31; Table 6). However, the most common S. sonnei integron pattern was “class 1+class 2 integron” (54.5%, 12/22; Table 6). The other integron patterns in Shigella are shown in Table 6.

Table 6.

The Status of Integron Patterns in This Study

Integron pattern	S. flexneri (n = 31)	S. sonnei (n = 22)
Integron pattern	Positive isolates (%)	Positive isolates (%)
Class 1 integron only	0 (0.0)	0 (0.0)
Class 2 integron only	1 (3.2)	6 (27.3)
Atypical class 1 integron only	0 (0.0)	0 (0.0)
Class 1+class 2 integron	12 (38.7)	12 (54.5)
Class 1+atypical class 1 integron	0 (0.0)	0 (0.0)
Class 2+atypical class 1 integron	1 (3.2)	0 (0.0)
Class 1+class 2+atypical class 1 integron	13 (41.9)	1 (4.5)
Total	27 (87.1)	19 (86.4)

Location and gene cassette arrays of integrons

To identify gene cassette arrays of integrons in both class 1 and class 2, we amplified each gene cassette region. All Shigella isolates with class 2 integrons (100%, 45/45; Table 7) were positive for cassette regions. However, only 11 (28.9%) of the 38 Shigella isolates with class 1 integrons contained cassette regions. Then, cassette regions of atypical class 1 integrons were analyzed and 39.4% (15/38) isolates showed positive (Table 7).

Table 7.

Characteristics of Novel Integrons Identified in This Study

Approximate length, kb	Integron type	S. flexneri	S. sonnei	Gene cassette(s) and order	Novel GenBank No.
0.75	Class 1	3	0	dfrA5	JN651401
1.6	Class 1	0	7	dfrA17-aadA5
2.2	Class 1	1	0	aacA4-cmlA1	JN651402
2.2	Class 2	26	19	dfrA1-sat1-aadA1
2.4	Atypical class 1	14	1	bla_oxa-30-aadA1

For class 1 integrons, a total of three different lengths of amplicons (0.75, 1.6, and 2.2 kb) were found. Analysis of sequences of the PCR products for each RFLP pattern showed one gene cassette (dfrA5) in the 0.75 kb amplicon, two aacA4-cmlA1 in the 2.2 kb amplicon, and dfrA17-aadA5 in the 1.6 kb size amplicon (Table 7). For class 2 integrons, three gene cassettes were present in the 2.2 kb amplicon with the same cassette array, dfrA1-sat1-aadA1. For a typical class 1 integrons, two gene cassettes present in the 2.4 kb amplicon with the array bla_oxa-30-aadA1 (Table 7).

Nucleotide sequence accession numbers

Two cassette arrays (dfrA5 and aacA4-cmlA1) in class 1 integron were first detected in S. flexneri isolates. In the GenBank database, the following accession numbers JN651401 for dfrA5 and JN651402 for aacA4-cmlA1 were retrieved.

Discussion

During the recent decades, resistance to several clinical antimicrobial agents has been observed among members of Shigella such as S. flexneri and S. sonnei strains.²⁴ In this study, most S. flexneri strains had acquired the resistance to tetracycline (100.0%), ampicillin (87.1%), amoxicillin (87.0%), amoxicillin/clavulanic acid (74.2%), piperacillin (71.0%), and trimethoprim/sulfamethoxazole (61.3%). Most S. sonnei strains showed resistance to tetracycline (95.5%), trimethoprim/sulfamethoxazole (95.5%), amoxicillin (91.0%), piperacillin (90.9%), and ampicillin (86.4%). The prevalence of resistance to antimicrobial agents was reported to be comparable among developing nations,^25,26 but higher than that of developed countries.²⁷ This may relate to the intensive clinical use of these drugs. This study has demonstrated that the empirical use of antimicrobial drugs to treat shigellosis in China is not appropriate.

In the management of Shigella infections, quinolones and third-generation cephalosporins are still the mainstream therapeutic options for adult patients. However, in this study, S. flexneri isolates showed 35.5% resistant rates to cefotaxime, cefepime, and ciprofloxacin, which were higher than previous reports in both developed and developing countries.^1,27–29 Resistance is lower among S. sonnei isolates; the resistant rates to the third-generation cephalosporins ranged from 4.5% to 18.2%, which were similar as other reports.³⁰ Interestingly, there were no ciprofloxacin-resistant S. sonnei, but we did observe isolates with decreased susceptibility. Similar results were also reported in S. sonnei isolates from Hangzhou area of China.³¹ In the recent meta-analysis of quinolone resistance in Shigella, the resistance to ciprofloxacin in S. sonnei isolates was lower than 1.0%.¹ Therefore, ciprofloxacin can be used to treat S. sonnei infection.

The production of ESBLs is the major defense mechanism against third-generation cephalosporins employed by Enterobacteriaceae. In this study, the CTX-M, SHV, and TEM families have been detected in Shigella species. 45.2% of S. flexneri and 100% of S. sonnei harbored CTX-M type ESBL and TEM-1 type, respectively. Similar reports from Anhui and Jiangsu Provinces located in Eastern China.^32,33 Simultaneously, the qnrA gene was detected in 19.4% (6/31) of S. flexneri and 59.1% (13/22) of S. sonnei strains, and the qnrB gene was detected in 12.9% (4/31) and 50.0% (11/22), respectively. The association of ESBL genes and PMQR genes had been previously reported in various countries, such as Spain, Germany, and India.^34–36 The PMQR genes are lowly prevalent in clinical isolates of Shigella. PMQR determinants along with β-lactamases, and class 1 integrons on a transferable plasmid may spread to other bacterial species by horizontal exchange, which lead to the emergence and spread of multidrug-resistant pathogens. This high prevalence of ESBL-producing and coexistence resistance mechanism poses a major threat for the dissemination. Therefore, tracking and monitoring of the rapid spread of ESBL and PMQR-producing Shigella isolates are urgently required.

Most Shigella isolates (>90.0%) in this study showed MDR, which was higher than previous reports.^25,37 Due to the rising cases of MDR, there is increased interest to understand the genetic basis of resistance to antimicrobial agents.

MDR strains are thought to involve the ability of integrons to capture and spread the genes associated with resistance.^38,39 It has been shown that S. sonnei isolates express class 2 integrons.^40,41 However, class 2 integrons were frequently detected in both S. flexneri isolates (83.9%) and S. sonnei (86.4%) isolates in this study, which was a little higher than that in Japan (76.9%),⁴² but lower than that in Korea (100.0%)⁴³ and Brazil (90.3%).⁴¹ All the class 2 integron-positive isolates in this study carried the same cassette array with three resistance genes, namely aminoglycoside adenylyltransferase (aadA1), streptothricin acetyltransferase (sat1), and dihydrofolate reductase (dfrA1), conferring resistance to spectinomycin/streptomycin, streptothricin, and trimethoprim, respectively. The presence of class 2 integrons in almost all the Shigella isolates could explain the high trimethoprim/sulfamethoxazole resistance.

Clusters comprised S. flexneri and S. sonnei isolates from different cities, and in some cases, they showed 100% genetic homology. In particular, within cluster A, one subtype (A6) included seven S. flexneri strains, of which were from Shanghai, Changchun, Jinan, and Changsha. This indicated that the A6 subtype was prevalent in these four cities of China. The other main subtype included four S. flexneri strains was A10, indicating that this subtype was prevalent in Nanjing area of China. There were three main PFGE subtypes among S. sonnei isolates, including F6, F2, and F5. The isolates of F6 subtype were from Jinan, Changchun, and Shanghai, demonstrating that it spread among these three cities in China. The other two subtypes, F2 and F5, were both prevalent in Shanghai area of China.

We also detected the class 1 and atypical class 1 integrons in all the isolates. Class 1 integrons were detected in 80.6% of S. flexneri and 59.1% of S. sonnei strains, which were significantly higher than those reported previously, including 3.8%, 3.2%, and 14.9% in Japan, Brazil, and Korea, respectively.^41–43 Class 1 and class 2 integrons were present in S. flexneri (38.7%) and S. sonnei (54.5%) isolates in this study. The widespread distribution and coexistence of class 1 and class 2 integrons in Shigella isolates may relate to the MDR phenotype. DNA sequencing of cassette regions of class 1 integrons revealed three different cassette arrays, including dfrA5, aacA4-cmlA1, and dfrA17-aadA5. To the best of our knowledge, the first two cassette arrays were first reported in S. flexneri isolates and got the GenBank accession numbers JN651401 and JN651402, respectively. These novel cassette arrays found in S. flexneri showed the resistance gene-capturing, collecting, and dissemination capacities of integrons. The class 1 integron gene cassette, dfrA17-aadA5, is a rare gene, which was only reported during a Shigella outbreak before.^43,44 However, in our previous study, it was prevalent in Enterobacteriaceae.¹³ Now, it is also prevalent in S. flexneri, reflecting the horizontal transfer of integrons within different species of Enterobacteriaceae.

S. flexneri strain YSH600 in Japan was found to harbor the atypical class 1 integron.⁴⁵ In this study, the atypical class 1 integrons were detected in nearly half of S. flexneri isolates, but only in one S. sonnei isolate. Sequences of the YSH6000 strain and that of the gene cassette bla_oxa-30-aadA1 found in this study were identical. This indicated the cassette array bla_oxa-30-aadA1 likely lead to the resistance to ampicillin, tetracycline, and chloromycetin. It is notable that our findings show that 41.9% S. flexneri strains had typical class 1 and class 2 integrons and atypical class 1 integrons, a rare phenomenon compared with previous studies. The coexistence of three different integrons in S. flexneri strains makes them more easily to capture more resistance genes. Moreover, if subjected to antibiotic selection pressure, integrons that are found in conjugatable plasmids are likely to capture and disseminate genes associated with resistance.²² PFGE typing of all Shigella strains revealed that cluster A6 was prevalent in S. flexneri strains from Shanghai, Changchun, Jinan, and Changsha, and A10 only spread in Nanjing. Among S. sonnei isolates, cluster F6 was prevalent in Jinan, Changchun, and Shanghai. Similarly, clusters F2 and F5 were only found in Shanghai. These data showed the dissemination of MDR S. flexneri and S. sonnei among several cities in China. Thus, these pathogenic MDR Shigella could prove to be a serious public health problem in China. More stringent guidelines are required for effective control of infections to guard against the spread of MDR Shigella strains.

Conclusion

Our study demonstrated the dissemination of MDR S. flexneri and S. sonnei with typical class 1 and class 2 integrons and atypical class 1 integrons in China. The presence of integrons, especially the coexistence of two or three different integrons in these strains, may have important clinical implications, as multiple gene cassettes could be captured easily leading to MDR. The emergence of resistance to third-generation cephalosporins and quinolones is worrisome. Increased surveillance and the development of adequate prevention strategies are warranted.

Footnotes

Acknowledgments

The authors thank the following nine hospitals that facilitated the collection of samples: Zhongshan Hospital Fudan University, The Second Hospital of Jilin University, Xiangya Hospital Central-South University, Shenzhen People's Hospital, Jinan Central Hospital Affiliated to Shandong University, Beijing Children's Hospital, Lanzhou University Second Hospital, Renmin Hospital of Wuhan University, and Southwest Hospital of Third Military Medical University. We also thank Nanjing Center for Disease Control and Prevention of China for the strain collection.

Disclosure Statement

No competing financial interests exist.

References

, Cao

, Pan

S.Y

, Zhuang

, R. Yu

, Peng, Qian

H.M

, Wei

, Zhao

, Liu

, and Tong

M.Q.

2012. Comparison of the prevalence and changing resistance to nalidixic acid and ciprofloxacin of Shigella between Europe-America and Asia-Africa from 1998 to 2009. Int. J. Antimicrob. Agents, 40:9–17.

Kotloff

K.L.

, Winickoff

J.P

, Ivanoff

, Clemens

J.D

, Swerdlow

D.L

, Sansonetti

P.J

, Adak

G.K

, and Levine

M.M.

1999. Global burden of Shigella infections: implications for vaccine development and implementation of control strategies. Bull. World Health Organ. 77:651–666.

Zhang

, Si

, Wang

, and Gong

2016. Patterns of bacillary dysentery in China, 2005–2010. Int. J. Environ. Res. Public Health, 13:164.

Qiu

, Xu

, Yang

, Wang

, Liang

, Li

, Yi

, Liu

, Cui

, Wu

, Xie

, Jia

, Wang

, Hao

, Jin

, Wang

, Sun

, and Song

2015. Shift in serotype distribution of Shigella species in China, 2003–2013. Clin. Microbiol. Infect. 21:252.e5–e8.

Wang

X.Y.

, Tao

, Xiao

, Lee

, Deen

, Gong

, Zhao

, Zhou

, Li

, Shen

, Song

, Ma

, Li

Z.M

, Wang

, Su

P.Y

, Chang

, Xu

J.H

, Ouyang

P.Y

, von Seidlein

, Xu

Z.Y

, and Clemens

J.D.

2006. Trend and disease burden of bacillary dysentery in China (1991–2000). Bull. World Health Organ. 84:561–568.

, Zhou

, Ke

, Pan

, Cao

, Huang

, Zhuang

, Liu

, and Tong

2015. Comparison of resistance to third-generation cephalosporins in Shigella between Europe-America and Asia-Africa from 1998 to 2012. Epidemiol. Infect. 143:2687–2699.

Qiu

, Xu

, Wang

, Yang

, Wang

, Zhang

, Liu

, Chen

, Liu

, Li

, Su

, Jia

, Wang

, Jin

, Keim

, Yuan

, Huang

, and Song

2012. Emergence of resistance to fluoroquinolones and third-generation cephalosporins in Shigella flexneri subserotype 1c isolates from China. Clin. Microbiol. Infect. 18:E95–E98.

Zhang

, Luo

, Li

, Lin

, Ma

, Hu

, Jin

, Ran

, and Cui

2011. Wide dissemination of multidrug-resistant Shigella isolates in China. J. Antimicrob. Chemother. 66:2527–2535.

Qin

T.T.

, Bi

R.R

, Fan

W.F

, Kang

H.Q

, Ma

, and Gu

2016. Novel mutations in quinolone resistance-determining regions of gyrA, gyrB, parC and parE in Shigella flexneri clinical isolates from eastern Chinese populations between 2001 and 2011. Eur. J. Clin. Microbiol. Infect. Dis. 35:2037–2045.

10.

Fàbrega

, Madurga

, Giralt

, and Vila

2009. Mechanism of action of and resistance to quinolones. Microb. Biotechnol. 2:40–61.

11.

Bennett

P.M.

1999. Integrons and gene cassettes: a genetic construction kit for bacteria. J. Antimicrob. Chemother. 43:1–4.

12.

, Tong

, Zhao

, Liu

, Ning

, Pan

, and Zhao

2007. Prevalence and characterization of class I integrons among Pseudomonas aeruginosa and Acinetobacter baumannii isolates from patients in Nanjing, China. J. Clin. Microbiol. 45:241–243.

13.

, Pan

, Wang

, Zhao

, Mei

, Huang

, and Tong

2008. Novel cassette arrays of integrons in clinical strains of Enterobacteriaceae in China. Int. J. Antimicrob. Agents. 32:529–533.

14.

DeLappe

, O'Halloran

, Fanning

, Corbett-Feeney

, Cheasty

, and Cormican

2003. Antimicrobial resistance and genetic diversity of Shigella sonnei isolates from western Ireland, an area of low incidence of infection. J. Clin. Microbiol. 41:1919–1924.

15.

Surdeanu

, Ciudin

, Pencu

, and Straut

2003. Comparative study of three different DNA fingerprint techniques for molecular typing of Shigella flexneri strains isolated in Romania. Eur. J. Epidemiol. 18:703–710.

16.

Clinical Performance Standards for Antimicrobial Susceptibility Testing. 2012. 22th Information Supplement. M100-S22. CLSI, Wayne, PA.

17.

Bhattacharjee

, Sen

M.R

, Anupurba

, Prakash

, and Nath

2007. Detection of OXA-2 group extended-spectrum-beta-lactamase-producing clinical isolates of Escherichia coli from India. J. Antimicrob. Chemother. 60:703–704.

18.

Edelstein

, Pimkin

, Palagin

, Edelstein

, and Stratchounski

2003. Prevalence and molecular epidemiology of CTX-M extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. Antimicrob. Agents Chemother. 47:3724–3732.

19.

Navonvenezia

, Chmelnitsky

, Hammermünz

, Leavitt

, Gold

H.S

, Schwaber

M.J

, and Carmeli

2005. Extended-spectrum beta-lactamases among enterobacter isolates obtained in Tel Aviv, Israel. Antimicrob. Agents Chemother. 49:1150–1156.

20.

Wang

, Chen

, Guo

, Xiong

, Hu

, Zhu

, and Zhang

2011. Occurrence of false positive results for the detection of carbapenemases in carbapenemase-negative Escherichia coli and Klebsiella pneumonia isolates. PLoS One, 6:e26356.

21.

Robicsek

, Strahilevitz

, Sahm

D.F

, Jacoby

G.A

, and Hooper

D.C.

2006. qnr Prevalence in ceftazidime-resistant Enterobacteriaceae isolates from the United States. Antimicrob. Agents Chemother. 50:2872–2874.

22.

White

P.A.

, McIver

C.J

, and Rawlinson

W.D.

2001. Integrons and gene cassettes in the Enterobacteriaceae. Antimicrob. Agents Chemother. 45:2658–2661.

23.

Pan

J.C.

, Ye

, Meng

D.M

, Zhang

, Wang

H.Q

, and Liu

K.Z.

2006. Molecular characteristics of class 1 and class 2 integrons and their relationships to antibiotic resistance in clinical isolates of Shigella sonnei and Shigella flexneri. J. Antimicrob. Chemother. 58:288–296.

24.

Niyogi

S.K.

2007. Increasing antimicrobial resistance—an emerging problem in the treatment of shigellosis. Clin. Microbiol. Infect. 13:1141–1143.

25.

Tariq

, Haque

, Ali

, Bashir

, Habeeb

M.A

, Salman

, and Sarwar

2012. Molecular profiling of antimicrobial resistance and integron association of multidrug-resistant clinical isolates of Shigella species from Faisalabad, Pakistan. Can. J. Microbiol. 58:1047–1054.

26.

Ahmed

, Hoque

, Elahi

M.S

, Endtz

H.P

, and Hossain

M.A.

2012. Bacterial aetiology of diarrhoeal diseases and antimicrobial resistance in Dhaka, Bangladesh, 2005–2008. Epidemiol. Infect. 140:1678–1684.

27.

Wong

M.R.

, Reddy

, Hanson

, Johnson

K.M

, Tsoi

, Cokes

, Gallagher

, Lee

, Plentsova

, Dang

, Krueger

, Joyce

, and Balter

2010. Antimicrobial resistance trends of Shigella serotypes in New York City, 2006–2009. Microb. Drug Resist. 16:155–161.

28.

Folster

J.P.

, Pecic

, Bowen

, Rickert

, Carattoli

, and Whichard

J.M.

2011. Decreased susceptibility to ciprofloxacin among Shigella isolates in the United States, 2006 to 2009. Antimicrob. Agents Chemother. 55:1758–1760.

29.

Taneja

, Mewara

, Kumar

, Verma

, and Sharma

2012. Cephalosporin-resistant Shigella flexneri over 9 years (2001–2009) in India. J. Antimicrob. Chemother. 67:1347–1353.

30.

, Bao

, Chen

, Cui

, Guo

, Wang

, Zhang

, Wang

, Tang

Y.W

, and Mao

2012. Genotypes and antimicrobial profiles of Shigella sonnei isolates from diarrheal patients circulating in Beijing between 2002 and 2007. Diagn. Microbiol. Infect. Dis. 74:166–170.

31.

X.Y.

, Pan

J.C

, Wang

H.Q

, Zhang

, Huang

Z.C

, and Gu

Y.M.

2009. Characterization of fluoroquinolone-resistant Shigella flexneri in Hangzhou area of China. J. Antimicrob. Chemother. 63:917–920.

32.

Shen

, Qian

, Gong

, Deng

, Dong

, Zhou

, and Guo

2013. High prevalence of antibiotic resistance and molecular characterization of integrons among Shigella isolates in Eastern China. Antimicrob. Agents Chemother. 57:1549–1551.

33.

Yang

, Chen

, Zhu

, Liu

, Cheng

, Hu

, Ye

, and Li

2013. Surveillance of antimicrobial susceptibility patterns among Shigella species isolated in China during the 7-year period of 2005–2011. Ann. Lab. Med. 33:111–115.

34.

González-López

J.J.

, Piedra-Carrasco

, Salvador

, Rodríguez

, Sánchez-Montalvá

, Planes

A.M

, Molina

, and Larrosa

M.N

. 2014. ESBL-producing Salmonella enterica serovar Typhi in traveler returning from Guatemala to Spain. Emerg. Infect. Dis. 20:1918–1920.

35.

Schmiedel

, Falgenhauer

, Domann

, Bauerfeind

, Prenger-Berninghoff

, Imirzalioglu

, and Chakraborty

2014. Multiresistant extended-spectrum β-lactamase-producing Enterobacteriaceae from humans, companion animals and horses in central Hesse, Germany. BMC Microbiol. 14:187.

36.

Muthuirulandi Sethuvel

D.P.

, Anandan

, Devanga Ragupathi

N.K

, Veeraraghavan

, Vinod

, and Walia

2015. Association of bla_CTX-M-15 and qnr genes in multidrug-resistant Salmonella Typhimurium and Shigella spp from India. J. Infect. Dev. Ctries. 9:1294–1297.

37.

Vrints

, Mairiaux

, Van Meervenne

, Collard

J.M

, and Bertrand

2009. Surveillance of antibiotic susceptibility patterns among Shigella sonnei strains isolated in Belgium during the 18-year period 1990 to 2007. J. Clin. Microbiol. 47:1379–1385.

38.

Rowe-Magnus

D.A.

, Guerout

A.M

, and Gu

Y.M.

2002. Bacterial resistance evolution by recruitment of super-integron gene cassettes. Mol. Microbiol. 43:1657–1669.

39.

, Gu

, Pan

, and Tong

2011. Epidemiology and molecular mechanism of integron-mediated antibiotic resistance in Shigella. Arch. Microbiol. 193:767–774.

40.

Gassama-Sow

, Diallo

M.H

, Boye

C.S

, Garin

, Sire

J.M

, Sow

A.I

, and Aïdara-Kane

2006. Class 2 integron-associated antibiotic resistance in Shigella sonnei isolates in Dakar, Senegal. Int. J. Antimicrob. Agents, 27:267–270.

41.

Peirano

, Agersø

, Aarestrup

F.M

, and dos Prazeres Rodrigues

2005. Occurrence of integrons and resistance genes among sulphonamide-resistant Shigella spp. from Brazil. J. Antimicrob. Chemother. 55:301–305.

42.

Ahmed

A.M.

, Furuta

, Shimomura

, Kasama

, and Shimamoto

2006. Genetic characterization of multidrug resistance in Shigella spp. from Japan. J. Med. Microbiol. 55:1685–1691.

43.

J.Y.

, Yu

H.S

, Kim

S.K

, Seol

S.Y

, Cho

D.T

, and Lee

J.C.

2003. Changes in patterns of antimicrobial susceptibility and integron carriage among Shigella sonnei isolates from southwestern Korea during epidemic periods. J. Clin. Microbiol. 41:421–423.

44.

Navia

M.M.

, Gascón

, and Vila

2005. Analysis of the mechanisms of resistance to several antimicrobial agents in Shigella spp. causing travellers' diarrhoea. Clin. Microbiol. Infect. 11:1044–1047.

45.

Luck

S.N.

, Turner

S.A

, Rajakumar

, Sakellaris

, and Adler

2001. Ferric dicitrate transport system (Fec) of Shigella flexneri 2a YSH6000 is encoded on a novel pathogenicity island carrying multiple antibiotic resistance genes. Infect. Immun. 69:6012–6021.