New structures simultaneously harboring class 1 integron and ISCR1-linked resistance genes in multidrug-resistant Gram-negative bacteria.

BACKGROUND: The connection structure of class 1 integron and insertion sequence common region 1 (ISCR1) is called "complex class 1 integrons" or "complex sul1-type integrons", which is also known to be associated with many resistance genes. This structure is a powerful gene-capturing tool kit that can mobilize antibiotic resistance genes. In order to look for and study the structure among clinical multidrug-resistant (MDR) Gram-negative isolates, 63 isolates simultaneously harbored class 1 integron and ISCR1-linked resistance genes were isolated from 2309 clinical non-redundant MDR Gram-negative isolates in Nanfang Hospital in 2008-2013. The connecting regions between the class 1 integrons and ISCR1 were examined using PCR and DNA sequencing to determine the structures in these isolates. RESULT: The two elements (the variable regions of the class 1 integron structures and the ISCR1-linked resistance genes) are connected in series among 63 isolates according to long-extension PCR and DNA sequencing. According to the kinds and permutations of resistance genes in the structure, 12 distinct types were identified, including 8 types that have never been described in any species. Several types of these structures are similar with the structures of other reports, but not entirely same.
CONCLUSION: This study is the first to determine the structure simultaneously harboring class 1 integron and ISCR1-linked resistance genes by detecting the region connecting class 1 integrons and ISCR1 in a large number of MDR bacteria. These structures carrying various resistance genes were closely associated with multidrug resistance bacteria in Southern China.

Background

The increasing use of antimicrobial agents to treat Gram-negative bacterial infections has led to an increase in antibiotic resistance. Consequently, formerly routine therapies for many infectious diseases caused by multidrug-resistant (MDR) Gram-negative bacteria are now compromised. MDR bacteria evolve relatively quickly because the main driving force is lateral gene transfer, which is facilitated by a wide range of mobile genetic elements. The majority of these elements are integrons and transposons (including unit transposons and insertion sequences) [1]. Insertion sequences with common regions (ISCRs) are a type of insertion sequence.

In two previous studies, a total of 1329 and 1447 multidrug Gram-negative bacteria isolated in 2008–2009, was investigated for an ISCR1 [2] and a class 1 integron [3] respectively. In this study, 2309 clinical non-redundant MDR Gram-negative isolates were isolated between 2008 and 2013 at Nanfang Hospital, a 2200-bed tertiary-level teaching hospital in Guangzhou, China. Here, strains which carry a physical linkage between class 1 integrons and ISCR1 were focused. The ISCR1 and class 1 integrons were characterized using PCR and DNA sequencing as the methods described previously [2, 3]. Of these strains, 63 isolates simultaneously harbored class 1 integrons and ISCR1-linked resistance genes were selected. The results were shown on the Table 1. Based on the results, the region connecting the ISCR1 and the 3′-CS of the integron and the overall structures were investigated.

Table 1

Characterisation of complex class 1 integron in multidrug-resistant Gram-negative bacteria and resistance profiles of sequenced strains

Species	No. of isolates	Class 1 integron	ISCR	Type of complex class 1 integron
Escherichia coli	2	aadB-aadA2	qnrA1- ampR	B
	1	dfrA14-arr-2-bla OXA-10-aadA1	bla DHA-1-ampR	C
	1	dfrA25	sapA-like-qnrB2	D
	1	aacA4-arr-3-dfrA27- aadA16	short chain dehydrogenase/reductase- qnrB6	G
Enterobacter cloacae	2	aadB-aadA2	bla CTX-M-9-insB	A
	5	aadB-aadA2	qnrA1- ampR	B
	1	aacA4-arr-3-dfrA27- aadA16	short chain dehydrogenase/reductase- qnrB6	G
	1	aacA4-bla OXA-101-catB3-arr-3	qnrA1- ampR	K
Enterobacter aerogenes	2	aacA4-arr-3-dfrA27- aadA16	short chain dehydrogenase/reductase- qnrB6	G
Klebsiella pneumoniae	1	aadB-aadA2	qnrA1- ampR	B
	2	dfrA25	sapA-like-qnrB2	D
	5	dfrA12-orfF-aadA2	sapA-like-qnrB2	E
	9	aacA4-arr-3-dfrA27- aadA16	sapA-like-qnrB2	F
	15	aacA4-arr-3-dfrA27- aadA16	short chain dehydrogenase/reductase- qnrB6	G
Klebsiella oxytoca	1	aacA4-arr-3-dfrA27- aadA16	short chain dehydrogenase/reductase- qnrB6	G
Proteus mirabilis	1	bla PSE-1	dfrA10	L
Acinetobacter spp.	5	aadB-aadA2	qnrA1- ampR	B
	3	catB3-qnrVC-like-aacA4	bla PER-1-GST-novel type ABC transporter	H
	1	catB-like-aadB-aadA24-like	bla PER-1-GST-novel type ABC transporter	J
Pseudomonas. aeruginosa	2	aadB-aadA2	qnrA1- ampR	B
	1	aacA4-like-bla OXA-101-aadA5	bla PER-1-GST-novel type ABC transporter	I
Stenotrophomonas. maltophilia	1	aadB-aadA2	qnrA1- ampR	B

This structure which is usually called “complex class 1 integrons” or “complex sul1-type integrons”, [4] is the large genetic element in which different class 1 integrons is associated with ISCR1. These elements are known to be associated with many resistance genes, encoding resistance to chloramphenicol, trimethoprim, quinolone, and β-lactam, [5] and have two notable structures. Besides, this structure is a powerful gene-capturing tool kit that can mobilize antibiotic resistance genes. The most of their structure comprises a typical class 1 integron with a 5′ conserved segment (5′-CS), a 3′-CS, and an intervening variable region (VR1), followed by a copy of ISCR1 and then by an ISCR1-linked resistance gene region (VR2), which accommodates a variety of resistance genes. This region is, in turn, followed by a repetition of the 3′-CS [6, 7].

Methods

According to the previous study, sixty three isolates simultaneously harbored class 1 integrons and ISCR1-linked resistance genes were selected for further analysis of the connecting the ISCR1 and the 3′-CS of the integron.

According the hypothesis connecting structure (Fig. 1) between the ISCR1 and the 3′-CS of the integron, primers A and B were designed to amplify the 2045-bp fragment at the junction between ISCR1 and the 3′-CS of the class 1 integrons and used to preliminarily confirm that ISCR1 was inserted downstream from the sul1 gene in the class 1 integrons. Primers I and X, which are specific for the VR1 and VR2 resistance genes investigated in this study, were used to identify the connecting regions: downstream of class 1 integrons and upstream of ISCR1. The amplification products of primers I and X were used as the templates in subsequent PCR analyses. Five set primers (Fig. 1) were used to verify this region with the ISCR1 and the class 1 integrons connected in series.

Fig. 1

The hypothesis connecting structure between the ISCR1 and the 3′-CS of the integron. a Integrase were amplified by PCR using primer pairs intI1-F and intI-R. class 1 integrons were amplified by 5′CS and 3′CS.ISCR1 were amplified by PCR using primer pairs ISCR1-F and ISCR1-R. ISCR1-linked genes were amplified by PCR using primer pairs ISCR1-F and sul1-R. b Lengths of the PCR products obtained with primers A and B. c Lengths of the PCR products obtained with a series primers (I-X)

Primers of the connection region of integron and ISCR1 are listed in Table 2 and were synthesized by Beijing Genomics Institute (Shenzhen, China). PCR amplifications were performed using 1.5 μL of template, 2 μL of 10 × PCR buffer, 4 μM of each primer stock solution, 4 mM of each dNTP, 1 U of Ex Taq DNA polymerase (TaKaRa Bio Inc., Tokyo, Japan), and sterile distilled water added to a final total volume of 20 μL. Amplification was performed using a Mastercycler® PCR System (Eppendorf International, Hamburg, Germany). Thermocycling parameters were 94 °C for 5 min, 35 cycles of 94 °C for 30 s, 55–60 °C for 5 min, and 72 °C for 2–5 min; a final extension step at 72 °C was added for 5 min. Different annealing temperature and extension of time depended on different length of the PCR amplicons.

Table 2

Primers for PCR amplification of the connection region of integron and ISCR1

Type of complex class 1 integron	No.	Primer name	Sequence (5′–3′)
A	I	aadA2 + bla CTX-9F	CGTTGCCTTGGTAGGTCC
	II	aadA2 + bla CTX-9 up R	TTGCTTGCCATAGTCATCTT
	IX	aadA2 + bla CTX-9 down F	CCCCAAGGAGCCCATTC
	X	aadA2 + bla CTX-9R	GGTATTCAGCGTAGGTTCAGT
B	I	aadA2 + qnrA1 F	GTTGTCCCGCATTTGGT
	II	aadA2 + qnrA1 up R	GGTTGAGCGAGAAGGTTTT
	IX	aadA2 + qnrA1 down F	GCGTGAGCTGCCACCAGAA
	X	aadA2 + qnrA1 R	TCTTATGGCTGACTTGATTGTAG
C	I	aadA1 + bla DHA-1 F	ATCTGGCTATCTTGCTGAC
	II	aadA1 + bla DHA-1 up R	TTCCGAGAAGGTGATTGC
	IX	aadA1 + bla DHA-1 down F	CCAACACTGCTCAACACT
	X	aadA1 + bla DHA-1 R	GGTGGCGATTGTGATTCT
D	I	dfrA25+ sapA F	ACGAAGCGATGGGTAGA
	II	dfrA25+ sapA up R	AGCCCTCACGAGTTGTTAT
	IX	dfrA25+ sapA down F	CAAGAAGCCCGACAAAT
	X	dfrA25+ sapA R	TGGGAGGTGCTGGATAA
E	I	aadA2+ sapA F	CGTTGCCTTGGTAGGTC
	II	aadA2+ sapA up R	AACCGCACAATCTCGTC
	IX	aadA2+ sapA down F	CGCTGCTGATAGACGAAG
	X	aadA2+ sapA R	TGGGAGGTGCTGGATAA
F	I	aadA16+ sapA F	GTTGTTCCTTGGCGTTATC
	II	aadA16+ sapA up R	TCAGCAATATCGGGATAGAG
	IX	aadA16+ sapA down F	AGACGATACGCTGACTCA
	X	aadA16+ sapA R	ATGACCGACTGCTTGATG
G	I	dfrA27 + short chain F	GCAATGAGGGAGCTAAAGA
	II	dfrA27 + short chain up R	TTGGGTTCAGGGTGCTAT
	IX	dfrA27 + short chain down F	CAAGAAGCCCGACAAATC
	X	dfrA27 + short chain R	TTCACGAGCATAGGCAATA
H	I	aacA4 + bla PER F	CCCGAGGTCACCAAGA
	II	aacA4 + bla PERup R	GCACCATCCCACATAAGA
	IX	aacA4 + bla PERdown F	AAGAGGGCGAAGACGA
	X	aacA4 + bla PER R	TCCATCAGGCAACAGAAT
I	I	aadA5 + bla PER F	ACTGGTCTCATTGCTCCTA
	II	aadA5 + bla PER up R	CGAAGAACCGCACAATCT
	IX	aadA5 + bla PER down F	CAACACTGCTCAACACTG
	X	aadA5 + bla PER R	ATTGGTTCGGCTTGACTC
J	I	aadA24 + bla PER F	CATCATTCCGTGGCGTTA
	II	aadA24 + bla PER up R	GACACCGAGACCAATAGC
	IX	aadA24 + bla PER down F	AATCCAACACTGCTCAACA
	X	aadA24 + bla PER R	CATCATTCCGTGGCGTTA
K	I	arr-3 + qnrA1 F	GGTAATCCAACACAGTCCTA
	II	arr-3 + qnrA1 up R	GTCCGCCTCAGCAATATC
	IX	arr-3 + qnrA1 down F	TCCAACACTGCTCAACAC
	X	arr-3 + qnrA1 R	CCAGAGTATCCGCAATCC
L	I	bla PSE-1 + dfrA10 F	TTATGGCGGCGTTAGATG
	II	bla PSE-1 + dfrA10 up R	CGAGACCAATAGCGGAAG
	IX	bla PSE-1 + dfrA10 down F	ATATTGAAGTCTGCGAACAC
	X	bla PSE-1 + dfrA10 R	CGTGCTCTGTGATAGTTGA
-	III	common 1 791 F	TATTGCTGAGGCGGACTG
	IV	common 1 791 R	CATTGGAGGAGGTCGTTG
	V	common 2 1054 F	GGCTTCCGCTATTGGTC
	VI	common 2 1054 R	TTGCTTGCCATAGTCATCTT
	VII	common 3 1727 F	TCGCCCACTCAAACAAA
	VIII	common 3 1727 R	GCTCCTCATCCGAAGTATCT
	A	-	CCTGTCGGTGTTGCTTAT
	B	-	GTTGCTTGCCATAGTCATC

Types of complex class 1 integron correspond to Fig. 2;

Primers No. correspond to Fig. 1

Fig. 2

Twelve distinct structures connecting the variable regions of the class 1 integron structures and the ISCR1-linked resistance genes in MDR Gram-negative bacteria. The types of the structures were marked from A to L. ISCR1 is represented by red boxes; integrase is represented by yellow boxes; the qacEΔ1/sul1 is denoted by gray boxes; Open reading frames are indicated with open boxes having their own individual color and the direction of their transcription is indicated with arrows

The PCR amplicons were purified and sequenced (Sanger capillary sequencing) at the Beijing Genomics Institute (Shenzhen, China). The resulting DNA sequences were analysed with the BLAST program at the NCBI homepage (http://www.ncbi.nlm.nih.gov/blast/).

Results

Thirteen genes (qnrA1, qnrB2, qnrB6, ampR, dfrA10, blaDHA-1, blaCTX-M-9, blaPER-1, insB, sapA-like, gst, and those encoding a novel ABC transporter and a short-chain dehydrogenase/reductase) were detected in the ISCR1-linked resistance gene arrays. The gene cassettes of the class 1 integrons found in the 63 isolates included those encoding resistance to trimethoprim (dfrA12, dfrA25, dfrA27), aminoglycosides (aadA2, aadA16, aadB, aac(6′)-Ib, aacA4), chloramphenicol (catB3, catB8), quinolone (qnrVC-like), and rifampicin (arr-3, arr-2).

In this study, the structures (the variable regions of the class 1 integron structures and the ISCR1-linked resistance genes) are connected in series among 63 isolates. We found 12 distinct structures connecting the ISCR1 and the class 1 integron, with a different gene-cassette variable regions, composed of the 5′-CS and the 3′-CS but displaying another unique variable region located between ISCR1 and the second copy of the 3′-CS (Fig. 2). The PCR products amplified from the regions connecting the first 3′-CS and ISCR1 are shown in Fig. 3.

Fig. 3

Agarose gel electrophoresis of PCR products obtained with primers I and X for 12 isolates. primers I and X and the region correspond to Fig. 1. The following types corresponding to the Fig. 2 were showed by lane: 1, marker; 2 A; 3, B; 4, C; 5, D; 6, E; 7, F; 8, G; 9, H; 10, I; 11, J; 12, K; 13, L; 14, marker

Discussion

To the best of our knowledge, this is the first study to investigate the structure connecting the ISCR1 and the 3′-CS of the integron in different species in clinical isolates of MDR Gram-negative bacteria on a large scale. It was found that 63 isolates simultaneously carried class 1 integrons and ISCR1. The variable regions of the class 1 integron structures to the ISCR1-linked resistance genes were linked successfully using long-extension PCR, suggesting that these two structures are connected in series.

Of the 12 type structures, 8 types (A, C, F, H, I, J, K and L type) were first found in any species according to the systematic search in PubMed and GenBank. The structure type K is similar to a structure In37::ISCR1::qnrA1 (accession No. AY259086) [8]. A genetic structure (type B), [aadB + aadA2]:ISCR1:[qnrA1 + ampR], is the same as a structure already reported In293::ISCR1::qnrA1 in E. cloacae, [9] which was also found in E. coli, K. pneumoniae, Acinetobacter spp., P. aeruginosa and S. maltophilia in this study. We detected the structure D type, [dfrA25]:ISCR1:[sapA-like + qnrB2], that was previously found and described in Salmonella isolates [10], but which has never before been described in E. coli or K. pneumoniae. The structure G has been detected in clinical K. pneumoniae (accession No. JF775516) isolates in a previous study [11] but was first detected in E. coli, E. cloacae, E. aerogenes and K. oxytoca. Besides, the structure G was found in about 32 % isolates (20/63) in this study. The structure E type was previously reported by Ziyong Zong et al. in 2010 from China [12] (accession No. NG037697).

Among the 12 type structures, the VR-2 was similar as previous surveys, whereas the VR-1 was very different from previous surveys. Although the array of gene cassettes in the VR-1 can be easily exchanged, the distribution of ISCR1::qnrB2 in our survey revealed the presence of structure D, E and F. Worldwide, different VR-1 arrays in the structures carrying ISCR1::qnrB2 clinical isolates have been reported eg. In2, In27, In54, In73, In207 and In585 [13]. About structure H, I and J, Ruirui Xia et al. described similar structures carrying blaPER-1 and qnrVC-like genes and made an exhaustive study [14].

Enterobacteriaceae strains carrying the structures are becoming more common [15-17]. In this study, nine distinct structures were identified among Enterobacteriaceae strains, including seven distinct structures connecting the ISCR1 and the class 1 integron that have never been described in any species. It should be noted that only four distinct structures were identified in MDR non-fermenting isolates. One possible explanation is that chromosomal resistance mechanisms, such as efflux pumps, are more common than laterally transferred genetic resistance factors in these genera in this bacterial population.

Conclusions

This is the first study to describe the structure connecting the ISCR1 and the class 1 integron in clinical MDR Gram-negative bacterial isolates in a large-scale study. In total, 12 distinct structures were described. Several types of these structures are similar with the structure of other reports, but not entirely same. This structure is a powerful gene-capturing tool that can mobilize antibiotic-resistance genes. Therefore, the structural analysis of the structure connecting the ISCR1 and the class 1 integron could guide treatment strategies and provide directions for future research into the mechanisms of bacterial antibiotic resistance.

Nucleotide sequence accession number

The nucleotide sequences of the structure A to L in this work have been submitted to the GenBank database and assigned accession No. JX880393, JX880383, KM111274, JX880388, KM111278, KM111276, KM111280, JX880386, KM111273, KM111272, KM111271 and KM111275.