Development of a Multiplex PCR Assay for Detection and Genogrouping of Neisseria meningitidis

Abstract

Neisseria meningitidis is a leading pathogen of epidemic bacterial meningitis and fulminant sepsis worldwide. Twelve different N. meningitidis serogroups have been identified to date based on antigenic differences in the capsular polysaccharide. However, more than 90% of human cases of N. meningitidis meningitis are the result of infection with just five serogroups, A, B, C, W135, and Y. Efficient methods of detection and genogrouping of N. meningitidis isolates are needed, therefore, in order to monitor prevalent serogroups as a means of disease control and prevention. The capsular gene complex regions have been sequenced from only seven out of the 12 serogroups. In this study, the capsular gene complexes of the remaining five serogroups were sequenced and analyzed. Primers were designed that were specific for N. meningitidis species and for the 12 individual serogroups, and a multiplex PCR assay using these specific primers was developed for N. meningitidis detection and genogrouping. The assay was tested using 15 reference strains covering all 12 serogroups, 143 clinical isolates, and 21 strains from closely related species or from species that cause meningitis. The assay could detect N. meningitidis serogroups and was shown to be specific, with a detection sensitivity of 1 ng of genomic DNA (equivalent to ∼4 × 10⁵ genomes) or 3 × 10⁵ CFU/ml in noncultured mock cerebrospinal fluid (CSF) specimens. This study, therefore, describes for the first time the development of a molecular protocol for the detection of all N. meningitidis serogroups. This multiplex PCR-based assay may have use for the clinical diagnosis and epidemiological surveillance of N. meningitidis.

INTRODUCTION

Neisseria meningitidis is an encapsulated Gram-negative bacterium and the leading pathogenic cause of epidemic bacterial meningitis and fulminant sepsis worldwide. Infections with N. meningitidis are a significant cause of mortality and morbidity in young children and adolescents (16, 23). The bacterial capsular polysaccharide (CPS) is the most important virulence factor for N. meningitidis (14, 20, 26, 28, 32). Twelve different serogroups, A, B, C, 29E, H, I, K, L, W135, X, Y, and Z, have been identified based on antigenic variation in the CPSs (6, 22). More than 90% of human cases of N. meningitidis meningitis are the result of infection with serogroups A, B, C, W135, and Y (21). Rare meningitis cases caused by serogroup K and Z infections have also been reported over the last few decades (11, 25). Detection and genogrouping of N. meningitidis isolates are important to monitor changes in the population of prevalent circulating serogroups for the purposes of disease control and prevention, vaccination strategy, and contact management (3, 12, 30).

Several genes, which include ctrA, porA, crgA, and 16S rRNA, have been widely used as the gene targets for N. meningitidis detection by PCR-based assays (7, 16, 33). In particular, genes such as sacB and siaD have been used to genogroup the most prevalent A, B, C, W135, and Y serogroups (3, 4, 10). However, to our knowledge, no molecular method that distinguishes all 12 N. meningitidis serogroups has ever been developed. The conventional serogrouping method that uses specific antisera is still in routine practice in many diagnostic laboratories; however, it is labor-intensive, and antisera for all serogroups are very hard to obtain. Therefore, developing a rapid, accurate, easy to use, and inexpensive method for the detection and genogrouping of N. meningitidis is in urgent need.

Variations in genes clustered in the chromosomal capsular gene complex are responsible for CPS differences in the N. meningitidis serogroups. A typical N. meningitidis capsular gene complex consists of regions A, B, C, D, and E. Regions A and C are located between the galE and tex genes on the chromosome and are transcribed independently (31). Region B is upstream of region A or downstream of region C (35). Genes in region A are responsible for the synthesis of the polysaccharide. Genes in region B, which include lipA and lipB, are involved in translocation and surface expression of capsule polymers. Region C contains four genes (ctrA, -B, -C, and -D) that are necessary for the transport of the capsule to the membrane (35). Region D is comprised of a cassette of genes, which include rmlA, -B, and -C and galE, and is not involved in capsule expression but is responsible for meningococcal lipooligosaccharide (LOS) biosynthesis (13). Region E contains only one gene, tex, that has been proposed to regulate CPS synthesis (19). The gene sequences in region A of the complex are different in individual serogroups, and those in regions B, C, D, and E are highly conserved between serogroups (32).

Out of the 12 serogroups, only the seven capsular gene complexes from serogroups A, B, C, 29E, W135, X, and Y have been characterized previously (8, 27). In this study, the capsular gene complexes of the five uncharacterized serogroups (H, I, K, L, Z) were sequenced, which made it possible to develop primers to differentiate genes specific for all 12 N. meningitidis serogroups. A multiplex PCR-based assay for N. meningitidis detection and genogrouping was developed based on “screening out” specific genes and primers for each N. meningitidis serogroup. This assay was tested on 179 bacterial isolates and proved to be specific, accurate, and sensitive.

MATERIALS AND METHODS

Bacterial strains.

The 179 strains used in this study are listed in Table 1. They include 15 reference strains and 143 clinical strains of the 12 targeted N. meningitidis serogroups (A, B, C, 29E, H, I, K, L, W135, X, Y, and Z) and 15 strains of other Neisseria species, 5 strains of Streptococcus pneumoniae, and one strain of Haemophilus influenzae. N. meningitidis strains were grown on chocolate agar plates at 37°C in the presence of 5% CO₂ in air for 24 h.

Table 1.

Bacterial strains used in this study^a

Serogroup, sample type, or species	Strain no.b (source of strain)	Total no. of strains
Neisseria meningitidis reference strain (n = 15) serogroup
A	29010 (CMCC), 29019 (CMCC)	2
B	29011 (CMCC), 29061 (CMCC)	2
C	29012 (CMCC)	1
29E	29034 (CMCC)	1
H	29031 (CMCC)	1
I	29044 (CMCC)	1
K	29047 (CMCC)	1
L	43828 (ATCC)	1
W135	29037 (CMCC), 29057 (CMCC)	2
X	M8210 (ICDC)	1
Y	29038 (CMCC)	1
Z	35562 (ATCC)	1
N. meningitidis clinical isolate (n = 143) sample type
CSF	23 (ICDC), 2 (SCDC)	25
Blood	4 (ICDC)	4
TS	61 (ICDC), 6 (SCDC)	67
CP	30 (SCDC)	30
BP	4 (SCDC)	4
Unknown	9 (ICDC), 4 (SCDC)	13
Strains from other species (n = 21)
Neisseria lactamica	3719 (CCM), 29114 (CMCC)	2
Neisseria flavescens	2827 (CCM)	1
Neisseria subflava	3482 (CCM), 4392 (CCM), 4400 (CCM), 29109 (CMCC), 29110 (CMCC), 29111 (CMCC), 29121 (CMCC)	7
Neisseria mucosa	3483 (CCM)	1
Neisseria sicca	4404 (CCM), 4405 (CCM)	2
Neisseria gonorrhoeae	No. 1 (DDTGH), no. 2 (DDTGH)	2
Streptococcus pneumoniae 6A	G1839 (NU)	1
Streptococcus pneumoniae 1	G1863 (NU)	1
Streptococcus pneumoniae 19A	G1921 (NU)	1
Streptococcus pneumoniae 23F	G1934 (NU)	1
Streptococcus pneumoniae 6B	G1940 (NU)	1
Haemophilus influenzae	58528 (CMCC)	1

^aCMCC, China's Medical Culture Center; ATCC, American Type Culture Collection; ICDC, Chinese Center for Disease Control and Prevention, Beijing, China; SCDC, Shanghai Center for Disease Control and Prevention; CCM, Czech Collection of Microorganisms; DDTGH, Dermatological Department of Tianjin General Hospital; NU, Nankai Univeristy, China; CSF, cerebrospinal fluid; CP, chocolate agar plate; BP, blood agar plate; TS, throat swab.

^bFor N. meningitidis clinical isolate sample types, the number of strains is given instead of strain number.

Genomic DNA extraction.

Genomic DNA was extracted with a DNA extraction kit (QIAamp DNA minikits; Qiagen, Hilden, Germany).

Capsular gene complex amplification, sequencing, and analysis.

Primers wl-14395 (5′-CGCCATTTCTTCCGCCAACACCA-3′) and wl-22861 (5′-CCAGCCCGAAAGTAGCCGATGC-3′) were designed based on the sequences of galE and gltS genes, respectively, and were used to amplify the capsular gene complexes of serogroups D, H, and K. The primers wl-14396 (5′-GCGTTTGCGTGCAGCATCGACT-3′) and wl-27419 (5′-GCCGAGAACCTTAACTGGCATATTGTGAA-3′) were designed based on tex and lipB gene sequences, respectively, and were used to amplify the capsular gene complexes of serogroups I, L, and Z. Long, high-fidelity PCR was carried out under the following conditions: denaturation step at 95°C for 15 s, annealing step at 62°C for 30 s, and extension step at 68°C for 10 min for 32 cycles. The PCR products were gel purified on UNIQ-10 columns (Sangon, Shanghai, China), and the sample DNA was sheared at speed code 2 (20 cycles) to the desired fragment length of 2 to 3 kb by HydroShear fragmentation (GeneMachines, CA). The fragments were then cloned into the pUC18 vector after end-blunting reaction with T₄ polymerase and Klenow fragments (New England BioLabs, MA) to produce shotgun banks.

The samples were sequenced on an ABI 3730 automated DNA sequencer (Applied Biosystems, CA), and sequencing data were assembled by use of the Staden package software (29). The Artemis program (www.sanger.ac.uk) was used to identify open reading frames (ORFs) and annotations (24). BLAST and PSI-BLAST software (http://blast.ncbi.nlm.nih.gov/Blast.cgi) were used to search several databases (1), which included GenBank (www.ncbi.nlm.nih.gov/GenBank), the Clusters of Orthologous Groups (COG; www.ncbi.nlm.nih.gov/COG/), and Pfam (pfam.sanger.ac.uk) protein motif databases (2, 34).

Primer design.

All the primers used in this study are listed in Table 2. The genes porA and ctrA were used as the gene targets for N. meningitidis species-specific primers. Sequences of N. meningitidis porA and ctrA genes were obtained from the GenBank database and aligned using ClustalX software (www.clustal.org). The universal primers were designed based on the most conserved regions of these two genes. Serogroup-specific primers were designed based on the sacD gene sequence for serogroup A, the siaD gene sequence for serogroups B and C, the synG/synF gene sequence for W135/Y, the cap29EH gene sequence for serogroup 29E, the wnmB gene sequence for serogroups I/K, the capZC gene sequence for serogroup Z, the lcbB gene sequence for serogroup L, the xcbA gene sequence for serogroup X, and the wnmA gene sequence for serogroup H. In addition, the primers used to differentiate serogroups I and K were based on the lipA gene sequence (sense primer) and galE gene sequence (antisense primer) of serogroup I. The primers that were used to differentiate serogroups W135 and Y were based on synF and synG gene sequences. At least two pairs of primers were designed for screening each serogroup.

Table 2.

Primers used in this study

Primer pair	Targeted gene	GenBank accession no.	Amplicon size (bp)	Targeted serogroup	Sequence of primer pairs (5′→ 3′)
wl-35663/38969	ctrA	GU391296	257	All	GTCGCGGTGATGTGGTTA/AATCTCTGCCTCACTGCCAT
wl-40218/40170	porA	AY319969	158	All	CTCATAGCCGCCCGTCA/GCGGTTTTGCCGGGAACTAT
wl-37416/37417	sacD	NC_003116	470	A	TTTTATTCTTAGATGTTGACGTTTT/ATGCCAGAAATGTTTAGGAGTT
wl-37410/37411	siaD	NC_003112	555	B	TTTTTAGCATATTCAGGAAAGG/TTCAATGTGGTTGACAACATCT
wl-37469/37470	siaD	NC_008767	381	C	TGTGCTAATCCCGCCTGA/AGAAAGCCGGGAATCGTT
wl-39429/39430	cap29EH	AJ576117	694	29E	TTGGCGGTTGAAACCTTAC/GCGTATCATGCTCCATTACCA
wl-39433/39434	wnmB	HQ437687	893	I/K	AATGTGTGTGCAGGAGCTTG/ATGCTTCGGTTGCCTGTC
wl-37709/37710	capZC	HQ437689	656	Z	TATTGGCCTGAGCACCG/AGGCGTACCGTCTGTAACTG
wl-37406/37407	lcbB	HQ437688	988	L	TTTGAATGTACCCTCTCCTCTG/TAACAGTCCTGATATCACTCCGTA
wl-39431/39432	xcbA	AY289931	777	X	CTCTCGTCTACCAAATGGCTAT/GTGAGGGTGACATCCGCTA
wl-36794/36795	wnmA	HQ437685	1454	H	TGATCTACCCAAGGCACATAC/CAATCGGCTTATTGAGGGT
wl-39435/39436	synG	EU038215	520	W/Y	AAGGTGAATCTTCCGAGCAG/CCATTGGAAATTTTCGCTT
wl-32526/45168	galE/lipA	HQ437687	5030	I	TATTGGTTCGCATACTGTCATT/GTCATAGACGATGGAATAGGGA
wl-37473/37475	synG	EU038216	476	W135	ACGGTATCTGATGAAATGCTG/TCATATACAACGATTGGAATATC

Specificity of primers.

Primers based on porA and ctrA gene sequences were used to amplify the DNA templates of N. meningitidis reference strains from the 12 serogroups, from Neisseria flavescens, Neisseria subflava, Neisseria mucosa, Neisseria lactamica, Neisseria sicca, Neisseria gonorrhoeae strains, and from S. pneumoniae and H. influenzae. The PCRs were carried out in a reaction mixture that contained 1× PCR buffer (50 mM KCl, 10 mM Tris-HCl [pH 8.3]), 2.5 mM MgCl₂, 0.4 mM deoxynucleoside triphosphate (dNTP), 1 U rTaq DNA polymerase, 0.1 μM each primer, and 50 to 100 ng of DNA template in a final volume of 20 μl. The PCR program used was as follows: a denaturation step at 94°C for 5 min and then 31 cycles of a denaturation step at 94°C for 30 s, an annealing step of 56°C for 45 s, and an extension step of 72°C for 1 min, with a final extension at 72°C for 5 min.

Development of the multiplex PCR-based assay.

This assay consisted of three multiplex PCRs. PCR 1 was used to detect serogroups A, B, C, 29E, and I/K, and PCR 2 was used to detect serogroups Z, L, X, H, and W135/Y. Reaction 3 was used to differentiate serogroups I and K and W135 and Y, which could not be distinguished by PCRs 1 and 2 (see Fig. S1 in the supplemental material). The PCR mixture for reactions 1, 2, and 3 contained the same reagents except for the primers, and the details are described in the supplemental material. The PCR program for reactions 1 and 2 was as follows: a denaturation step at 94°C for 5 min and then 24 cycles of a denaturation step at 94°C for 30 s, an annealing step at 59°C for 45 s, and an extension step at 72°C for 1 min, with a final extension at 72°C for 5 min. The PCR program for reaction 3 was the same as that for reactions 1 and 2 except that the annealing step was 54°C for 45 s and the extension step was 72°C for 3 min.

Nucleotide sequence accession numbers.

The DNA sequences of the capsular gene complexes of N. meningitidis serogroups H, I, K, L, and Z have been deposited in the GenBank database under the accession numbers HQ437685, HQ437686, HQ437687, HQ437688, and HQ437689, respectively.

RESULTS

Capsular gene complexes of six N. meningitidis serogroups.

The capsular gene complexes of the remaining five serogroups, H, I, K, L, and Z, were sequenced in this study. DNA sequences of 23,488, 24,676, 27,709, 18,453, and 21,871 bp in length were obtained from serogroups H, I, K, L, and Z, respectively. These gene complexes contained 19, 19, 22, 14, and 15 ORFs, respectively (see Fig. S2 in the supplemental material). The average GC contents of the sequences were 45.13, 48.04, 45.03, 45.62, and 46.97% in serogroups H, I, K, L, and Z, respectively. The functions of each ORF in these capsular gene complexes were predicted based on sequence homology by search of the listed databases and are summarized in Tables S1 to S5 in the supplemental material.

Genes specific for N. meningitidis detection and genogrouping.

ctrA or porA genes have been used individually to identify N. meningitidis (15, 16). However, we found that our primers based on the ctrA sequence produced PCR products from both N. meningitidis and N. lactamica, and those based on the porA sequence gave PCR products from both N. meningitidis and N. gonorrhoeae (data not shown). Therefore, it was necessary to use a combination of these two genes as the targets to distinguish N. meningitidis from other species. Serogroup-specific genes were chosen based on the capsular gene sequences of all 12 N. meningitidis serogroups from those genes responsible for different CPS structures. These genes included ones that encode glycosyltransferase (sacD, wnmB, lcbB), polysialyltransferase (siaD), polymerase (synF, synG), phosphotransferase (capZC, wnmA, xcbA), and d-arabinose-5-phosphate isomerase (cap29EH). However, the gene sequences responsible for CPS synthesis in serogroups I and K, and those in serogroups W135 and Y, were too similar to be useful for differentiation. Genes lipA and galE, whose location is different in serogroups I and K, were used to differentiate the two serogroups. For serogroups W135 and Y, a minor difference between synF and synG sequences was used to differentiate these two serogroups.

Primers based on the ctrA and porA gene sequences gave both of the two expected PCR products in N. meningitidis, but only one or no PCR product in other species. Primers based on serogroup-specific genes could amplify the expected PCR products in the corresponding serogroup(s) but not in other serogroups. Primers that differentiated the serogroups I and K, or serogroups W135 and Y, gave PCR products only for serogroups I and W135, respectively, and no PCR product was detected for serogroups K or Y. The positions of the serogroup-specific primers are shown in Fig. S2 and S3 in the supplemental material.

Development of a multiplex PCR-based assay.

Three groups of primer sets were made based on the screened specific primers. The primer pair efficiency in each group was determined on the basis of achieving amplicons of the expected sizes at a range of different primer concentrations. The primer concentration that resulted in high-signal products was used as described in Materials and Methods. At the optimized primer concentration ratio, the DNA of strains that belonged to serogroups A, B, C, 29E, and I/K produced the expected PCR products of distinct sizes (470 bp for serogroup A, 555 bp for B, 381 bp for C, 694 bp for 29E, 893 bp for I/K, and 257 and 158 bp for ctrA and porA of all the tested serogroups) in the group 1 multiplex PCRs (Fig. 1). The DNA from serogroups Z, L, X, H, and W135/Y produced the expected PCR products (656 bp for Z, 988 bp for L, 777 bp for X, 1,454 bp for H, 520 bp for W135/Y, and 257 and 158 bp for ctrA and porA of all the tested serogroups) in the group 2 multiplex PCRs (Fig. 1). Serogroups I and W135 gave expected PCR products of 5,030 bp and 476 bp, respectively, in the group 3 multiplex PCRs (Fig. 2).

Fig 1.

Agarose gel electrophoresis of the PCR products obtained from multiplex PCRs 1 and 2 in the assay. Lane MW, 100-bp DNA ladder marker; lane A, serogroup A; lane B, serogroup B; lane I/K, serogroup I/K; lane 29E, serogroup 29E; lane C, serogroup C; lane H, serogroup H; lane L, serogroup L; lane X, serogroup X; lane W135/Y, serogroup W135/Y; lane Z, serogroup Z.

Fig 2.

Agarose gel electrophoresis of the PCR products obtained from multiplex PCR 3. Lanes MW, DNA marker DL15,000; lane I, serogroup I; lane K, serogroup K; lane W135, serogroup W135; lane Y, serogroup Y.

Specificity of the multiplex PCR-based assay.

The developed multiplex PCR-based assay was tested with the 15 reference strains that belonged to the 12 N. meningitidis serogroups, with 97 clinical strains that belonged to A, B, C, W135, 29E, X, or Y or nongroupable isolates from different locations and years; with 15 strains of closely related species such as N. flavescens, N. subflava, N. mucosa, N. lactamica, N. sicca, and N. gonorrhoeae; and with six strains from species also reported to cause meningitis that included five S. pneumoniae isolates (belonging to serotypes 1, 23F, 6A, 6B, and 19A, reported to be more virulent) and one H. influenzae isolate (Table 1; see also Table S6 in the supplemental material). All the reference strains were detected with 100% specificity. Among the 97 clinical isolates, 90 of known serotype were detected and confirmed by the specific antiserum serogrouping. Five out of the seven strains that were classified as nongroupable by antisera could be classified as belonging to a specific serogroup by the multiplex PCR assay (see Table S6). For strains that belonged to other Neisseria species, only one or no PCR products that corresponded to segments of ctrA or porA genes could be detected. There was no PCR product obtained for S. pneumoniae and H. influenzae strains. These results indicate that the multiplex PCR-based assay is specific for individual N. meningitidis serogroups.

Double-blinded test to verify the multiplex PCR assay.

A double-blinded test was performed in order to verify the stability and specificity of the assay. Forty-six clinical isolates were used for the test (see Table S6 in the supplemental material). The isolates were also characterized by specific antisera at the Center for Disease Control and Prevention, Shanghai, China. The detection results of the PCR assay were consistent with the results by antiserum serogrouping except for the isolate NM033 (see Table S6). The reason for the different detection results for isolate NM033 using antisera and the multiplex PCR assay are unknown, and the strain needs to be further analyzed.

Detection sensitivity of the multiplex PCR assay.

The detection sensitivities for genomic DNA and isolates in noncultured mock CSF specimens were both determined. Serial dilutions of genomic DNA (10 ng, 1 ng, 100 pg, 10 pg, and 1 pg per ml) of N. meningitidis serogroups A, B, C, 29E, and Z reference strains were used to test the sensitivity of the multiplex assay. As little as 1 ng of DNA (equivalent to ∼4 × 10⁵ genomes) could be detected by the assay. Pure cultures of strain NM055 (serogroup C, isolated from a patient) were mixed with 1 ml CSF specimens from healthy people with the concentrations of 10⁷, 10⁶, 10⁵, 10⁴, to 10³ CFU/ml, respectively, and tested with the multiplex assay. The lower limit of detection was determined to be 3 × 10⁵ CFU/ml. All the reference strains of other serogroups could also be detected by this assay at the level of ∼4 × 10⁵ genomes or 3 × 10⁵ CFU/ml. Therefore, ∼4 × 10⁵ genomes or 3 × 10⁵ CFU/ml strains in the mock CSF specimen were the lower limits for the detection using this assay.

DISCUSSION

Molecular protocols for the detection of N. meningitidis have been reported previously. However, serogrouping of N. meningitidis is dependent mainly on methods that use specific antisera. The capsular gene complexes from seven out of the 12 N. meningitidis serogroups have been identified previously. The complexes of the remaining five serogroups were sequenced in this study, and these sequences enabled the design of primers for further genogrouping. As far as we know, this study is the first to describe a molecular protocol for the detection and genogrouping of all 12 N. meningitidis serogroups.

In the multiplex PCR-based assay developed in this study, primers for ctrA and porA genes were combined to be used for N. meningitidis identification. Genes that encode glycosyltransferase, polysialyltransferase, polymerase, phosphotransferase, and d-arabinose-5-phosphate isomerase in the capsular gene complexes were used as the serogroup-specific genes, as these genes were responsible for differences in the CPSs between the serogroups. For serogroups W135 and Y, it was reported that the amino acid 310 in synF and synG was responsible for the different capsular forms of the two serogroups (5). A primer pair based on minor divergence between synF and synG sequences was used to differentiate serogroups W135 and Y. The antisense primer covered three consistently different positions (L324V, A325V, N327D) between the two genes, and the fragment amplified contained six of the nine consistently different positions (including G310P) of the two genes.

A total of 179 strains, which included the N. meningitidis reference strains, clinical isolates from different locations and years, those of closely related species, and those of species that can be detected in blood or CSF causing meningitis, were used to characterize the specificity of the assay. The results corresponded well with the findings by the traditional serology-based method. Some isolates previously classified as nongroupable were also able to be genogrouped by our assay. The total time to perform this assay was only 2 to 4 h, following bacterial isolation, and a PCR block is the only major equipment necessary. The accurate and efficient detection of a pathogenic microorganism is crucial for the prevention and effective treatment of disease, especially in the case of pathogen outbreaks. Therefore, this multiplex PCR-based assay has the advantage for applications in clinical diagnosis and epidemiological surveillance.

Although the detection sensitivity (4 × 10⁵ genomes or 3 × 10⁵ CFU/ml) is close to the level of N. meningitidis in most clinical CSF specimens (10⁵ CFU/ml) (17), a better sensitivity will be helpful for the assay to be applied. It was reported that real-time PCR assays for N. meningitidis had a sensitivity less than 100 genomes (250 fq) (9, 18); therefore, real-time PCR assays for the detection and genogrouping of all N. meningitidis serogroups will be more applicable and can be developed in future studies.