DNA Microarray Characterization of Pathogens Associated with Sexually Transmitted Diseases

This study established a multiplex PCR-based microarray to detect simultaneously a diverse panel of 17 sexually transmitted diseases (STDs)-associated pathogens including Neisseria gonorrhoeae, Chlamydia trachomatis, Mycoplasma genitalium, Mycoplasma hominis, Ureaplasma, Herpes simplex virus (HSV) types 1 and 2, and Human papillomavirus (HPV) types 6, 11, 16, 18, 31, 33, 35, 39, 54 and 58. The target genes are 16S rRNA gene for N. gonorrhoeae, M. genitalium, M. hominism, and Ureaplasma, the major outer membrane protein gene (ompA) for C. trachomatis, the glycoprotein B gene (gB) for HSV; and the L1 gene for HPV. A total of 34 probes were selected for the microarray including 31 specific probes, one as positive control, one as negative control, and one as positional control probe for printing reference. The microarray is specific as the commensal and pathogenic microbes (and closely related organisms) in the genitourinary tract did not cross-react with the microarray probes. The microarray is 10 times more sensitive than that of the multiplex PCR. Among the 158 suspected HPV specimens examined, the microarray showed that 49 samples contained HPV, 21 samples contained Ureaplasma, 15 contained M. hominis, four contained C. trachomatis, and one contained N. gonorrhoeae. This work reports the development of the first high through-put detection system that identifies common pathogens associated with STDs from clinical samples, and paves the way for establishing a time-saving, accurate and high-throughput diagnostic tool for STDs.


Introduction
The major causative agents of sexually transmitted diseases (STDs) are Neisseria gonorrhoeae, Chlamydia trachomatis, Mycoplasma genitalium, Mycoplasma hominis, Ureaplasma, Herpes simplex virus (HSV), Human papillomavirus (HPV), Human immunodeficiency virus (HIV), Treponema pallidum and Trichomoniasis vaginalis [1]. STDs are among the most common human infectious diseases and pose a major public health concern globally. In 2008, there were an estimated 110 million prevalent STIs (sexually transmitted infection) among women and men in the United States. Of these, more than 20% of infections (22.1 million) were young people aged 15 to 24 years [2]. From 2004From -2007 in China,the reported incidence of gonorrhea was declining, whereas that of syphilis,AIDS and the infection rates of HIV increased [3],During 2008-2012, the morbidity of STDs in Tianjin, P. R. China, showed an decreasing trend. The number of reported STD cases steadily decreases from 7,010 (incidence 62.87 per 10,000) in 2008 to 4,895 (incidence 36.14 per 10,000) in 2012 [4].
One of the most troublesome aspects of STD treatments is that STDs appear to be asymptomatic during the early infection stages, and the diseases resulting from different infectious agents often appear in similar symptom making clinical diagnoses difficult and unreliable in many cases [5]. Traditionally, microscopic examination and culturing of the pathogens are considered as the "gold standards" for the identification of the pathogens associated with STDs, and these procedures have been tailored to identify specific organisms, e.g., N. gonorrhoeae [6,7], C. trachomatis [7], mycoplasmas, ureaplasmas [8] and HSV [9]. However, the culture process takes three to seven days and sometimes even weeks to complete [10][11][12]. In addition, HPV and T. pallidum are difficult to be cultured [13,14].
Other strategies used in the detection of STD-associated pathogens include immunological assays and DNA amplification or hybridization [15,16]. Serological assays such as enzyme immunoassays (EIA) and/or direct immunofluorescence assays (DFA) provide rapid results and eliminate the need for cell culture-based assays, but these approaches sometimes do not rule out cross-reactive epitopes or distinguish between ongoing or past infections since antibody also responses to antigens that resulted from past infections [17]. Nucleic acid amplification is a very sensitive detection strategy, and can theoretically be used for the detection of as little as a few gene copies due to its high sensitivity. The commercial AMPLICOR CT/NG Amplification Kit licensed by Food and Drug Administration (FDA) of USA was used clinically to diagnose C. trachomatis and N. gonorrhoeae infections [18]. In China, PCR-fluorescence diagnostic kits have been used and certificated by the SFDA (State Food and Drug Administration, P. R. China, www.sfda.gov.cn) for STD diagnosis. However, each of these products is limited to the detection of only one or two pathogen(s) at one time. Although several systematic studies of microarray [19][20][21] have been conducted on the pathogens responsible for STDs, none of the studies is able to examine the most common clinical pathogens N. gonorrhoeae, C. trachomatis, Ureaplasma, M. hominis, M.genitalium, HSV and HPV simultaneously.
In this study, we introduce a multiplex, PCR-based oligonucleotide microarray to detect not only N. gonorrhoeae, C. trachomatis, M. hominis, M. genitalium, Ureaplasma, HSV, and HPV simultaneously, but also different serotypes of HSV types 1 and 2, and HPV types 6,11,16,18,31,33,35,39, 54 and 58. The microarray can be used as a high throughput tool for screening sexually transmitted diseases.

Bacterial strains and clinical specimen collection
The 92 strains used for microarray specificity testing included 64 type strains and 28 clinical isolates (Table 1). Total five groups of 344 clinical specimens, including cervical and urethral swab specimens, genital warts, ulcers, and blister specimens, were obtained from three domestic hospitals of Tianjin, General Hospital of Tianjin Medical University, Tianjin Union Medical Center, and Changzheng Hospital in Tianjin, P. R. China. The clinical samples were kept in

Primer design
The 16S rRNA gene was used as the target gene for the detection of N. gonorrhoeae, M. genitalium, M. hominism, Ureaplasma; the major outer membrane protein gene (ompA) for the detection of C. trachomatis; the glycoprotein B gene (gB) for HSV; and the L1 gene for HPV. The sequences of the target genes selected were obtained from GenBank. Related sequences were aligned using the CLUSTALX software. Based on the variable regions in the generated alignments, five compatible primer pairs for the multiplex PCR were designed using Primer Premier 5.0 software ( Table 2). One primer pair for the amplification of the human β-globin gene [22] was selected as the positive control.

Cloning
The target gene fragments of the selected organisms associated with respective STDs were amplified and cloned into the pGEM-T Easy vector (Promega, MA, USA), and transformed into competent E. coli DH5α cells. The clones were used to optimize the conditions for multiplex PCR to evaluate the sensitivity of the multiplex PCR and microarray assay and to screen the specificity of the probes.

Target genes and optimization of multiplex PCR
Specific primer pairs for amplification of target genes were determined, including wl-5744 and wl-5839 for 16S rRNA sequence of N. gonorrhoeae, wl-5967 and wl-5968 for 16S rRNA sequence of M. genitalium, M. hominis, and Ureaplasma, wl-5963 and wl-9516 for ompA gene of C. trachomati, wl-9940 and wl-10219 for gB gene of HSV, wl-10247 and wl-10248 for L1 gene of HPV, and wl-9502 and wl-9503 for the positive control β-globin gene, respectively.

Two-step multiplex PCR and target gene labeling
The first PCR amplification step was carried out in a 25 μl reaction volume containing 1x PCR Buffer (70 mM KCl, 14 mM Tris-HCl, pH 8.3), 1.5 U Taq DNA polymerase (Sangon Corporation, Shanghai, China), 2.5 mM MgCl 2 , 400 μM each dNTPs, and primers at different concentrations ( Table 2). The amplification was started at 80°C for 10 min then denatured at 95°C for 3 min followed by 35 cycles at 94°C for 45 s, 50°C for 2 min and 68°C for 1 min and a final step at 72°C for 10 min. An aliquot of 3 μl of PCR product was run on an agarose gel to confirm the size of the amplified DNA. The remaining reaction mixture was stored at -20°C.
For the second PCR step, the same PCR mixture was used except that 0.15 nM cyanine dye Cy3 (or Cy5)-dUTP (Amersham Biosciences UK Ltd., Little Chalfont, England) was included, the reverse primers were applied, and 3 μl of the PCR products from the first step were used as the template. The PCR conditions were the same as those described for the first PCR step. All labeled DNA samples were stored at -20°C in the dark.

Oligonucleotide probe design
For each group or biotype of pathogens tested, 1-3 probes were designed using the OligoArray 2.0 based on GenBank sequences. One probe based on the human β-globin gene was designed as the positive control and one negative control probe and one positional reference and printing control probe were also used as described by Li et al., 2006 [23]. Each probe was 5'-amino-modified with a 16 poly(T)s tail followed by a stretch of specific sequence (synthesized by AuGCT Biotechnology Corporation, Beijing, China). All oligonucleotide probes are listed in Table 3.

DNA array preparation
The probes were dissolved in 50% dimethyl sulfoxide (DMSO) to a final concentration of 1 μg/ μl and printed onto aldehyde group-modified glass slides (CEL Corporation, USA) using a SpotArray72 (Perkin-Elmer Corporation, CA, USA). Each probe was spotted in triplicate and each slide consisted of six microarrays framed with a 12 μl Geneframe (Beijing Capital Biochip Corporation, Beijing, China) which constituted individual reaction chambers. A schematic diagram of the probe positions of the single microarray is shown in Fig 1.

Hybridization process
The hybridization was performed as follows: a volume of 8 μl of labeled target single-stranded PCR products were mixed with 8 μl of hybridization buffer {25% formamide, 0.1% sodium

Data acquisition and analysis
Slides were scanned with laser beam at 532 nm for Cy3 (or 635 nm for Cy5) nm using the 4100A biochip scanner (Axon Corporation, CA, USA). A positive detection result was reported when all the probes of the given target gene generated hybridization signals above the signalto-noise-ratio threshold (3.0).

Nucleotide sequence and microarray accession numbers
The microarray dataset was deposited into the Gene Expression Omnibus database under the accession number GSE69508.

Amplicons of target genes
The size of the amplicons, ranged from 260-733 base pairs (bp) in length, of the pathogens are as follows: C. trachomati, 733 bp; N. gonorrhoeae, 648 bp; M. genitalium, M. hominis, and Ureaplasma, 520 bp; HSV, 407 bp; HPV, 260 bp; and β-globin, 326 bp. The results indicated that the primers were specific for the respective target genes and compatible in the multiplex PCR reaction.

Probe screening and specificity
For each sample, at least three hybridization reactions were replicated to demonstrate the reproducibility of the microarray method. The DNA microarray was tested using 24 strains of the target species and 68 strains of commensal microbes that are either common to the genitourinary tract or genetically close related to the target species (Table 1). From 85 oligonucleotide probes initially tried in the microarray, 34 probes were selected as suitable for the microarray including 31 probes for specific genes, one as positive control probe, one as negative control probe, and one as positional control probe for printing reference ( Table 3). The hybridization results for different pathogens are shown in Fig 2. The microarray specifically identified the 17 target strains. N. gonorrhoeae produced positive signals with its specific probes of OA-2331 and OA-2332, as well as the positive control probe OA-2340 and the positional and printing control probe Cy3 but not with the other probes (Fig 2a). Similarly, Ureaplasma produced positive signals with its specific probes of OA-2062 (Fig 2b); M. hominis, with OA-2055 and OA-2056 (Fig 2c); C. trachomatis, with OA-2102 and OA-2101 (Fig 2d); HSV type 1, with OA-2334 and OA-2335 (Fig 2e); HSV type 2, with OA-2338 and OA-2339 (Fig 2f); HPV type 6, with OA-2312 and OA-2313 (Fig 2g); HPV type 11, with OA-2314 and OA-2315 (Fig 2h); HPV type 16, with OA-2316 and OA-2317 (Fig 2i); HPV type 18, with OA-2318 and OA-2319 (Fig 2j); HPV type 31; with OA-2322 and OA-2323 (Fig 2k); HPV type 35, with OA-2327 (Fig 2l); HPV type 39, with OA-2655 and OA-2659 ( Fig  2m); HPV type 54, with OA2330 (Fig 2n); HPV type 58, with OA-2321 (Fig 2o); M. genitalium, with OA-2051, OA-2052, and OA-2053 (Fig 2p); and HPV type 33, with OA-2324 and OA-2325 (Fig 2q).
Multiplex PCR assays carried out with DNA from unrelated organisms (Table 1) did not amplify products except when Escherichia coli, Enterococcus faecalis, Enterococcus faecium, Klebsiella pneumoniae, Citrobacter freundii, and Mycoplasama pneumoniae were used as templates, however, none of these products hybridized to the microarray probes. None of the strains closely related to the target species, except M. pneumoniae which shares 98% homology of the 16S rRNA sequence with M. genitalium, hybridized to the specific probes on the microarray. It was reported that M. pneumoniae are rarely associated with genitourinary infections [8]. The degenerate base code is as follows: W = A or T, Y = C or T, R = A or G, S = C or G, K = G or T, M = A or C. doi:10.1371/journal.pone.0133927.t003 Detection sensitivity Ten-fold serial dilutions, ranging from 10 0 −10 7 copies of target DNA obtained from N. gonorrhoeae, C. trachomatis, Ureaplasma, HSV type 2 and HPV type 11 in addition to the human βglobin gene, were used to test the sensitivity of the multiplex PCR and the microarray. The sensitivity of the multiplex PCR was at 10 3 -10 4 copies and that of the microarray at 10 2 -10 3 copies for C. trachomatis. The microarray is 10 times more sensitive than that of the multiplex PCR (Fig 3a and 3b).

Discussion
Primers and probes specific to STD pathogens were designed for the microarray. The N. gonorrhoeae-specific primers were based on the 16S rRNA gene sequence, which is a more specific than the one described by Farrell [24], since the primers in this study did not generate any amplicons from closely related N. gonorrhoeae strains. The primers used for the detection of HPV, Mycoplasma and Ureaplasma were also improved for specificity from those described by Jenkins and Yoshida [8,25]. The primers designed for the detection of C. trachomatis and HSV were designed from specific sequences encoding the major outer membrane protein and glycoprotein B, respectively [6,26,27]. The specificity of this microarray was further enhanced by the pathogen-specific probes included in the chip. The most difficult problem associated with multiplex PCR was the uneven amplification of targeting products, leading to poor sensitivity. In order to overcome the problem, the multiplex PCR parameters, including the primer concentrations, the PCR buffer, and the annealing time and temperatures, were optimized to ensure the sensitivity of optimized multiplex PCR equals to that of conventional PCR. Following the optimized multiplex PCR analysis, single-strand primer extension was employed to fluorescently label respective products. This two-step method improved the hybridization efficiency and reliability.   (1) 10 7 copies, (2) 10 6 copies, (3) 10 5 copies, (4) 10 4 copies (5) 10 3 copies, (6) 10 2 copies, and (7) 10 1 copies (b)Detection sensitivity of C. trachomatis with the microarray. (1) 10 6 copies, (2) 10 5 copies, (3) 10 4 copies (4) 10 3 copies, (5) 10 2 copies, and (6) 10 1 copies. It is important to employ an internal control to verify DNA extraction efficiency from the samples and performance of all of the components in the reaction [28]. In our system, amplification of the human β-globin gene verified the efficiency of the DNA extraction and hybridization. Inability to amplify the human β-globin gene product or to detect the corresponding fluorescence signal indicates that the DNA concentration in the sample was too low or that the inhibitors were present in the reactions.
The microarray is able to detect the presence of STDs related 17 pathogens, N. gonorrhoeae, C. trachomatis, M. hominis, M. genitalium, Ureaplasma, HSV types 1 and 2, and HPV types 6, 11, 16, 18, 31, 33, 35, 39, 54 and 58 simultaneously within 7.5 hours from DNA preparation. As the data showed in Fig 2, the assay has differentiated successfully 17 pathogens based on the specificity of the probes applied with 10 times better resolution than that of the multiplex PCR. Compared to the traditional methods, the high throughput microarray assay has a number of advantages, First, it is particularly useful for the diagnosis of multiple infections simultaneously; second, it could be used to detect not only bacterial pathogens, but Chlamydia, Mycoplasma, Ureaplasma, and viruses (HSV and HPV) as well; trd, it is able to inspect the most prevalence of serogroups of HPV in the samples; and fourth,it is very unlikely to get false positive results, as this technology requires two specific steps, first, PCR reaction using specific primers, and second, hybridization applying specific probes. The methodology underlying in this microarray provides a general mean to detect STD pathogens and could be applied in other diagnosis.
This study offers a high throughput, low cost diagnosis tool, alternative to conventional, time-consuming, labor-intensive methods, to detect STDs related pathogens quickly and reliably. The major limitation of the study is that we need to expand the probes to include all the STDs pathogens, adding HIV and Treponema pallidum. In conclusion, the development of a new multiplex PCR-based microarray assay for a high-throughput platform for the detection of multiple infectious agents from multiple samples simultaneously is presented. These data suggested that the microarray analysis developed is a reliable, sensitive, and specific approach for the diagnosis of STD-associated pathogens.