3208 Natalie Damaso1,2,3 Lauren Martin1,2,3 Priyanka Kushwaha3 DeEtta Mills1,2 1 Department
of Biological Sciences, Florida International University, Miami, FL, USA 2 International Forensic Research Institute, Florida International University, Miami, FL, USA 3 Department of Chemistry and Biochemistry, Florida International University, Miami, FL, USA
Received February 10, 2014 Revised August 21, 2014 Accepted August 22, 2014
Electrophoresis 2014, 35, 3208–3211
Short Communication
F-108 polymer and capillary electrophoresis easily resolves complex environmental DNA mixtures and SNPs Ecological studies of microbial communities often use profiling methods but the true community diversity can be underestimated in methods that separate amplicons based on sequence length using performance optimized polymer 4. Taxonomically, unrelated organisms can produce the same length amplicon even though the amplicons have different sequences. F-108 polymer has previously been shown to resolve same length amplicons by sequence polymorphisms. In this study, we showed F-108 polymer, using the ABI Prism 310 Genetic Analyzer and CE, resolved four bacteria that produced the same length amplicon for the 16S rRNA domain V3 but have variable nucleotide content. Second, a microbial mat community profile was resolved and supported by NextGen sequencing where the number of peaks in the F-108 profile was in concordance with the confirmed species numbers in the mat. Third, equine DNA was analyzed for SNPs. The F-108 polymer was able to distinguish heterozygous and homozygous individuals for the melanocortin 1 receptor coat color gene. The method proved to be rapid, inexpensive, reproducible, and uses common CE instruments. The potential for F-108 to resolve DNA mixtures or SNPs can be applied to various sample types—from SNPs to forensic mixtures to ecological communities. Keywords: Capillary electrophoresis / F-108 polymer / Microbial community / SNP / SSCP DOI 10.1002/elps.201400069
Environmental microbial community mixtures are difficult to analyze and current molecular profiling methods do not always support sequence queries of differences within or between samples. For community DNA analyses, the current practices for differentiating mixtures require either traditional cloning and sequencing, or NextGen sequencing, both of which can be expensive and time consuming [1, 2]. Therefore, rapid DNA profiling techniques such as terminal RFLP [3], denaturing gradient gel electrophoresis [4], and amplicon length heterogeneity PCR methods [5] using denaturing polymers or conditions are often used to query DNA community mixtures. Amplicon separation using denaturing conditions and polymers such as performance optimized polymer 4 (POP-4), can underestimate the true sequence diversity hidden within each amplicon. This is because taxonomically unrelated organisms can produce a length ampli-
Correspondence: Dr. DeEtta Mills, Department of Biological Sciences, Florida International University, 11200 SW 8th Street, OE 167, Miami, FL 33199, USA E-mail:
[email protected] Fax: +1-305-348-1986
Abbreviations: F-108, Pluronic F-108 (tri-block copolymer); mc1r, melanocortin 1 receptor; POP-4, performance optimized polymer 4; POP-CAP, performance optimized polymer conformation analysis polymer
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
con that is not representative of the varied base sequence contained within the amplicon [6, 7]. SSCP is a technique that has been highly applicable in community profiling [8], but can often be difficult to optimize for CE. This technique has commonly been used to detect unknown mutations or SNPs in short amplicons. During separation ssDNA fragments partially renature and form folded conformations due to intramolecular interactions between the bases under nondenaturing gel electrophoresis. Different electrophoretic mobility based on the secondary DNA structures can result in separation of DNA strands with a resolution of one base [9]. A polymer that has recently been used as a CE-SSCP sieving matrix to enhance resolution is the F-108 triblock copolymer, poly(ethyleneoxide)-poly(propyleneoxide)poly(ethyleneoxide). The PPO forms hydrophobic micelles that do not directly interact with DNA, but induce micelle formation resulting in a high-density sieving matrix that will separate DNA amplicons with different base content. Previous study by Shin et al. 2010 discovered that the PEO chain length increased the resolution due to the hydrophilic mesh that interacts with DNA and, together, it results in a highdensity sieving matrix [10]. The objectives of this research were as follows: First, to validate the F-108 polymer for the ABI CE using four model organisms that display the same length amplicon for 16S rRNA hypervariable domains under denaturing conditions but have varying base composition; second, was to assess the
www.electrophoresis-journal.com
Electrophoresis 2014, 35, 3208–3211
polymer for resolving more complex mixtures from an environmental source using a hot spring microbial mat sample; and third, to assess the polymer for resolving SNPs, within particular genes. The melanocortin 1 receptor (mc1r) gene, which codes for the production of the black pigmentation (eumelanin) was used [11]. Purified DNA from four bacteria species, Salmonella enterica, Escherichia coli, Pseudomonas aeruginosa, and Staphylococcus aureus were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and diluted to a 10 ng working stock. All four model organisms produced the same length amplicon within the 16S rRNA V3 domain, but have varying nucleotide sequences. A cyanobacteria-dominated microbial mat from Hunter’s Hot Spring in Lakeview, Oregon was collected and stored R solution (Life Technologies, Grand Island, in RNALater NY, USA) at 4°C until processed. Extraction of the comR SPIN Kit for Soil munity DNA was done using FastDNA (MP BioMedical, Solon, OH, USA) as per manufacturer’s directions. To test the mc1r SNPs, equine hair samples were collected noninvasively and DNA was extracted from 10 to 20 mane hairs with intact roots using the Qiagen QIAmp DNA Micro kit (Qiagen, Valencia, CA, USA) as per manufacturer’s directions. The hypervariable domain V3 of the 16S rRNA gene was amplified using bacteria-specific 338F (5 ACTCCTACGGGAGGCAGCAG-3 ) and the reverse primer 518R (5 -ATTACCGCGGCTGCTGG-3 ). The forward primer was labeled with the fluorochrome, HEX, and HPLC purified (Integrated DNA Technologies, Skokie, IL, USA). PCR reactions were performed in a 20 L reaction as previously described [12]. The PCR protocol for the isolates was modified for the F-108 polymer by using 5 ng DNA and 0.2 M of each primer and only 30 s intervals for the denaturation, annealing, and extension time. Hunter’s Hot Spring community samples followed the same protocols used for the bacteria isolates but with one modification for the annealing temperature (54°C). The forward primer MC1RF (5 -CCTACCTCGGGCTGACCACCAA-3 ) and the reverse primer MC1R-R (5 -GAGAGGACACTAACCACCCAGATG3 ) were used for the mc1r gene as previously described [13, 14]. The forward primer was labeled with the HEX fluorochrome (Integrated DNA Technologies). PCR products were prepared for each polymer by mixing 1 L of the PCR product from each sample with Hi-DiTM Formamide (Life Technologies, Carlsbad, CA, USA) (11.5 L for POP-4;
CE and CEC
3209
13.5 L for F-108) and 0.5 L GeneScan ROXTM 500 size standard, heated at 95°C for 5 min, and snapped cooled on ice for 2 min. Each bacterial isolate’s PCR product was loaded individually and then as a mixture by combining all four products post-PCR. A total of 90 samples were tested for reproducibility. Next, the microbial mat and equine SNP samples provided an assessment of limitations and resolution of the F-108 polymer. DNA separation was performed on an ABI Prism 310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) using a 50 m internal diameter, 40 cm distance from inlet to detection window, noncoated capillary (Polymicro Technologies, Phoenix, AZ, USA). Pluronic F-108 (cat #542342, Sigma−Aldrich, St. Louis, MO, USA) was prepared as previously described [15]. POP-4 separation parameters were specified by the manufacturer. As expected, the POP-4 polymer separated the four PCR products based on length and displayed the same length peak (192 bp) for each bacterium. F108 separation conditions were taken from the literature [15] with some minor modifications: injection voltage of 15 kV for 5 s, 15 kV electrophoresis voltage, 40 min run time, and running temperature of 38°C. Since conditions for each CE instrument can vary slightly, these modifications were found to be optimal under our laboratory conditions. When the same PCR products were separated using the F-108 polymer, four separate peaks were produced, representing the four bacteria (Fig. 1). Next, an environmental mixture from the hot spring microbial mat was separated using both the POP-4 and F-108 polymer using the same parameters that had previously been optimized for each polymer. The POP-4 separation had a profile with approximately six peaks above the level of background fluorescence (Fig. 2a). F-108 profile showed a community with approximately 15 peaks (Fig. 2b) that were detected above the baseline of 50 relative fluorescent units set for all analyses. When the same mat community was sequenced using NextGen sequencing of the 16S ribosomal DNA, it confirmed that the community consisted of 15 bacterial species (unpublished data, available upon request), correlating to the number of peaks resolved by F-108. To test the resolution of the F-108 polymer for SNPs, amplicons with known SNPs were assessed. POP-4 confirmed a single peak at 456 base pairs [13], but when resolved with F-108, it detected the heterozygote SNPs and produced two separate peaks (Fig. 3). The equine samples tested had previously been verified using the commercial SNapShotۚ kit (Life
Figure 1. Separation using F-108 polymer for four bacteria (S. enterica, E. coli, P. aeruginosa, S. aureus) mixed post-PCR. Bacteria peaks, shown as filled peaks, were assigned mobility values relative to the internal standard, ROX 500, shown as open peaks. y-Axis, relative fluorescence units; x-axis, data points. All species produced a 192 bp amplicon using the POP-4 polymer.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
3210
N. Damaso et al.
Figure 2. (a) Separation using POP-4 polymer for the hot spring microbial mat produced six peaks. (b) Separation using F-108 polymer for the microbial mat produced 15 peaks. NextGen results were represented by mostly uncultured microbes: 20% uncultured cyanobacteria, 40% uncultured environmental bacteria, and the rest listed as (one representative of each) having their closest relatives to be Planktothricoides sp., Roseiflexus sp., uncultured Bacteroidetes sp., uncultured, CFB group bacterium, uncultured Chlorobi sp., and uncultured Chloroflexus sp. Minimum threshold (50 relative fluorescent units) represented as the horizontal line; y-axis, relative fluorescence units; x-axis, data points.
Technologies) customized for the mc1r gene and other coat color markers. All the heterozygous and homozygous individuals for the mc1r gene were correctly identified. The coat color was also known from each of the donor horses, further verifying the results. Previous published research by the Shin group was successfully verified using the four model organisms in this study. Their study also compared the F-108 polymer with POP-conformation analysis polymer (POP-CAP) to distinguish 12 pathogenic bacterial isolates that were mixed postPCR using this same CE-SSCP method. While F-108 polymer was able to distinguish all 12 isolates, POP-CAP produced only six peaks [15]. Once the method was reproduced for this study, the question became: Can it separate complex mixtures of unknown templates from the environment? Previous studies that used CE-SSCP with commercial polymers, such as POP-CAP and GeneScan, were not able to get this degree of resolution and none have used it for microbial community analysis. King et al. 2005 looked at different universal primers for amplifying the 16S rDNA variable domains. These amplicons were separated using the commercial native GeneScan polymer to find the best primer set that showed the greatest resolution. Their study mixed 12 pure culture DNAs but none of the primer sets could separate all 12 bacteria because several peaks would coelute or resulted in distorted peak shapes. They did conclude that the best primer set was the V3 domain, which produced 11 peaks with P. aeruginosa coeluting with other species [9]. In the current study, the V3 domain of the 16S rRNA was also queried and found to be optimal for F-108 CE-SSCP analysis and was able to distinguish all 15 bacteria from the mat community that were correlated to the NextGen sequencing results. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Electrophoresis 2014, 35, 3208–3211
Figure 3. Separation using F-108 polymer for (a) horse 1 (heterozygote), (b) horse 2 (homozygote), and (c) horse 3 (heterozygote). y-Axis, relative fluorescence units; x-axis, data points. All horses produced a 456 bp amplicon using the POP-4 polymer.
Finally, the F-108 polymer was able to resolve the equine SNPs that differed by one nucleotide substitution within the mc1r gene. Usually, sensitivity of SNP detection decreases with increasing fragment length. Therefore, one of the limitations for SSCP has been that short sequences (⬍400 bp) are easily resolved but larger fragments are resolved at a reduced rate [9, 16, 17]. However, there have been studies, such as research conducted by Choi et al. 2014, that enhanced the sensitivity of CE-SSCP. F-108 polymer was used to detect SNPs by the incorporation of reaction condition-optimized ligation probes to amplify the region of the tp53 gene, encoding for the p53 protein [18]. The present study differs from the Choi study, as it was able to detect the SNPs simply by using a standard PCR product separated with the F-108 polymer. The F-108 polymer was able to resolve the equine SNPs within amplicons that were over 400 bp in length and still distinguished the one base polymorphism without the additional step of designing a probe assay for the amplicon. Utilization of the F-108 polymer and CE can provide greater resolution of mixtures based on sequence differences within an amplicon. This could have a significant impact on the forensic community by using this polymer for identification of biothreat agents—critical to homeland security. Moreover, this study showed that the F-108 could rapidly profile SNPs in equine coat color genes [19, 20]. Assessing more coat color SNPs could provide a phenotypic description that can be used in forensic investigations for equine-associated crimes. For example, in the past few years, horses in MiamiDade and Broward counties, FL, have been illegally and www.electrophoresis-journal.com
Electrophoresis 2014, 35, 3208–3211
inhumanly slaughtered and the meat was sold in the black market. Being able to rapidly profile coat color from seized horsemeat will now provide additional phenotypic information about the slaughtered horse and supplement the standard equine DNA typing to help identify the stolen horse and increase the match probability in these cases. The model systems tested in this study assessed the resolution of the F-108 nondenaturing polymer and validated its utility for microbial community profiling and genotyping SNPs. To our knowledge, this is the first account of using CE-SSCP method with F-108 polymer for a microbial community analysis from an environmental sample as well as SNP analysis for coat color genes in horses. Multiplexing the coat color SNP markers and using F-108 for detection is ongoing in our laboratory. Further analyses are needed to see if F-108 CE-SSCP could enhance the resolution for SNP detection (i.e., the C to G transversion found in DYS392; http://www.cstl.nist.gov/strbase/STRseq.htm) for human DNA forensic analyses so DNA sequencing would not be required. This present study using F-108 polymer proved to be rapid, reproducible, and was able to enhance sequence resolution for several applications. This polymer could easily be integrated into CE separations in any laboratory—for forensic-, ecological-, or medical-based assessments of sequence differences. We would like to acknowledge the Forensic DNA Profiling Facility, International Forensic Research Institute, Florida International University for their instrumental support and Beatrice Kallifatidis, MSFS, for her support and guidance. ND thanks the McNair Graduate Fellowship for their support in funding. This work was funded in part by NSF CBET-0853746 to DM, NSF-OCE #1208784 to DM, and Sigma Xi Grants-inAids #G20130315163604 to ND. We would like to acknowledge Dr. Linda McGown, Rensselaer Polytechnic Institute, for her training in separation science to ND. The authors have declared no conflict of interest.
References [1] Rastogi, G., Sani, R. K., in: Ahmad, I., Ahmad, F., Pichtel, J. (Eds.), Microbes and Microbial Technology, Springer, New York 2011, pp. 29–57.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
CE and CEC
3211
[2] Rincon-Florez, V. A., Carvalhais, L. C., Schenk, P. M., Diversity 2013, 5, 581–612. [3] Liu, W. T., Marsh, T. L., Cheng, H., Forney, L. J., Appl. Environ. Microbiol. 1997, 63, 4516–4522. [4] Muyzer, G., de Waal, E. C., Uitterlinden, A. G., Appl. Environ. Microbiol. 1993, 59, 695–700. [5] Litchfield, C., Gillevet, P., J. Ind. Microbiol. Biotechnol. 2002, 28, 48–55. [6] Hou, A., Williams, H. N., in: Anderson, J. T., Davis, C. A. (Eds.), Wetland Techniques, Springer, Netherlands 2013, pp. 59–92. [7] Neilson, J. W., Jordan, F. L., Maier, R. M., J. Microbiol. Methods 2013, 92, 256–263. [8] Hiibel, S. R., Pruden, A., Crimi, B., Reardon, K. F., J. Microbiol. Methods 2010, 83, 286–290. [9] King, S., McCord, B., Riefler, G., J. Microbiol. Methods 2005, 60, 83–92. [10] Shin, G. W., Hwang, H. S., Seo, S. W., Oh, M., Ryu, C. Y., Salvo, C. J., Feldman, S., Doh, J., Jung, G. Y., J. Sep. Sci. 2010, 33, 1639–1643. [11] Jackson, I. J., Annu. Rev. Genet. 1994, 28, 189–217. [12] Moreno, L. I., Mills, D., Fetscher, J., John-Williams, K., Meadows-Jantz, L., McCord, B., J. Microbiol. Methods 2011, 84, 388–393. [13] Kakoi, H., Tozaki, T., Nagata, S., Gawahara, H., KirimaSuda, I., J. Anim. Breed. Genet. 2009, 126, 425– 431. ´ [14] Rieder, S., Taourit, S., Mariat, D., Langlois, B., Guerin, G., Mamm. Genome 2001, 12, 450–4550. [15] Shin, G. W., Hwang, H. S., Oh, M., Doh, J., Jung, G. Y., Electrophoresis 2010, 31, 2405–2410. [16] Ren, J., J. Chromatogr. B Biomed. Sci. Appl 2000, 741, 115–128. [17] Kourkine, I. V., Hestekin, C. N., Barron, A. E., Electrophoresis 2002, 23, 1375–1385. [18] Choi, W., Shin, G. W., Hwang, H. S., Pack, S. P., Jung, G. Y., Jung, G. Y., Electrophoresis 2014, 35, 1196–1203. [19] Ranade, K., Chang, M., Ting, C., Pei, D., Hsiao, C., Olivier, M., Pesich, R., Hebert, J., Chen, Y. D., Dzau, V. J., Curb, D., Olshen, R., Risch, N., Cox, D. R., Botstein, D., Genome Res. 2001, 11, 1262–1268. [20] Royo, L. J., Fernandez, I., Azor, P. J., Alvarez, I., Perez-Pardal, L., Goyache, F., J. Anim. Sci. 2008, 86, 1291–1295.
www.electrophoresis-journal.com
