Comparison of six commercial kits to extract bacterial chromosome and plasmid DNA for MiSeq sequencing

Abstract

We compared commercial kits for extraction of genomic DNA from the Gram-negative bacterium Klebsiella pneumoniae for subsequent Miseq sequencing. Purification of DNA was based on matrix binding (silica or anion exchange resin) or differential precipitation (salting out), respectively. The choice of extraction kit had little effect on sequencing quality and coverage across drastically different replicons, except for an apparent depletion of small plasmids (<5 kb) during precipitation-based extractions. Sequencing coverage provided copy-number estimates for small plasmids that were consistently higher than those from quantitative real-time PCR.

Comparison of six commercial kits to extract bacterial chromosome and plasmid DNA for MiSeq sequencing

OPEN

R

Laura Becker, Matthias Steglich, Stephan Fuchs, GuidoWerner & Ulrich Nbel,

We compared commercial kits for extraction of genomic DNA from the Gram-negative bacterium Klebsiella pneumoniae replicons, except for an apparent depletion of small plasmids (<

Recently, DNA sequencing technologies have undergone major improvements in terms of sequencing speed, throughput, and associated costs1. Due to this development, next generation sequencing is about to be integrated into routine practice in clinical microbiology laboratories2,3. Bacterial whole-genome sequencing in the diagnostic context enables pathogen identication and strain genotyping with ultimate discriminatory power for detection of transmission chains and outbreaks4,5. Furthermore, it promises to enable the prediction of a microbes phenotype, including antibiotic resistance and virulence68.

For preparation of microbial DNA for sequencing, robust extraction methods are required. Commercially available DNA extraction kits are usually preferred, as they provide superior reproducibility, quality control, and potential for automation. These kits rely on dierent principles for DNA purication, including solution- and solid-phase-based protocols9. While the latter make use of DNA-adsorbing materials (e.g. silica-membranes, silica-covered magnetic beads, or anion-exchange columns), which specically bind DNA and subsequently release it to an appropriate buer, solution-based (salting out) protocols are based on precipitation of DNA. The purity of DNA was previously reported to have eects on the reproducibility of sequencing library preparations10

and on the evenness of sequencing read distribution along the sequenced genome11. Moreover, depending on the extraction method applied, the purication of DNA molecules is known to be inuenced by their specic size, nucleotide composition, topology, and association with proteins12,13. Since individual bacterial genomes frequently consist of several replicons that may vary in size and copy number by orders of magnitude (e. g., chromosomes, plasmids), dierential extraction efficiency and unequal sequence representation may be expected. For compensation, costly increased overall sequencing coverage may be required to ensure reliable detection of diagnostically relevant polymorphisms and genes (e. g., predictive markers for antimicrobial resistance).

Numerous studies have documented that PCR-based analyses of microbial community composition may be aected by the DNA extraction methods applied, due to their dierential efficiency for diverse microorganisms (for recent examples, see1417). In contrast, comparative analyses of DNA extraction protocols for (meta-)genomic investigations or diagnostics are scarce18. To our best knowledge, the suitability of DNA extraction kits relying on dierent technical principles for purication of DNA from bacterial cultures to be used in genomic sequencing has not been systematically assessed.

In the present study, we compared the performance of six commercially available kits for extraction of DNA, namely Genomic-tip 20/G, MagAttract HMW DNA Kit, MasterPure DNA Purication Kit, Wizard Genomic DNA Purification Kit, DNeasy Blood & Tissue Kit and Plasmid Mini Kit, for subsequent Illumina Miseq

SCIENTIFIC REPORTS

1

www.nature.com/scientificreports/

Extraction Method

Genomic-tip 20/G Qiagen anion-exchange column

(gravity) 8.1 8 h (45min) 4109 9.8 (3.5) 1.77 (0.06)

MagAttract HMW

DNA Kit Qiagen

silica-membrane column

(spin) 3.2 3h (45min) 2109 10.9 (1.3) 1.72 (0.05)

Plasmid Mini Kit Qiagen alkaline lysis, anion-exchange

column (gravity) 5.2 3h 50min (40min) 18109 0.50 (0.2) 1.67 (0.03)

Table 1. Summary of DNA extraction kit characteristics. *List price on manufacturer web page in July 2015. **Approximate time to complete DNA extraction from three samples. SD: standard deviation from three independent DNA extractions.

sequencing. Experiments were performed using a clinical Klebsiella pneumoniae isolate, whose fully sequenced genome consists of a 5,278 kb chromosome, one large plasmid (362 kb), and two small plasmids (4.8 kb and 3.8kb)4,19.

Characteristics of the kits compared

Klebsiella pneumoniae 23412 was inoculated into 40ml of brain-heart-infusion broth and incubated at 37C and shaking at 140 rpm for 15 hours, resulting in 4 109 bacterial cells per milliliter. This culture was aliquoted and the biomass was pelleted by centrifugation and stored at 20C prior to DNA extraction. Basic characteristics of the kits compared are listed in Table1. Extraction costs per sample varied from 1.10 to 8.10, with salting-out kits being the least expensive (Table1). Extractions were performed according to the manufacturers protocols, and DNA was eluted or redissolved, respectively, in nuclease free water. All kits required similar hands-on time (3560minutes for three samples), but the lengths of incubation periods and total completion times varied more widely (Table1).

While DNA extracted with the Genomic-tip, MasterPure and MagAttract kits met the A260/A280 absorbance ratio (1.82.0) recommended for preparation of Nextera XT libraries (Illumina)20, other kits deviated from this range (Table1). DNA yields determined by using an assay based on uorescence (PicoGreen, Molecular Probes) varied considerably (Table1), but all kits supplied sufficient amounts of DNA (i. e., 1ng) for Nextera XT library preparation. Pulsed-eld gel electrophoresis (PFGE) indicated that the MagAttract and Genomic-tip kits provided the largest DNA fragments (up to 300 kb) (Fig.1). Accordingly, the 362-kb plasmid was not visible as a distinct band in any of the DNA extracts (Fig.1). In contrast, the two small plasmids were visually detectable as bands of 4 and 5kb, respectively, in DNA extracts from all kits applying binding of DNA to some matrix (Fig.1).

Sequencing libraries were prepared using the Nextera XT kit (Illumina) according to the manufacturers protocol. Capillary electrophoresis (applying the High Sensitivity DNA kit on an Agilent 2100 Bioanalyzer) did not reveal any fragment-size dierences between libraries prepared from the dierent extracts (not shown). Libraries were sequenced on a MiSeq machine (Illumina) using v3 reagents with 2 300 cycles according to the manufacturers instructions. In resulting sequencing reads, 75 to 89% of bases had quality scores Q30, and this proportion was independent from the DNA extraction method (data not shown). Sequencing reads were aligned to the reference genome sequence (concatenated chromosomal and plasmid sequences, accession nos. CP011313 to CP011316) by using BWA-SW (version 0.7.12-r1039, default parameters21). Read alignments (BAM les processed with SAMtools22) did not reveal any signicant dierences in read length and read span (insert size) between extraction kits (determined using a Python script23, no signicant dierence compared to the DNeasy kit, p0.02; data not shown).

Sequencing coverage of chromosome and plasmid DNA

Sequencing coverage along the reference genome sequence was determined from the read alignment by applying the sequence viewer module in Geneious 7.1.4 soware24 and normalized for the overall number of reads per library (Fig.2). All DNA extraction kits resulted in 29 to 56-fold average coverage of the chromosome and the 362-kb plasmid (Fig.2; no signicant dierence compared to the DNeasy kit, p 0.02). Further, the two small plasmids achieved higher coverage than the chromosome in all cases (Fig.2); for example, based on the Genomic-tip extract, the mean coverage was 34 (standard deviation, 3.7) for the chromosome and 2,164 (standard deviation, 982) for the 3.8-kb plasmid (Fig.2). Strikingly, however, DNA extraction with salting-out kits (MasterPure, Wizard Genomic) when compared to other kits resulted in 712 fold lower coverage of both small plasmids (p< 0.02; Fig.2). The true ratio of plasmid and chromosome copy numbers within the bacterial cells cannot easily be determined. By using quantitative real-time PCR25 (qPCR), we estimated that the two small

Manufacturer

Principle

Yield [g] (SD)

Costs per sample*[]

Purity [A260/280] (SD)

Completion time*(hands-on-time)

Cell count

DNA-binding magnetic beads,silica-based 4.4 2h 40min (1h) 2109 10.3 (6.6) 1.83 (0.05)

MasterPure DNA

Purication Kit Epicentre salting-out 1.1 2h 10min (35min) 0.4109 3.3 (1.0) 1.82 (0.03)

Wizard Genomic DNA

Purication Kit Promega salting-out 2.0 3h (35min) 4109 18.1 (7.5) 1.58 (0.01)

DNeasy Blood &

Tissue Kit Qiagen

SCIENTIFIC REPORTS

2

www.nature.com/scientificreports/

Figure 1. Pulsed-eld gel electrophoresis of K. pneumoniae 234-12 DNA extracted with six dierent kits.

DNA was extracted from aliquots of the same overnight culture according to the manufacturers protocols, and 10l of resulting extracts were loaded per lane. M1 Size standard Salmonella Braenderup lysed in agarose plug, DNA digested with XbaI27. 1 K. pneumoniae 234-12 lysed in agarose plug28, DNA digested with S1 nucleasefor presentation of the linearized 362-kb plasmid29. 2 Genomic-tip 20/G. 3 MagAttract HMW DNA Kit. 4 MasterPure DNA Purication Kit. 5 Wizard Genomic DNA Purication Kit. 6 DNeasy Blood & Tissue Kit. 7 Plasmid Mini Kit. M2 GeneRuler 1kb Plus DNA Ladder (Thermo Scientic).

Figure 2. Eect of DNA extraction kit on the sequencing coverage of the chromosome and the three plasmids of K. pneumoniae 23412. Means and standard deviations from three independent experiments are reported. For statistical analysis, two-tailed students t-tests with Bonferroni correction were performed (global signicance level = 0.10). Asterisks indicate a statistically signicant dierence compared to the DNeasy kit (p-value below 0.02). One gene was excluded from this evaluation, because orthologues were found on both the chromosome and the 3.8-kb plasmid.

plasmids were 23 times more abundant than the chromosome in a crude extract (based on boiling a resuspended bacterial cell pellet for 10min. and subsequent centrifugation to remove cell debris), and 39 times more abundant than the chromosome in matrix-based extracts (Suppl. Fig. S1). The slight enrichment of small plasmids may be caused by their preferential binding to anion exchange resins or silica matrices12,13. In contrast, small plasmids apparently got depleted during extractions with salting-out kits, since their copy numbers were estimated by qPCR to be lower than in the crude extract and even lower than that of the chromosome (down to only 20%; Suppl. Fig. S1). Interestingly, copy numbers of small plasmids appeared almost ten-fold higher based on sequencing coverage results when compared to qPCR results (Fig.3, Suppl. Fig. S1). Since it was previously reported that qPCR may underestimate the copy number of supercoiled plasmids in contrast to molecules linearized by restriction digestion26, we assume that our sequencing coverage results provide more precise estimates of actual copy numbers. Independent from the DNA extraction kits used, sequencing coverage was very even along each of the replicons (Fig.3). In any case, all kits yielded sufficient coverage for all replicons. Hence, the more balanced coverage of chromosome and plasmids achieved with salting-out protocols may be considered advantageous for economic reasons, as smaller overall sequencing output is required.

Conclusion

In conclusion, all DNA extraction kits tested yielded satisfactory MiSeq sequencing results. Our investigation was limited to a single bacterial isolate and species, which prohibits wide generalization. However, it was notable that the choice of extraction kit had little eect on sequencing read quality and on the evenness of sequencing

SCIENTIFIC REPORTS

3

www.nature.com/scientificreports/

Figure 3. Coverage along replicons. Bars correspond to the mean coverage of 10,000 nucleotides for the chromosome, 1,000 nucleotides for the 362-kb plasmid and 100 nucleotides for the two small plasmids, respectively. Pointed regions (red arrows) contain a gene occurring on both the 3.8-kb plasmid and chromosome.

coverage. In cases where a dierential coverage of smaller plasmids (<5kb in our case) may be considered negligible, the choice of DNA extraction kit can be guided largely by other factors including extraction costs, extraction time and potential for automation.

1. van Dijk, E. L., Auger, H., Jaszczyszyn, Y. & Thermes, C. Ten years of next-generation sequencing technology. Trends in genetics : TIG 30, 418426, doi: 10.1016/j.tig.2014.07.001 (2014).

2. Didelot, X., Bowden, R., Wilson, D. J., Peto, T. E. & Crook, D. W. Transforming clinical microbiology with bacterial genome sequencing. Nature reviews. Genetics 13, 601612, doi: 10.1038/nrg3226 (2012).

3. Peacock, S. J. & Weinstock, G. M. Microbial sequencing to improve individual and population health. Genome Med 6, 103, doi: 10.1186/s13073-014-0103-5 (2014).4. Haller, S. et al. What caused the outbreak of ESBL-producing Klebsiella pneumoniae in a neonatal intensive care unit, Germany 2009 to 2012? Reconstructing transmission with epidemiological analysis and whole-genome sequencing. BMJ Open 5, e007397, doi: 10.1136/bmjopen-2014-007397 (2015).5. Harris, S. R. et al. Whole-genome sequencing for analysis of an outbreak of meticillin-resistant Staphylococcus aureus: a descriptive study. The Lancet. Infectious diseases 13, 130136, doi: 10.1016/S1473-3099(12)70268-2 (2013).

6. Gordon, N. C. et al. Prediction of Staphylococcus aureus antimicrobial resistance by whole-genome sequencing. J Clin Microbiol 52, 11821191, doi: 10.1128/JCM.03117-13 (2014).

7. Holden, M. T. et al. A genomic portrait of the emergence, evolution, and global spread of a methicillin-resistant Staphylococcus aureus pandemic. Genome research 23, 653664, doi: 10.1101/gr.147710.112 (2013).

8. Zankari, E. et al. Genotyping using whole-genome sequencing is a realistic alternative to surveillance based on phenotypic antimicrobial susceptibility testing. J Antimicrob Chemother 68, 771777, doi: 10.1093/jac/dks496 (2013).

9. Tan, S. C. & Yiap, B. C. DNA, RNA, and protein extraction: the past and the present. Journal of biomedicine & biotechnology 2009, 574398, doi: 10.1155/2009/574398 (2009).

10. Lamble, S. et al. Improved workows for high throughput library preparation using the transposome-based Nextera system. BMC Biotechnol 13, 104, doi: 10.1186/1472-6750-13-104 (2013).

SCIENTIFIC REPORTS

4

www.nature.com/scientificreports/

11. van Heesch, S. et al. Systematic biases in DNA copy number originate from isolation procedures. Genome biology 14, R33, doi: 10.1186/gb-2013-14-4-r33 (2013).12. Melzak, K. A., Sherwood, C. S., Turner, R. F. B. & Haynes, C. A. Driving Forces for DNA Adsorption to Silica in Perchlorate Solutions. Journal of Colloid and Interface Science 181, 635644, http://dx.doi.org/10.1006/jcis.1996.0421 (1996).

13. Budelier, K. & Schorr, J. In Current Protocols in Molecular Biology (John Wiley & Sons, Inc., 2001).14. Peng, X. et al. Comparison of direct boiling method with commercial kits for extracting fecal microbiome DNA by Illumina sequencing of 16S rRNA tags. J Microbiol Methods 95, 455462, doi: 10.1016/j.mimet.2013.07.015 (2013).

15. Kennedy, N. A. et al. The impact of dierent DNA extraction kits and laboratories upon the assessment of human gut microbiota composition by 16S rRNA gene sequencing. PLos One 9, e88982, doi: 10.1371/journal.pone.0088982 (2014).

16. Henderson, G. et al. Eect of DNA extraction methods and sampling techniques on the apparent structure of cow and sheep rumen microbial communities. PLos One 8, e74787, doi: 10.1371/journal.pone.0074787 (2013).

17. Wst, P. K. et al. Estimates of Soil Bacterial Ribosome Content and Diversity Are Significantly Affected by the Nucleic Acid Extraction Method Employed. Appl Environ Microbiol 82, 25952607, doi: 10.1128/AEM.00019-16 (2016).

18. Wesolowska-Andersen, A. et al. Choice of bacterial DNA extraction method from fecal material inuences community structure as evaluated by metagenomic analysis. Microbiome 2, 19, doi: 10.1186/2049-2618-2-19 (2014).

19. Becker, L. et al. Complete Genome Sequence of a CTX-M-15-Producing Klebsiella pneumoniae Outbreak Strain from Multilocus Sequence Type 514. Genome Announc 3, doi: 10.1128/genomeA.00742-15 (2015).

20. Illumina. Nextera XT DNA Library Preparation Guide, Part # 15031942 Rev. C, October (2012).21. Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 17541760, doi: 10.1093/bioinformatics/btp324 (2009).22. Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 20782079, doi: 10.1093/bioinformatics/btp352 (2009).

23. Li, W. Automatically estimate insert size of the paired-end reads for a given SAM/BAM file. https://gist.github.com/ davidliwei/2323462, accessed in March 2015 (2015).

24. Kearse, M. et al. Geneious Basic: an integrated and extendable desktop soware platform for the organization and analysis of sequence data. Bioinformatics 28, 16471649, doi: 10.1093/bioinformatics/bts199 (2012).

25. Skulj, M. et al. Improved determination of plasmid copy number using quantitative real-time PCR for monitoring fermentation processes. Microb Cell Fact 7, 6, doi: 10.1186/1475-2859-7-6 (2008).

26. Providenti, M. A., OBrien, J. M., Ewing, R. J., Paterson, E. S. & Smith, M. L. The copy-number of plasmids and other genetic elements can be determined by SYBR-Green-based quantitative real-time PCR. J Microbiol Methods 65, 476487, doi: 10.1016/j. mimet.2005.09.007 (2006).

27. Hunter, S. B. et al. Establishment of a universal size standard strain for use with the PulseNet standardized pulsed-field gel electrophoresis protocols: converting the national databases to the new size standard. J Clin Microbiol 43, 10451050, doi: 10.1128/ JCM.43.3.1045-1050.2005 (2005).

28. Ribot, E. M. et al. Standardization of pulsed-eld gel electrophoresis protocols for the subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne pathogens and disease 3, 5967, doi: 10.1089/fpd.2006.3.59 (2006).

29. Barton, B. M., Harding, G. P. & Zuccarelli, A. J. A general method for detecting and sizing large plasmids. Anal Biochem 226, 235240, doi: 10.1006/abio.1995.1220 (1995).

We thank Kirstin Ganske and Sibylle Mller-Bertling for excellent technical assistance. This project has received funding from the European Unions Horizon 2020 program under grant agreement no. 643476 and from the German Federal Ministry of Health (IIA5-2513NIK006/321-4471-02/129).

Author Contributions

L.B., G.W. and U.N. designed the experiments. L.B. performed the experiments. L.B., M.S. and S.F. analyzed the data. L.B. and U.N. wrote the manuscript. All authors reviewed the manuscript.

Additional Information

Accession codes: Sequence data were submitted to the European Nucleotide Archive (http://www.ebi.ac.uk/ena) under accession number PRJEB10820.

Supplementary information accompanies this paper at http://www.nature.com/srep

Competing nancial interests: The authors declare no competing nancial interests.

How to cite this article: Becker, L. et al. Comparison of six commercial kits to extract bacterial chromosome and plasmid DNA for MiSeq sequencing. Sci. Rep. 6, 28063; doi: 10.1038/srep28063 (2016).

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

SCIENTIFIC REPORTS

5