Ancient DNA: Methods and Protocols (29 page)

5. Do not vortex mixtures containing enzymes as this may reduce enzyme activity.

6. USER enzyme is a convenient mixture of UDG and endoVIII sold by NEB. The exact ratio of the enzymes is proprietary, but if USER is not available, it can be replaced in the 50 m L repair reaction by 2

m L endoVIII (10 U/

m L) and 1-m L UDG

(5 U/ m L) (NEB unit defi nitions).

7. It is critical to add the ligase only after the other components have been mixed in order to avoid contact between ligase and locally high concentrations of incompletely mixed DNA inserts or adaptors. This could lead to increased chimeric insert or adaptor dimer formation. Since the Quick ligase buffer contains PEG-6000, which is highly viscous, be aware that it can take slightly more pipetting/fl icking than usual to thoroughly mix the ligation reagents.

8.
Bst
polymerase has a strong strand displacement function and can use the unligated 3’ OH of the DNA insert fragment as a primer for second strand synthesis of the adaptor. It will displace the short arm of the partially double-stranded adaptor and fi ll in the missing par
t (Fig. 1
(v)).

9. Unlike previous studies, this protocol performs the fi ll-in step in solution and does not use streptavidin beads, as we have found that beads are unnecessary. By removing this step, no material is lost between fi ll-in and library amplifi cation, increasing yield and reproducibility and at the same time simplifying the protocol.

10. We recommend Amplitaq Gold DNA polymerase (Applied

Biosystems) for the primary library amplifi cation as we have found it to work well in the Thermopol buffer that is used for the fi ll-in polymerase.

11. All steps subsequent to adaptor ligation can be performed outside the aDNA cleanroom. Take care, however, to avoid library cross-contamination.

12. The yield of libraries prepared from aDNA can vary dramatically depending on the sample, from virtually zero to 10 12

ligated insert molecules. In any case, however, 12 cycles of primary amplifi cation should be enough to produce over 1,000

18 Preparation of Next-Generation Sequencing Libraries from Damaged DNA 153

copies of each unique starting molecule unless the starting concentration is very high and PCR is saturated early in the exponential phase (in which case amplifi cation is less necessary anyway).

13. A 100-m L Phusion PCR reaction of an aDNA sequencing

library will produce a maximum of ~1 m g of amplifi ed library DNA. If higher amounts of library are required, multiple parallel PCRs can be performed.

14. An accurate dilution series is crucial for quantifi cation of your library. For convenience, we recommend storing the standards in eight-tube 0.2-mL PCR strips to allow multichannel

pipetting into the qPCR reactions. If the strips are not siliconized/with low-retention, add 0.05% Tween-20 to the TE

buffer to avoid DNA sticking to the tube walls over time.

15. The extraction negative (blank) library and negative library control (water only) will not be DNA-free, but should display only short adaptor-artifacts when visualized on a gel after PCR.

A smear of longer molecules may indicate contamination. The concentration of molecules in these control libraries should be lower than in the sample library.

References

1. Kircher M, Kelso J (2010) High-throughput 7. Burbano HA, Hodges E, Green RE, Briggs DNA sequencing—concepts and limitations.

AW, Krause J, Meyer M, Good JM, Maricic T,

Bioessays 32:524–536

Johnson PL, Xuan Z, Rooks M, Bhattacharjee

2. Metzker ML (2010) Sequencing technolo—

A, Brizuela L, Albert FW, de la Rasilla M,

gies—the next generation. Nat Rev Genet

Fortea J, Rosas A, Lachmann M, Hannon GJ,

11:31–46

Paabo S (2010) Targeted investigation of the

3. Blow MJ, Zhang T, Woyke T, Speller CF,

Neandertal genome by array-based sequence

Krivoshapkin A, Yang DY, Derevianko A,

capture. Science 328:723–725

Rubin EM (2008) Identifi cation of ancient 8. Maricic T, Whitten M, Paabo S (2010) remains through genomic sequencing. Genome

Multiplexed DNA sequence capture of mito—

Res 18:1347–1353

chondrial genomes using PCR products. PLoS

4. Briggs AW, Good JM, Green RE, Krause J,

One 5:e14004

Maricic T, Stenzel U, Lalueza-Fox C, Rudan P,

9. Green RE, Krause J, Briggs AW, Maricic T,

Brajkovic D, Kucan Z, Gusic I, Schmitz R,

Stenzel U, Kircher M, Patterson N, Li H, Zhai

Doronichev VB, Golovanova LV, de la Rasilla

W, Fritz MH, Hansen NF, Durand EY,

M, Fortea J, Rosas A, Paabo S (2009) Targeted

Malaspinas AS, Jensen JD, Marques-Bonet T,

retrieval and analysis of fi ve Neandertal mtDNA

Alkan C, Prufer K, Meyer M, Burbano HA,

genomes. Science 325:318–321

Good JM, Schultz R, Aximu-Petri A, Butthof

5. Green RE, Krause J, Ptak SE, Briggs AW,

A, Hober B, Hoffner B, Siegemund M,

Ronan MT, Simons JF, Du L, Egholm M,

Weihmann A, Nusbaum C, Lander ES, Russ C,

Rothberg JM, Paunovic M, Paabo S (2006)

Novod N, Affourtit J, Egholm M, Verna C,

Analysis of one million base pairs of Neanderthal

Rudan P, Brajkovic D, Kucan Z, Gusic I,

DNA. Nature 444:330–336

Doronichev VB, Golovanova LV, Lalueza-Fox

6. Briggs AW, Stenzel U, Johnson PL, Green RE,

C, de la Rasilla M, Fortea J, Rosas A, Schmitz

Kelso J, Prufer K, Meyer M, Krause J, Ronan

RW, Johnson PL, Eichler EE, Falush D, Birney

MT, Lachmann M, Paabo S (2007) Patterns of

E, Mullikin JC, Slatkin M, Nielsen R, Kelso J,

damage in genomic DNA sequences from a

Lachmann M, Reich D, Paabo S (2010) A draft

Neandertal. Proc Natl Acad Sci USA 104:

sequence of the Neandertal genome. Science

14616–14621

328:710–722

154

A.W. Briggs and P. Heyn

10. Miller W, Drautz DI, Ratan A, Pusey B, Qi J,

15. Hofreiter M, Serre D, Poinar HN, Kuch M,

Lesk AM, Tomsho LP, Packard MD, Zhao F,

Pääbo S (2001) Ancient DNA. Nat Rev Genet

Sher A, Tikhonov A, Raney B, Patterson N,

2:353–359

Lindblad-Toh K, Lander ES, Knight JR, Irzyk

16. Hofreiter M, Jaenicke V, Serre D, Haeseler Av

GP, Fredrikson KM, Harkins TT, Sheridan S,

A, Paabo S (2001) DNA sequences from mul—

Pringle T, Schuster SC (2008) Sequencing the

tiple amplifi cations reveal artifacts induced by

nuclear genome of the extinct woolly mam—

cytosine deamination in ancient DNA. Nucleic

moth. Nature 456:387–390

Acids Res 29:4793–4799

11. Rasmussen M, Li Y, Lindgreen S, Pedersen JS,

17. Heyn P, Stenzel U, Briggs AW, Kircher M,

Albrechtsen A, Moltke I, Metspalu M, Metspalu

Hofreiter M, Meyer M (2010) Road blocks

E, Kivisild T, Gupta R, Bertalan M, Nielsen K,

on paleogenomes—polymerase extension

Gilbert MT, Wang Y, Raghavan M, Campos

profi ling reveals the frequency of blocking

PF, Kamp HM, Wilson AS, Gledhill A, Tridico

lesions in ancient DNA. Nucleic Acids Res

S, Bunce M, Lorenzen ED, Binladen J, Guo X,

38:e161

Zhao J, Zhang X, Zhang H, Li Z, Chen M, 18. Briggs AW, Stenzel U, Meyer M, Krause J, Orlando L, Kristiansen K, Bak M, Tommerup

Kircher M, Paabo S (2010) Removal of deamiN, Bendixen C, Pierre TL, Gronnow B,

nated cytosines and detection of in vivo methy—

Meldgaard M, Andreasen C, Fedorova SA,

lation in ancient DNA. Nucleic Acids Res

Osipova LP, Higham TF, Ramsey CB, Hansen

38:e87

TV, Nielsen FC, Crawford MH, Brunak S,

Sicheritz-Ponten T, Villems R, Nielsen R, 19. Krause J, Briggs AW, Kircher M, Maricic T, Krogh A, Wang J, Willerslev E (2010) Ancient

Zwyns N, Derevianko A, Paabo S (2010) A

human genome sequence of an extinct Palaeo—

complete mtDNA genome of an early modern

Eskimo. Nature 463:757–762

human from Kostenki, Russia. Curr Biol

20:231–236

12. Reich D, Green RE, Kircher M, Krause J, 20. Krause J, Fu Q, Good JM, Viola B, Shunkov Patterson N, Durand EY, Viola B, Briggs AW,

MV, Derevianko AP, Paabo S (2010) The com—

Stenzel U, Johnson PL, Maricic T, Good JM,

plete mitochondrial DNA genome of an

Marques-Bonet T, Alkan C, Fu Q, Mallick S,

unknown hominin from southern Siberia.

Li H, Meyer M, Eichler EE, Stoneking M,

Nature 464:894–897

Richards M, Talamo S, Shunkov MV, 21. Meyer M, Briggs AW, Maricic T, Hober B, Derevianko AP, Hublin JJ, Kelso J, Slatkin M,

Hoffner B, Krause J, Weihmann A, Paabo S,

Paabo S (2010) Genetic history of an archaic

Hofreiter M (2008) From micrograms to pico—

hominin group from Denisova Cave in Siberia.

grams: quantitative PCR reduces the material

Nature 468:1053–1060

demands of high-throughput sequencing.

13. Gilbert MT, Drautz DI, Lesk AM, Ho SY, Qi

Nucleic Acids Res 36:e5

J, Ratan A, Hsu CH, Sher A, Dalen L, 22. Margulies M, Egholm M, Altman WE, Attiya Gotherstrom A, Tomsho LP, Rendulic S,

S, Bader JS, Bemben LA, Berka J, Braverman

Packard M, Campos PF, Kuznetsova TV,

MS, Chen YJ, Chen Z, Dewell SB, Du L,

Shidlovskiy F, Tikhonov A, Willerslev E,

Fierro JM, Gomes XV, Godwin BC, He W,

Iacumin P, Buigues B, Ericson PG, Germonpre

Helgesen S, Ho CH, Irzyk GP, Jando SC,

M, Kosintsev P, Nikolaev V, Nowak-Kemp M,

Alenquer ML, Jarvie TP, Jirage KB, Kim JB,

Knight JR, Irzyk GP, Perbost CS, Fredrikson

Knight JR, Lanza JR, Leamon JH, Lefkowitz

KM, Harkins TT, Sheridan S, Miller W,

SM, Lei M, Li J, Lohman KL, Lu H, Makhijani

Schuster SC (2008) Intraspecifi c phylogenetic

VB, McDade KE, McKenna MP, Myers EW,

analysis of Siberian woolly mammoths using

Nickerson E, Nobile JR, Plant R, Puc BP,

complete mitochondrial genomes. Proc Natl

Ronan MT, Roth GT, Sarkis GJ, Simons JF,

Acad Sci U S A 105:8327–8332

Simpson JW, Srinivasan M, Tartaro KR, Tomasz

14. Stiller M, Knapp M, Stenzel U, Hofreiter M,

A, Vogt KA, Volkmer GA, Wang SH, Wang Y,

Meyer M (2009) Direct multiplex sequencing

Weiner MP, Yu P, Begley RF, Rothberg JM

(DMPS)—a novel method for targeted high-

(2005) Genome sequencing in microfabricated

throughput sequencing of ancient and highly

high-density picolitre reactors. Nature 437:

degraded DNA. Genome Res 19:1843–1848

376–380

High-Throughput Sequencing

Michael Knapp , Mathias Stiller , and Matthias Meyer Abstract

Molecular barcoding is an essential tool to use the high throughput of next generation sequencing platforms optimally in studies involving more than one sample. Various barcoding strategies allow for the incorporation of short recognition sequences (barcodes) into sequencing libraries, either by ligation or polymerase chain reaction (PCR). Here, we present two approaches optimized for generating barcoded sequencing libraries from low copy number extracts and amplifi cation products typical of ancient DNA studies.

Key words:
Next generation high-throughput sequencing , Direct multiplex sequencing , Multiplex PCR , DNA capture , Ancient DNA , Barcoding

1. Introduction

1.1. Background

High-throughput sequencing produces enormous amounts of

sequence data compared to traditional Sanger sequencing
( 1 )
. For many studies, even the smallest lane on any next generation sequencing (NGS) instrument will produce excessive amounts of sequence data for a single sample. Moreover, if larger numbers of samples are to be analyzed, the cost soon becomes prohibitive if a full single lane is used per sample. It is therefore important to be able to pool multiple samples and sequence them in a single lane.

As the information about sample origin of individual sequence reads is lost in all NGS approaches, this requires barcoding techniques, in which a specifi c tag is attached to all DNA fragments allowing them to be sorted bioinformatically after sequencing
( 2 )
.

The most effi cient way to produce a barcoded sequencing

library is to amplify a genomic target region using polymerase chain reaction (PCR) with target-specifi c primers that include a Beth Shapiro and Michael Hofreiter (eds.),
Ancient DNA: Methods and Protocols
, Methods in Molecular Biology, vol. 840, DOI 10.1007/978-1-61779-516-9_19, © Springer Science+Business Media, LLC 2012

155

156

M. Knapp
et al.

sequencing adapter and barcode tail. While this approach is fast and simple, it has some drawbacks that limit its use. A complete barcoded set of primers is needed for each sample that is to be sequenced, which makes this strategy expensive for population level studies. It is also not suitable to barcode shotgun sequencing libraries.

Most other barcoding strategies rely on ligation of barcodes or barcoded sequencing adapters to amplifi ed or nonamplifi ed target DNA molecules. While more template DNA is lost than in a barcoding protocol using tailed PCR primers, this approach is suitable for all double-stranded DNA. The manufacturers of all major NGS

platforms provide protocols for adapter ligation and further protocols are available in the scientifi c literature (e.g.,
( 3, 4
) ). Most of these, however, are not designed for ancient DNA applications.

Here, we present two protocols that were developed with specifi c ancient DNA applications in mind. Protocol 1 was optimized for producing barcoded sequencing libraries from highly degraded, low copy number DNA extracts. Protocol 2 was designed for barcoding preamplifi ed multiplex PCR products, but can also be used for barcoding regular PCR products. Both protocols are derived from the original 454 library preparation protocol by Margulies
et al.
( 5
) .