Ancient DNA: Methods and Protocols (25 page)

2. A SYBR Green-based qPCR assay relies on a highly effi cient primer set that ideally generates amplicons of less than 250 bp (an arbitrary cut-off that is meant to ensure that all assays function effi ciently). Careful consideration must be given to ensure that the PCR reaction results in a minimum of homo-and hetero—

dimers (as the dimers will also bind SYBR Green). Researchers can refer to a number of online tools to assist in primer design.

It is well accepted that different primer sets have different sensitivities and hence different quantitation limits (the point at which the shifts in
C
values cease to respond as expected). It T

is advisable that a number of primer combinations are used and tested across a range of dilutions to fi nd the assay that is best suited to the application.

3. In some instances, absolute quantitation (exact copy numbers) may be required, and there are a number of ways to generate a standard curve. Standards can be useful in determining the qPCR assay sensitivity, defi ned as the absolute number of template molecules that a given qPCR assay can reproducibly detect. While amplicons or plasmids can be used as standards (using A260 or PicoGreen, Invitrogen, staining to determine concentrations), in our experience synthetic oligonucleotides are more reliable and reproducible. However, in other aDNA applications, absolute quantitation may not be required. For instance, the relative amount of DNA in each extract can easily be compared using only
C
values (free of inhibition) without T

an estimated absolute number. In fact, the considerable contamination risk involved with the use of high copy number standards might cancel out some of the benefi ts.

130

M. Bunce
et al.

4. A synthetic oligonucleotide that is approximately the same amplicon length as your assay should be employed as a synthetic standard. Many providers can synthesize long oligonucleotides, and HPLC purifi cation is not normally required.

The primer sequences should be incorporated into the 5 and 3 ¢

ends (see Fig. 2a
). The internal sequence can mimic the target sequence but should include an identifi able run of nucleotides (e.g., an insertion) so that it can be differentiated from that of your assay target. For contamination control purposes, it is also advisable to include uracil residue(s) in the middle of the sequence. This enables the standard to be inactivated by UNG

if requir
ed (see Fig. 2a ).

5. Make aliquots of the SYBR reagent in a clean room to avoid freeze–thaw cycles. Each PCR will require ~0.6 μ L of the reagent (see below). The SYBR is stable for approximately 1

year, and should be stored in the dark wrapped in foil. The 10,000× Gel stain is a cost-effective means by which to obtain SYBR Green. Note that the fi nal concentration of SYBR

Green in the PCR is not 1× of the original 10,000× stock.

The described concentration that equates to an approximate concentration of 0.12× is optimal for use in qPCR. A Rox dye (Invitrogen catalogue number 12223–012) can also be used

as an internal qPCR normalizer if required. Dilute the Rox dye 1:500 in DMSO, and use approximately 0.3 μ L per 25 μ L

reaction (this will need to be optimized for your qPCR

platform).

6. We prefer AmpliTaq Gold DNA polymerase (Applied Biosystems) for our aDNA qPCR assays, as the enzyme performs well on

aDNA templates

( 4
) . However, other commercially available hot-start
Taq
polymerases may be equally effi cient. We have trialed a number of proofreading polymerases both alone and as part of blends, all of which have performed poorly in qPCR

assays.

7. Because each oligonucleotide standard is single-stranded, you will need to add twice the determined volume to obtain a double-stranded equivalent. For example, 2 μ L of a 1 × 10 6 standard will be required to simulate 1 μ L of a 1 × 10 6 double-stranded standard.

8. The sensitivity and detection limits will differ between primer sets. However, no amplifi cation should be observed at
C
val-T

ues representing less than one standard template molecule. If this happens, the absolute copy number derived from the standard curve is not reliable. In our experience with oligonucleotide standards, the fewest number of copies that can be detected is between 2 and 20, and a reproducible quantitative result can be achieved at around 50–100 copies.

16 Quantitative Real-Time PCR in aDNA Research

131

9. All primer sets will have different sensitivities, detection, and quantitation limits, and it is important to explore these variables to ensure the fi delity and validity of your qPCR assay.

10. Within the aDNA clean room, it is advisable to separate the area (and pipettes) where PCR setup occurs from the area

where aDNA extract is added to the PCR tubes (e.g., using

two different PCR hoods or glove boxes). This setup will minimize the risk that reagents and equipment become cross—

contaminated with DNA.

11. The preferred time to conduct qPCR is when aDNA is freshly extracted. This maintains consistency between extracts, as storage time and freeze–thaw cycles will infl uence the number of amplifi able templates. Long-term storage of the DNA extract can result in DNA degradation and/or binding of DNA to the surfaces of plastic tubes.

12. Ensuring a good seal may seem trivial, but one problematic outcome of having a non-sealed tube (or plate seal) is that the SYBR dye will volatilize and condense on the camera (or in the well). This will generate anomalous qPCR results, and may

also damage the qPCR detection unit. For this reason it is not advisable to repeatedly use the same wells in a thermocycler block, especially during troubleshooting.

13. The decision to implement a two-step or three-step qPCR is assay-specifi c. Many manufacturers advocate a two-step strategy, citing evidence that with a two-step procedure more reliable data is obtained during the exponential phase. However, some qPCR assays may simply not be suited to this approach.

14. Briefl y, a melt curve analysis is an endpoint assay where reactions are heated to 60°C, imaged for SYBR, and then increased to the next temperature (typically 1°C steps) until 95°C. When the double-stranded DNA in the reaction denatures, a drop in fl uorescence is detected. Therefore, melt-curves can be a very valuable tool in the detection of primer dimer and nonspecifi c amplifi cation products, as these will tend to denature at different temperatures (i.e., if they have different lengths and/or a different GC-content) than the tar
get sequence (Fig.

3b ).

Melt curves can eliminate the need to run an agarose gel,

thereby saving time and decreasing the concentration of amplicon aerosols circulating in the laboratory.

15. An in-depth discussion of the analysis of qPCR outputs is beyond the scope of this chapter. The
C
(or absolute numbers T

if using a standard) generated by qPCR is only reliable if the
assay is validated

( 11
) . The presence of inhibition, primer dimer, nonspecifi c amplifi cation, and poor laboratory techniques will compromise the data fi delity. Once a DNA isolation protocol has been optimized, the
C
values can be used to T

132

M. Bunce
et al.

identify (or discard) samples that have, for example, a relatively high, medium, or low template copy number. It is often the case (at least with bone and eggshell) that samples showing high copy number also yield longer amplicons, work for nuclear DNA targets, and show less allelic drop-out when amplifying nuclear microsatellites or sex-linked loci
( 9, 10 )
.

Acknowledgments

MB was supported by the Australian Research Council as a Future Fellow (FT0991741). We thank Jayne Houston and James Haile for helpful discussions and Beth Shapiro for valuable editorial inputs.

References

1. Bustin SA (2004) A-Z of quantitative PCR. 7. Gilbert MT, Binladen J, Miller W, Wiuf C, International University Line, La Jolla, CA

Willerslev E, Poinar H, Carlson JE, Leebens-Mack

2. Bustin SA (2000) Absolute quantifi cation of

JH, Schuster SC (2007) Recharacterization of

mRNA using real-time reverse transcription

ancient DNA miscoding lesions: insights in the era

polymerase chain reaction assays. J Mol

of sequencing-by-synthesis. Nucleic Acids Res

Endocrinol 25:169–193

35:1–10

3. Bustin SA, Benes V, Nolan T, Pfaffl MW (2005)

8. Meyer M, Briggs AW, Maricic T, Hober B,

Quantitative real-time RT-PCR—a perspective.

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

J Mol Endocrinol 34:597–601

Hofreiter M (2008) From micrograms to pico—

4. Rohland N, Hofreiter M (2007) Comparison

grams: quantitative PCR reduces the material

and optimization of ancient DNA extraction.

demands of high-throughput sequencing.

Biotechniques 42:343–352

Nucleic Acids Res 36(1):e5

5. Oskam CL, Haile J, McLay E, Rigby P, 9. Allentoft M, Schuster S, Holdaway R, Hale M, Allentoft ME, Olsen ME, Bengtsson C,

McLay E, Oskam C, Gilbert MT, Spencer P,

Miller GH, Schwenninger JL, Jacomb C, Walter

Willerslev E, Bunce M (2009) Identifi cation of

R, Baynes A, Dortch J, Parker-Pearson M,

microsatellites from an extinct moa species

Gilbert MT, Holdaway RN, Willerslev E,

using high-throughput (454) sequence data.

Bunce M (2010) Fossil avian eggshell pre—

Biotechniques 46:195–200

serves ancient DNA. Proc Biol Sci 277:

10. Allentoft ME, Bunce M, Scofi eld RP, Hale ML,

1991–2000

Holdaway RN (2010) Highly skewed sex ratios

6. Pruvost M, Schwarz R, Correia VB, Champlot

and biased fossil deposition of moa: ancient

S, Braguier S, Morel N, Fernandez-Jalvo Y,

DNA provides new insight on New Zealand’s

Grange T, Geigl EM (2007) Freshly exca—

extinct megafauna. Quat Sci Rev 29:753–762

vated fossil bones are best for amplifi cation

11. Bustin SA (2010) Why the need for qPCR

of ancient DNA. Proc Natl Acad Sci USA

publication guidelines?—The case for MIQE.

104:739–744

Methods 50:217–226

Multiplex PCR Amplifi cation of Ancient DNA

Mathias Stiller and Tara L. Fulton Abstract

Multiplex PCR allows the simultaneous amplifi cation of up to dozens of target fragments in a single PCR.

It is therefore a powerful tool to obtain many kilobases of continuous sequence from minute amounts of ancient DNA (aDNA), which usually must be amplifi ed in multiple short and overlapping fragments.

Because signifi cantly less template is required compared to amplifying each fragment separately, multiplex PCR is particularly benefi cial when the fossil material itself, or access to the fossil material, is limited. The recently refi ned two-step multiplex PCR protocol consists of a fi rst-step reaction (the actual multiplex PCR) that then acts as the template for the second-step PCR. During the second step, nested primers are used in individual amplifi cation reactions. Although the same set of primers can be used in both steps, using a nested set in the second step adds an additional level of selectivity and specifi city, minimizing PCR

artifacts. This is particularly important when complex mixtures of template DNA, such as aDNA extracts, are amplifi ed.

Key words:
Polymerase chain reaction , Two-step multiplex PCR , Singleplex PCR , Monoplex PCR , Overlapping fragments , Nested primer , Ancient DNA

1. Introduction

The polymerase chain reaction (PCR) offers specifi city and single-molecule sensitivity, making it an excellent method for analyzing ancient DNA (aDNA). However, each PCR requires at least one template molecule of the desired genomic region. Depending on DNA preservation, more or less DNA extract will be required to begin each amplifi cation reaction. Multiple amplifi cations of short, overlapping fragments may be necessary to reconstruct long, informative DNA sequences. This process is performed in replicate to account for possible sequence errors due to miscoding lesions in the template molecule
( 1 )
. Ignoring economical constraints, the total number of PCR amplifi cations that can possibly be performed will depend on the amount of the sample that exists and the 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_17, © Springer Science+Business Media, LLC 2012

133

134

M. Stiller and T.L. Fulton

preservation of that sample. When analyzing unique or rare samples (such as hominids or carnivores), the amount of sequence data that can be obtained from a given specimen may be limited.

Multiplex PCR was fi rst established in 1988 to simultaneously amplify multiple loci in the human dystr
ophin gene ( 2 )
and has been used subsequently in a variety of applications including patho-gen identifi cation, gender screening, linkage analysis, template quantitation, genetic disease diagnosis, forensics, and population genetic studies (see
( 3
) for a comprehensive overview). Although low-number multiplex PCR has been used previously to amplify aDNA
( 4, 5
)
, the recently developed two-step approach
( 6, 7
) is the only approach to date that is able to overcome some of the major obstacles of aDNA research.

Using the two-step approach, Krause and colleagues
( 6 )
were able to reconstruct the complete mitochondrial genome (16,770

base pairs, bp) of a 12,000-year-old wooly mammoth using the equivalent of only 200 mg of bone, based on two primary (fi rst-step) multiplex reactions of 23 primer sets each. Since 2006, the approach has been used successfully to amplify both mitochondrial
( 8– 13
)
and nuclear aDNA ( 14, 15 )
from various extinct species.