ISSN 1755-098X MOLECULAR ECOLOGY RESOURCES20191602062051... · 531 Historical and non-invasive samples: a study case of genotyping errors in newly isolated microsatellites for the

Information on this journal (including instructions to authors) canbe accessed at http://wileyonlinelibrary.com

The journal is covered by AGRICOLA, Chemical Abstracts, Current Awareness Biological Sciences and Current Contents.

This journal is available online at Wiley Online Library.Visit wileyonlinelibrary.com to search the articles and register for table of contents email alerts.

RESOURCE ARTICLES

DNA Barcoding and Molecular Diagnostics437 A DNA mini-barcode for land plants

D. P. Little447 Identification of aphids of Aphis frangulae-group living on

Lamiaceae species through DNA barcodeG. E. Massimino Cocuzza & V. Cavalieri

458 The D1-D2 region of the large subunit ribosomal DNA asbarcode for ciliatesT. Stoeck, E. Przybos & M. Dunthorn

469 Design of Mini-barcode for Catfishes for assessment ofarchival biodiversityM. J. Bhattacharjee & S. K. Ghosh

478 Identification of Swedish mosquitoes based on molecularbarcoding of the COI gene and SNP analysisC. Engdahl, P. Larsson, J. Näslund, M. Bravo, M. Evander, J. O. Lundström, C. Ahlm & G. Bucht

489 Species and hybrid identification of sturgeon caviar: a newmolecular approach to detect illegal tradeE. Boscari, A. Barmintseva, J. M. Pujolar, P. Doukakis, N. Mugue & L. Congiu

499 DNA barcoding of Orchidaceae in KoreaH. M. Kim, S.-H. Oh, G. S. Bhandari, C.-S. Kim & C.-W. Park

508 Recovery of DNA barcodes from blackfly museumspecimens (Diptera: Simuliidae) using primer sets thattarget a variety of sequence lengthsL. M. Hernández-Triana, S. W. Prosser, M. A. Rodríguez-Perez, L. G. Chaverri, P. D. N. Hebert & T. Ryan Gregory

Molecular and Statistical Advances519 Evaluating the role of inbreeding depression in

heterozygosity-fitness correlations: how useful are tests foridentity disequilibrium?M. Kardos, F. W. Allendorf & G. Luikart

531 Historical and non-invasive samples: a study case ofgenotyping errors in newly isolated microsatellites for the lesser anteater (Tamandua tetradactyla L., Pilosa)C. L. Clozato, N. Moraes-Barros, F. R. Santos & J. S. Morgante

541 Parentage and sibship inference from markers inpolyploidsJ. Wang & K. T. Scribner

Permanent Genetic Resources554 Microsatellite markers from the Ion Torrent: a multi-

species contrast to 454 shotgun sequencingC. P. Elliott, N. J. Enright, R. J. N. Allcock, M. G. Gardner, E. Meglécz, J. Anthony & S. L. Krauss

569 Characterization and high cross-species transferability ofmicrosatellite markers from the floral transcriptome ofAspidistra saxicola (Asparagaceae)D. Huang, Y. Zhang, M. Jin, H. Li, Z. Song, Y. Wang & J. Chen

578 Development of a Nasonia vitripennis outbred laboratorypopulation for genetic analysisL. VAN DE Zande, S. Ferber, A. DE Haan, L. W. Beukeboom,J. VAN Heerwaarden & B. A. Pannebakker

588 A resource of single-nucleotide polymorphisms forrainbow trout generated by restriction-site associated DNAsequencing of doubled haploidsY. Palti, G. Gao, M. R. Miller, R. L. Vallejo, P. A. Wheeler, E. Quillet, J. Yao, G. H. Thorgaard, M. Salem & C. E. Rexroad III

597 Applicability of RAD-tag genotyping for interfamilialcomparisons: empirical data from two cetaceansA. Viricel, E. Pante, W. Dabin & B. Simon-Bouhet

606 Characterizing DNA preservation in degraded specimensof Amara alpina (Carabidae: Coleoptera)P. D. Heintzman, S. A. Elias, K. Moore, K. Paszkiewicz & I. Barnes

616 The use of archived tags in retrospective genetic analysis offishS. Bonanomi, N. Overgaard Therkildsen, R. Berg Hedeholm,J. Hemmer-Hansen & E. E. Nielsen

622 The first transcriptome and genetic linkage map for Asian arowanaX. Y. Shen, H. Y. Kwan, N. M. Thevasagayam, S. R. S. Prakki, I. S. Kuznetsova, S. Y. Ngoh, Z. Lim, F. Feng, A. Chang & L. Orbán

636 Characterization of the house sparrow (Passer domesticus)transcriptome: a resource for molecular ecology andimmunogeneticsR. Ekblom, P. Wennekes, G. J. Horsburgh & T. Burke

MOLECULARECOLOGY RESOURCES VOLUME 14, NUMBER 3, MAY 2014

MO

LECU

LAR

EC

OLO

GY R

ESOU

RC

ESV

OL

UM

E 14, N

UM

BE

R 3, M

AY

2014, pp

. 437–666

MOLECULARECOLOGYRESOURCES

VOLUME 14NUMBER 3

MAY2014

ISSN 1755-098X

Published by Wiley Blackwell

men_14_3_oc_Layout 1 4/3/2014 9:54 AM Page 1

Characterizing DNA preservation in degraded specimens ofAmara alpina (Carabidae: Coleoptera)

PETER D. HEINTZMAN,*1 SCOTT A. ELIAS,† KAREN MOORE,‡ KONRAD PASZKIEWICZ‡ and

IAN BARNES*2

*School of Biological Sciences, Royal Holloway University of London, Egham, TW20 0EX, UK, †Department of Geography, Royal

Holloway University of London, Egham, TW20 0EX, UK, ‡Exeter Sequencing Service, Biosciences, College of Life &

Environmental Sciences, University of Exeter, Exeter, EX4 4QD, UK

Abstract

DNA preserved in degraded beetle (Coleoptera) specimens, including those derived from dry-stored museum and

ancient permafrost-preserved environments, could provide a valuable resource for researchers interested in species

and population histories over timescales from decades to millenia. However, the potential of these samples as

genetic resources is currently unassessed. Here, using Sanger and Illumina shotgun sequence data, we explored

DNA preservation in specimens of the ground beetle Amara alpina, from both museum and ancient environments.

Nearly all museum specimens had amplifiable DNA, with the maximum amplifiable fragment length decreasing

with age. Amplification of DNA was only possible in 45% of ancient specimens. Preserved mitochondrial DNA

fragments were significantly longer than those of nuclear DNA in both museum and ancient specimens. Metage-

nomic characterization of extracted DNA demonstrated that parasite-derived sequences, including Wolbachia and

Spiroplasma, are recoverable from museum beetle specimens. Ancient DNA extracts contained beetle DNA in

amounts comparable to museum specimens. Overall, our data demonstrate that there is great potential for both

museum and ancient specimens of beetles in future genetic studies, and we see no reason why this would not be

the case for other orders of insect.

Keywords: ancient DNA, Coleoptera, DNA preservation, metagenomics, museum DNA, shotgun sequencing

Received 20 August 2013; revision received 15 November 2013; accepted 19 November 2013

Introduction

The beetles (Insecta: Coleoptera) are the most speciose

insect order (40%; >350 000 species) (Gullan & Cranston2010) and are found in nearly all terrestrial ecosystems,

fulfilling a great variety of niches (Grove & Stork 2000;

New 2007). They are therefore a major focus of biological

investigation, in which a genetic approach is often

required. To conduct genetic studies, specimens need to

be collected. However, due to the inhospitable nature of

some regions, such as the arctic, there may be limited

access and opportunity for collecting fresh specimens.

Due to more than two centuries of collection effort,

hundreds of millions of insect specimens have been

deposited in museum collections (Arino 2010), thereby

reducing the need to sample directly from difficult

regions (Schaefer et al. 2009). In addition, these speci-

mens can add a temporal perspective, on a decadal to

centennial timescale (Rowe et al. 2011), to genetic studies

(Harper et al. 2006; Hartley et al. 2006). However, DNA

from museum specimens is highly degraded (Wandeler

et al. 2007), which has no doubt dissuaded researchers

from utilizing them. Previous genetic studies have gener-

ally been small in scale and limited in scope, with no

standardized data that characterize the DNA present in

these remains (Goldstein & Desalle 2003; Gilbert et al.

2007a; Thomsen et al. 2009; Castalanelli et al. 2010; Gib-

son et al. 2012). A large-scale assessment of the propor-

tion and preservation of endogenous DNA is therefore

required. This would inform researchers keen to exploit

these genetic resources, as well as museum curators who

may be concerned about potentially destructive genetic

analyses (Mandrioli 2008).

Ancient beetle specimens, which are routinely recov-

ered from permafrost and other environments (Elias

Correspondence: Peter D. Heintzman, Fax: 1-831-459-5353;

E-mail: [email protected] Present address: Department of Ecology and Evolutionary

Biology, University of California, Santa Cruz, 1156 High Street,

Santa Cruz, CA 95064, USA2Present address: Earth Sciences Department, Natural History

Museum, Cromwell Road, London, SW7 5BD, UK

© 2013 John Wiley & Sons Ltd

Molecular Ecology Resources (2014) 14, 606–615 doi: 10.1111/1755-0998.12205

2010), are also potential genetic resources that have the

capacity to further extend the temporal aspect of beetle

DNA studies to millennial timescales. Recent proof of

concept studies has demonstrated the presence of beetle

ancient (a)DNA from a variety of environments and ages

(Reiss 2006; Willerslev et al. 2007; King et al. 2009; Thom-

sen et al. 2009), with permafrost considered to have the

greatest potential (Reiss 2006). The next step in beetle

aDNA research is to characterize the DNA present in

permafrost-preserved remains, which would assess their

potential as a future genetic resource.

The aim of this study therefore was to characterize

DNA preservation in beetle remains from both museum

and ancient environments. In characterizing DNA pres-

ervation from these two environments, we wished to

estimate the potential for recovery of DNA from very

small samples, overall success rate of PCR amplification,

retrievable fragment length using both PCR and Illumina

shotgun sequencing, as well as using the latter to exam-

ine the extent and nature of contamination, and potential

recovery of the metagenome. Altogether, these data

provide an overview of DNA preservation in both dry-

stored museum and ancient permafrost-preserved beetle

remains. Amara alpina (Carabidae: Coleoptera) was

selected as the study taxon, as it is well represented in

permafrost deposits, especially those of Beringia (Elias

et al. 2000) from which aDNA of a multitude of taxa has

been recovered (Shapiro & Cooper 2003) and is well

represented in the museum collections of Europe and

North America. Additionally, A. alpina is the most

cold-adapted ground beetle (Bennike et al. 2000) and is

important for palaeoclimatic and palaeoenvironmental

reconstruction (Elias 2010).

Materials and methods

Specimen provenance, age and collection

Specimens were divided into three age classes: modern

(10 years old) and ancient (permafrost deposits). Allspecimens, including age and provenance, are listed in

Table S1 (Supporting information). Modern specimens

(n = 6) were collected from four Beringian (Chukotka,Alaska, Yukon) localities between 2002 and 2004.

Museum specimens (n = 133), which had either beenpinned or glued to a card mount, were gathered from

the Canadian National Collection of Insects, Arachnids

and Nematodes (CNC) in April 2010 and the Swedish

Museum of Natural History (NRM) in May 2011. These

specimens were originally collected between 9 and

136 years before DNA extraction was conducted and

originated from 89 localities in Scandinavia (n = 28),Russia (n = 29) and North America (n = 76). A single

hind leg (including femur, tibia, tarsi), which is not

informative for morphological identification, was

removed for each analysis.

Ancient specimens (n = 148), consisting of completeor broken isolated sclerites, were collected from 15 Berin-

gian sites between 2003 and 2006 (Table S2, Supporting

information). Specimens ranged in age from >560 000 to5900 calendar years old, based on either radiocarbon

dating of associated plant remains, relative tephra dating

of layers above or below beetle-bearing sediments, or

through rough stratigraphic correlation. Specimens were

isolated from the sediment and stored at room tempera-

ture (Elias 1994). Prior to DNA extraction, ancient speci-

mens were stored at �20 °C to reduce furtherdegradation (Reiss 2006).

DNA extraction

Degraded specimens (museum, ancient) were extracted,

and PCR reactions were prepared, following standard-

ized aDNA handling protocols (Cooper & Poinar 2000;

Hofreiter et al. 2001; Gilbert et al. 2005; Wandeler et al.

2007). These included using sterile, dedicated ancient

DNA laboratories (Royal Holloway, NRM) and conduct-

ing independent replication (Data S1, Supporting infor-

mation). DNA was extracted using the QIAamp DNA

Micro kit (Qiagen), with a modified version of the tissues

protocol. This included using carrier RNA and conduct-

ing the final eluting step twice (50 lL each). Overnightlysis was conducted for 16 to 20 hours. Tween-20 was

added to DNA extracts (final concentration of 0.05%) to

prevent DNA adhering to the plastic tube surface (Fau-

lds et al. 2004). Extraction controls were used in a ratio of

one control to five samples. Specimens were not disinte-

grated prior to extraction as initial investigation indi-

cated that this did not affect the DNA recovery

likelihood. Modern DNA was extracted by Mack (2008)

using the same method, with the exception that complete

specimens were digested instead of a single hind leg

[following Gilbert et al. (2007a)]. Two specimens (one

modern, one ancient) were extracted in a previously

published study (Thomsen et al. 2009).

PCR and Sanger sequencing

Three markers were targeted: mitochondrial COI, and

multicopy nuclear 28S and ITS1. Novel overlapping

species-specific primer sets were designed using Oligo

v6.8 (Rychlik & Rychlik 2005). Primer design templates

were based on data from Gilbert et al. (2007a), Mack

(2008), Thomsen et al. (2009) and GenBank. PCR ampli-

fication conditions, primer sequences, sets and anneal-

ing temperatures are listed in Data S1 and Table S3

(Supporting information). Purification and sequencing


DNA PRESERVATION IN DEGRADED BEETLE SPECIMENS 607

of PCR products followed Brace et al. (2012). Sequencing

was also conducted at the NRM using an ABI 3100

genetic analyser and BigDye Terminator chemistry

(v1.1). Sequence data were quality-checked manually in

Sequencher v4.7 (Gene Codes) and compared against the

Basic Local Alignment Search Tool (BLAST) database

and A. alpina reference sequences (GenBank: KF695187-

KF695189) to ensure that the correct target had been

amplified. To detect any potential cross-sample contami-

nation, sequence data from overlapping fragments were

compared for each specimen. Due to the small sample

size of the modern data set, data from modern and

museum specimens were combined and are hereafter

referred to as museum. All stated fragment lengths

include primer sequence.

Specimens were considered to yield mitochondrial

(mt)DNA or nuclear (nu)DNA if one or more fragments

were recovered for each data type. All specimens that

failed to yield any DNA were tested with a minimum of

three or two primer sets for mtDNA and nuDNA,

respectively. The use of species-specific primers could

impede amplification if specimens had been taxonomi-

cally misidentified prior to PCR, leading to DNA recov-

ery rates being underestimated. Therefore, to assess

primer set specificity, primers were also tested on a spec-

imen from a congeneric species (A. glacialis). One ancient

specimen that yielded mtDNA was excluded from the

analysis of DNA recovery rates, due to extract exhaus-

tion prior to testing for the presence of nuDNA.

All mtDNA PCR products were sequence verified as

A. alpina. A total of 142 nuDNA PCR products were

sequenced, with at least one nuDNA fragment being

sequence verified from 71 of the 187 specimens reported

to have yielded nuDNA (museum: 45/136, ancient: 26/

51).

Amplification success was defined as the number of

mtDNA (COI) fragments recovered from eight overlap-

ping PCR primer sets, which had products of between

124 and 196 bp. Different combinations of these primers,

which amplified longer fragments (287 to 514 bp), were

tested on the museum specimens (Table S3, Supporting

information). A combination that would have amplified

four overlapping fragments was treated as four frag-

ments retrieved. For museum specimens, age was deter-

mined using either the collection date or other data

(e.g. specimen collected on the same collecting trip as a

specimen with an explicit collection date) stated on the

specimen label (Table S1, Supporting information).

Dated museum specimens were binned into 50-year time

intervals. Ancient specimens were binned into twenty

thousand year (ka) time intervals with age determined

by calibrated dates (Table S2, Supporting information).

Specimens binned as >60 ka ranged in age from >100 to>560 ka. To assess whether older specimens yielded

fewer DNA fragments, two-tailed Kruskal–Wallis tests

were employed in SPSS v19. Specimens that were

undated or imprecisely dated (e.g. Late Pleistocene,

‘LP’), as well as museum specimens that did not yield

mtDNA, were excluded from Kruskal–Wallis tests.

To assess the longest amplifiable mtDNA fragment,

museum specimens were initially tested for a 446 bp

fragment. Successfully amplified specimens were tested

for longer fragments and those that failed to amplify a

product were tested for sequentially shorter fragments.

A linear regression was performed in SPSS to test for a

relationship between the longest amplifiable fragment

and age. The specimens excluded from the Kruskal–Wal-

lis tests were also excluded from the regression analysis.

Illumina library preparation, shotgun sequencing andanalysis

Six samples (two from each age class), with excellent

DNA recovery based on Sanger sequencing, were used

for shotgun sequencing (Table S1, Supporting informa-

tion). DNA libraries were constructed using a modified

version of the Meyer & Kircher (2010) protocol, without

the initial DNA fragmentation step, and using QIAquick

spin columns (Qiagen) for all purification steps. These

modifications are listed in Data S1, Supporting informa-

tion. Amplified and indexed libraries were quantified

using a spectrophotometer and gel electrophoresis.

These were then diluted to equimolar concentrations,

pooled and sequenced on the Illumina HiSeq-2000

platform at the Exeter Sequencing Facility, using a

single lane of 2 9 100 cycles on a paired-end flow cell,

following the manufacturer’s instructions.

Raw HiSeq data were filtered to remove poor-quality

reads and sequencing artefacts using the standard (Blan-

kenberg et al. 2010) and FASTX toolkits (all v1.0.0 unless

otherwise stated), respectively, on the Galaxy server

(Goecks et al. 2010). Reads were binned by sample using

the FASTX Barcode Splitter, with a single base mismatch

permitted. For each data set, forward and reverse reads

were merged using SeqPrep, which combined the two

reads if overlap was detected and removed adapter

sequence. SeqPrep output consisted of the merged and

remaining unmerged reads, with between 87.2 and

99.7% of filtered reads being successfully merged. A

detailed methodology and pipeline of shotgun data qual-

ity control is available in Data S1 and Fig. S1 (Supporting

information).

To characterize the fragment length distribution of

endogenous DNA from A. alpina, merged reads were

aligned to 15 short reference sequences using Bowtie2

(Langmead & Salzberg 2012). These references included

eight mtDNA (658 to 16823 bp) and seven nuDNA (183

to 1043 bp) sequences that were either downloaded from


608 P . D . HEINTZMAN ET AL .

GenBank or originated from the Sanger data of this study

(Table S4, Supporting information). At least one repre-

sentative sequence was chosen for each locus available

for Amara, as well as the three available carabid mitoge-

nomes (as of 14/09/2012). A reference genome was not

used here, as the two available Coleoptera genomes

belong to taxa [Tribolium castaneum, Dendroctonus ponder-

osae (Richards et al. 2008; Keeling et al. 2013)] that

diverged from A. alpina more than 250 million years ago

(McKenna & Farrell 2009) and were therefore deemed

insufficiently similar to act as reference sequences.

Duplicate sequences were not removed from BAM files,

due to biases observed during duplicate removal (Data

S2 and Table S5, Supporting information). To ensure the

reasonable assignment of reads to reference sequences,

BAM files were indexed and visually inspected using

Tablet v1.12.09.03 (Milne et al. 2010). Fragment length

information was extracted from aligned reads in the

BAM files and pooled into two categories for each sam-

ple (mtDNA and nuDNA). Distributions were plotted

using a five-point centred moving average to smooth

length distributions. Descriptive statistics were calcu-

lated, and comparisons of fragment length distributions

between DNA categories were performed using t-tests in

SPSS. Appropriate t-statistics were selected using the

results of Levene’s test for equality of variances.

To assess the metagenome of degraded specimens,

merged and remaining unmerged read files were concat-

enated, and the ‘Collapse’ tool in Galaxy was used to

remove PCR duplicates. Reads were assembled into con-

tigs using a de novo assembly tool (clc_novo_assemble) in

the CLC assembly cell v4.0.6-beta, with the minimum

contig size set to 40 bp. Contigs were produced in this

way to increase the likelihood of robust taxonomic

assignment (Prufer et al. 2010) and to further collapse

PCR duplicates. Contig sequences were identified taxo-

nomically using BLAST v2.2.25 (database downloaded

on 23/08/2012 and gi_taxid_nucl.bin downloaded on

03/09/2012). Resulting taxonomic assignments were

visualized in MEGAN v4.70.3 (Huson et al. 2011), with

parameters set to minScore = 50, minComplexity = 0.44,topPercent = 10, winScore = 0 and minSupport = 5.These parameters ensured that contigs with low-quality

BLAST hits were excluded. Taxonomic information for

each of the specimens was combined, normalized and

collapsed to the rank of Class where possible. The twelve

most abundant classes were scrutinized further by

assessing the major composite genera, which each

comprised >2% of the identifiable contigs across allspecimens.

Results and discussion

DNA recovery rate

Nearly, all museum specimens (96%) yielded both

mtDNA and nuDNA (Table 1). Previous dry-stored

museum beetle DNA studies have reported a 46 to 100%

recovery rate (Goldstein & Desalle 2003; Gilbert et al.

2007a; Thomsen et al. 2009), which demonstrates the rela-

tive effectiveness of the methods employed here. The

three specimens that did not yield DNA were all from

the CNC, collected from the same locality, in the same

year, by the same collector and represent all the tested

specimens from this locality. We speculate that DNA-

degrading substances, such as ethyl acetate, may have

been used to kill and/or temporarily store these speci-

mens (Reiss et al. 1995; Dillon et al. 1996; Gilbert et al.

2007a).

In ancient specimens, 26% yielded both mtDNA and

nuDNA, whereas 54% yielded neither, and ~10% yieldedeither mtDNA or nuDNA only (Table 1). This is the first

reported recovery of nuDNA from a permafrost-pre-

served invertebrate and the second report of mtDNA

recovered from permafrost-preserved invertebrate ma-

crofossils [following Thomsen et al. (2009)]. MtDNA and

nuDNA were recovered from localities aged up to 55 000

and 41 000 radiocarbon years before present, respec-

tively, which represents the oldest ancient invertebrate

DNA recovered from macrofossils for both DNA types

(King et al. 2009; Thomsen et al. 2009). These recovery

rates are low compared with recovery rates from perma-

frost-preserved bone, which range between 22 and 80%

(Barnes et al. 2002, 2007; Shapiro et al. 2004; Campos et al.

2010a,b). This may be due to the very small size of beetle

sclerites (

magnitude smaller than the material used in bone-based

studies (Barnes et al. 2002, 2007; Shapiro et al. 2004; Cam-

pos et al. 2010a,b), therefore resulting in fewer template

molecules available for amplification. In addition, bones

are often ground into a fine powder to increase the DNA

yield. However, due to the potential loss of specimen

material, this was not conducted here. Alternatively,

thermal age analysis indicates that aDNA preservation

may be an order of magnitude poorer in beetle remains

when compared to bone (King et al. 2009), which may be

due to apatite reducing the fragmentation rate of DNA

in bone (Lindahl 1993).

Misidentification was not detected in ancient speci-

mens that yielded DNA (n = 68). It is unlikely that spe-cies-specific primer sets would have prevented any

potentially misidentified specimens from yielding DNA,

as primer sets were also successfully tested on a conge-

neric species (A. glacialis). Therefore, the assertion made

by Quaternary entomologists that even single broken

sclerites can be reliably identified (Coope 2004; Elias

2010) is supported by the genetic data.

DNA preservation by age

All mtDNA fragments were successfully amplified from

80.6% of museum specimens, whereas 3.6% yielded none

(Fig. 1a). The five specimens that did not yield mtDNA

were two of the oldest specimens from the NRM and the

three CNC specimens that did not yield any DNA. The

15.8% of specimens that amplified one to seven

fragments were almost exclusively >100 years old. Therelationship between museum specimen age and amplifi-

cation success was highly significant (two-tailed

Kruskal–Wallis test: v2 = 58.995, d.f. = 2, P < 0.001). Thismay indicate that the concentration of amplifiable DNA

decreases with age, due to the continuing occurrence of

strand breaks or other forms of damage (Lindahl 1993),

or the accumulation of interstrand cross-links through

time reducing amplification opportunity (Hansen et al.

2006). Decreasing amplification success with specimen

age has also been observed in other museum beetles (Gil-

bert et al. 2007a; Thomsen et al. 2009; Gibson et al. 2012),

as well as other insect orders (Watts et al. 2007; Strange

et al. 2009; van Houdt et al. 2010; Ugelvig et al. 2011;

Andersen & Mills 2012).

In the 37% of ancient A. alpina specimens that yielded

mtDNA, the majority of specimens yielded either one to

two fragments (16%) or seven to eight (15%; Fig. 1b). The

observation that around 15% of specimens yielded all, or

nearly all, of the mtDNA fragments tested demonstrates

the potential of ancient specimens for future PCR-based

studies. In contrast to the museum specimens, there was

a weak, but nonsignificant relationship between ancient

specimen age and the number of amplifiable fragments

(two-tailed Kruskal–Wallis test: v2 = 7.042, d.f. = 3,P = 0.071). The ancient specimens originated from geo-graphically widespread localities and could have there-

fore been subjected to differing preservation conditions,

which may explain the lack of a significant relationship

between age and amplification success. However, it

should be noted that specimens of >60 ka all failed toyield aDNA, but the sample size of these specimens was

small (n = 5). Reduced amplification success withincreasing specimen age in nonfrozen sediment-

preserved beetles has been observed elsewhere (King

et al. 2009).

Age is a highly significant predictor of the longest

amplifiable fragment (Fig. 1c) (linear regression:

R2 = 0.441, F = 98.612, d.f. = 1.125, P < 0.001). This sug-gests that strand breaks and/or interstrand cross-links

are still occurring for decades after the specimen has

been desiccated and stored. However, the regression

Fragments retrieved876543210

Spec

imen

s (%

)

80

60

40

20

0

Fragments retrieved876543210

Spec

imen

s (%

) 60

40

20

0

Collection year1990195019101870

Long

est f

ragm

ent (

bps) 600

400

200

0

(a) (b) (c)

Fig. 1 Amplification success of mitochondrial DNA from (a) museum and (b) ancient specimens of A. alpina based on the number of

amplifiable fragments, and (c) the longest amplifiable fragment retrieved from museum specimens by collection year. Colours indicate

specimen age: (a) collection year, blue: 1951–2004; red: 1901–1950; off-white: 1851–1900; green: no date, and (b) calibrated age, red:1–20 ka; orange: 21–40 ka; green: 41–60 ka; blue: >61 ka; grey: no date. In (c), the trend line equation is y = 1.946x � 3410.417, and thedotted lines illustrate 95% confidence intervals.



analysis may have been affected by several limitations in

the data set, including potential differences in efficiency

between primer sets and the noncontinuous nature of

the fragment lengths targeted. Comparable relationships

between age and maximum retrievable fragment length

have been shown in specimens of other insect orders

(Strange et al. 2009; Ugelvig et al. 2011). Intriguingly,

these studies targeted microsatellites, suggesting that

this observation may be applicable to mtDNA and

nuDNA, as well as a variety of different insect orders.

The regression analysis suggests that shorter mtDNA

fragments, of length viable for both PCR and next-gener-

ation sequencing-based approaches (70 bp), may be

retrievable from museum specimens of late 18th century

age. This is earlier than the oldest recovered museum

insect mtDNA (1820) (Thomsen et al. 2009) and encom-

passes the vast majority of entomological specimens

housed in collections, indicating their potential utility as

genetic resources.

Fragment length distributions

A total of 43.3 million paired-end reads were obtained

from the HiSeq run, of which 38 million passed quality

control measures and were assigned to a sample. 64 424

reads (0.17% of filtered reads) were identified as originat-

ing from A. alpina (Table S5, Supporting information).

However, due to the use of short reference sequences,

these reads represent a tiny fraction of the total number

of reads that originated from A. alpina.

The length distributions of the identified reads indi-

cate that nuDNA mean fragment length (79 to 109 bp) is

significantly shorter than mtDNA (61 to 101 bp; Fig. 2;

t-tests: P < 0.001; Table S6, Supporting information).These length distributions and mean length values are

similar between all samples and comparable to other

museum and ancient DNA data sets (Gilbert et al. 2007b,

2008; Miller et al. 2009; Rasmussen et al. 2011; Kircher

2012). The disparity between mtDNA and nuDNA mean

fragment length is in agreement with recent studies of

vertebrates from permafrost and sediment deposits (Sch-

warz et al. 2009; Allentoft et al. 2012), which suggest that

nuDNA degrades at a faster rate than mtDNA. This may

be due to the circular configuration of mtDNA making it

less accessible to exonucleases (Allentoft et al. 2012), the

double membrane of the mitochondrion offering addi-

tional protection (Schwarz et al. 2009) or the interaction

of nuDNA and histones facilitating strand breaks

(Binladen et al. 2006).

Metagenomic analysis

Contigs from modern and museum samples ranged in

length from 40 to 7935 bp, with N50 values of between

125 and 184 bp, of which 0.4 to 0.5% could be taxonomi-

cally identified (Table S7, Supporting information). The

taxonomic profiles of these samples are comparable

(Fig. 3), even considering the large age range of these

samples (9 to 138 years) and their independent storage

histories in separate museums (Table S1, Supporting

information). This suggests that the museum metage-

nome of historical dried beetle, and perhaps insect, speci-

mens may be fairly consistent, regardless of age or

storage collection.

0

0.01

0.02

0 50 100 150

0

0.01

0.02

Modern-1 Ancient-1Museum-1

Modern-2 Museum-2 Ancient-2

200Frag. length (bps)

Frag. length (bps)200150100500

Prop

. of r

eads

Prop

. of r

eads


Prop

. of r

eads

0.01

0

0.02


Prop

. of r

eads

0.01

0

0.02

0.03


Prop

. of r

eads

0

0.02

0.04


Prop

. of r

eads

0

0.02

0.04

Fig. 2 Fragment length distributions of DNA from three age classes (modern, museum, ancient) of A. alpina. The proportion of reads is

the number of reads for a given fragment length divided by the total number of reads in that sample. Fitted lines are based on a five-

point centred moving average. Solid line: nuclear DNA, dotted line: mitochondrial DNA.



The Insecta made up 38 to 49% of identifiable contigs

(Fig. 3; Table S7, Supporting information), which were

classified as belonging to model species in the Lepidop-

tera, Hymenoptera, Diptera, Hemiptera and Coleoptera

(Table S8, Supporting information). These model insects

have their nuclear genomes in the BLAST database and

are therefore biased towards during taxonomic assign-

ment. An exception is Abax, which does not have a refer-

ence genome, but is taxonomically close to A. alpina

(both Carabidae: Harpalinae). The number and variety of

component taxa further demonstrates the issues associ-

ated with the lack of a suitable reference genome, which,

together with the very low proportion of contigs that

could be assigned, would suggest that a substantial num-

ber of A. alpina contigs were not identified (Prufer et al.

2010). The single-leg extracted museum samples contain

comparable proportions of insect DNA to the whole-

specimen extracted modern samples, which suggests

that a single leg is sufficient for yielding museum DNA

from insects.

Parasites and commensals constitute the majority of

the bacterial contigs in the modern and museum

samples. Alphaproteobacterial contigs were mainly

composed of Wolbachia. The Wolbachia contigs for mod-

ern-1 were assigned to the wPip strain, whereas contigs

for the other modern and museum samples were

assigned to wRi. These strains belong to different

Wolbachia supergroups [wRi: A, wPip: B; (Klasson et al.

2009)], which would therefore suggest two separate

infection histories in this species. Contigs identified to

the bacterial class Mollicutes, which comprise 19% in

modern-2 and are found in very low proportions in the

museum samples, are made up of Spiroplasma. These

are arthropod commensals, which can be pathogenic

(Regassa & Gasparich 2006). As a commensal, Spiroplas-

ma are found in the arthropod gut and become patho-

genic when they enter the hemolymph (Regassa &

Gasparich 2006). Given that the modern samples had

far higher proportions of Spiroplasma compared with

the museum samples and that DNA was extracted

from whole specimens rather than legs in the former, it

is inferred that Spiroplasma detected here were com-

mensals. Another major bacterial parasite of beetles,

Rickettsia (Duron et al. 2008), was not detected in any of

the samples analysed. We see no obvious reason why

this metagenomic approach would not be applicable to

the investigation of parasitism in museum specimens of

other insect orders, as well as other arthropod groups.

The remaining identifiable contigs in the modern and

museum samples were assigned to the Mammalia (9 to

17%; Homo, Mus), Actinopterygii (Danio), Eudicotyledons

and Saccharomycetes. The origins of these contaminant

contigs are speculated to be from human handling

(Homo), the museum environment (Eudicotyledons,

InsectaArachnida

MammaliaActinopterygii SaccharomycetesEudicotyledons

AlphaproteobacteriaBetaproteobacteria GammaproteobacteriaMollicutes

ActinobacteriaFlavobacteriia

Prop

. of i

dent

ified

con

tigs (

%)

0

10

20

30

50

Modern-1

40

Modern-2 Museum-1 Museum-2 Ancient-1 Ancient-2

Fig. 3 Metagenomic composition of DNA from three age classes (modern, museum, ancient) of A. alpina. The twelve most abundant

classes and phyla across all data sets are illustrated, in descending order. Invertebrates: red, vertebrates: orange, plants and fungi: yel-

low, Proteobacteria: green, other bacteria: blue.



Saccharomycetes, Mus) and potential misassignments in

the BLAST database.

The two ancient samples comprise contigs that range

in length from 40 to 11617 bp, with N50 values of

between 174 and 214 bp, of which 16 to 19% could be

taxonomically identified (Table S7, Supporting informa-

tion). These samples have profoundly different

taxonomic profiles to the modern and museum samples,

with 0.2 to 0.3% of identified contigs assigned as Insecta

(Fig. 3; Table S7, Supporting information). There were

contigs assigned to Amara in both ancient samples. Given

the large taxonomic diversity of insects and the limited

amount of genetic information available for Amara, we

would not expect any contigs to be falsely identified to

the target genus, if the insect contigs were an artefact of

background contamination. Mammalian contamination

was low in the ancient samples (0.2 to 2%).

The vast majority of identified contigs from the

ancient samples were assigned to the Proteobacteria and

Actinobacteria. The ratios of these bacterial groups differ,

with the Alpha- and Betaproteobacteria (30 to 38% of

identified contigs) and Actinobacteria (38%) dominating

in ancient-1 and ancient-2, respectively (Fig. 3). These

bacterial compositions are consistent with the prove-

nance of these samples (permafrost); the component gen-

era suggest the samples were from glacial or periglacial

soils/sediments, near aquatic sources (Table S8, Support-

ing information). This was inferred from Polaromonas

[glacial and periglacial deposits (Darcy et al. 2011)],

Caulobacter [aquatic or semi aquatic habitats (Laub et al.

2007)] and the remaining genera being typical of a soil/

sediment environment (Janssen 2006; Philippot et al.

2007; Doughari et al. 2011). This complements the locality

information for these samples (Goldbottom Creek and

Titaluk River, in permanently frozen sediment) and

opens up the possibility of identifying potentially

unknown or dubious provenance information based on

the bacterial metagenome of ancient samples. However,

the diversity and ubiquity of many environmental bacte-

ria would only allow for general inferences to be made.

Although the proportion of endogenous DNA from

A. alpina in these chitinous remains seems to be around

two orders of magnitude lower than other permafrost or

cold-preserved tissues, such as bone and hair (Poinar

et al. 2006; Gilbert et al. 2008; Miller et al. 2008; Lindqvist

et al. 2010), this is due to BLAST bias, which resulted

from the lack of a suitable reference genome. Based on

the proportion of contigs identified, modern and

museum samples have an average of 165 times more

insect DNA than ancient samples. However, based on

the proportion of all contigs, this ratio decreases from

165:1 to 4:1, with the disparity between these two ratios

being due to a far greater proportion of contigs identi-

fied in the ancient samples (Table S7, Supporting

information). This suggests that the amount of insect

DNA in permafrost-preserved remains is comparable to

the amount found in museum specimens, although

direct comparison between the museum and the ancient

samples may be problematic if insect DNA in the ancient

samples was outcompeted by environmental DNA

during shotgun sequencing.

Conclusions

Using museum and ancient specimens of the ground

beetle A. alpina, this study has explored DNA preserva-

tion in degraded beetle specimens and assessed their

potential utility for future genetic studies. Museum spec-

imens have great potential, with nearly all specimens

containing amplifiable endogenous DNA. Furthermore,

it was possible to recover parasite sequences, which indi-

cates that museum specimens could be routinely used to

study host–parasite associations over decadal to centen-

nial timescales (Tsangarasa & Greenwood 2012). Ampli-

fiable endogenous DNA could only be recovered from

45% of ancient specimens, which would suggest that

large quantities of specimens would be required for

large-scale analyses. The vast majority of extracted DNA

was identified as originating from environmental bacte-

ria, although this was probably due to BLAST bias. The

major hurdle to future degraded insect DNA research

will be the availability of appropriate reference genome

sequences for maximum exploitation of recovered

sequence data. International initiatives, such as the 5000

insect genomes project (i5k), are anticipated to provide

such reference genomes over the next 5 years.

Acknowledgements

The authors thank Svetlana Kuzmina and Philip Thomsen for

specimens and DNA extracts; Yves Bousquet, Anthony Davies

(both CNC) and Johannes Bergsten (NRM) for access to the

museum specimens; Selina Brace, Jessica Thomas, Edwin van

Leeuwen, Roland Preece, Aurelien Ginolhac, Matthias Meyer

and Martin Kircher for technical advice and assistance; and Tom

Gilbert, Stephen Rossiter and three anonymous reviewers for

comments on earlier drafts of this work. PDH acknowledges

funding from a RHUL Reid Scholarship, the RHUL Research

Strategy Fund and SYNTHESYS (SE-TAF-1185; SYNTHESYS

receives funding from the European Community–ResearchInfrastructure Action under FP7 ‘Capacities’ Specific Pro-

gramme). Details on the i5k initiative can be found at: http://

www.arthropodgenomes.org/wiki/i5K.

References

Allentoft ME, Collins M, Harker D et al. (2012) The half-life of DNA in

bone: measuring decay kinetics in 158 dated fossils. Proceedings of the

Royal Society of London Series B-Biological Sciences, 279, 4724–4733.



Andersen JC, Mills NJ (2012) DNA extraction from museum specimens

of parasitic hymenoptera. PLoS ONE, 7, e45549.

Arino AH (2010) Approaches to estimating the universe of natural

history collections data. Biodiversity Informatics, 7, 81–92.

Barnes I, Matheus P, Shapiro B, Jensen D, Cooper A (2002) Dynamics of

Pleistocene population extinctions in Beringian brown bears. Science,

295, 2267–2270.

Barnes I, Shapiro B, Lister A et al. (2007) Genetic structure and extinction

of the woolly mammoth, Mammuthus primigenius. Current Biology, 17,

1072–1075.

Bennike O, Bj€orck S, B€ocher J, Walker IR (2000) The Quaternary arthro-

pod fauna of Greenland: a review with new data. Bulletin of the Geologi-

cal Society of Denmark, 47, 111–134.

Binladen J, Wiuf C, Gilbert MTP et al. (2006) Assessing the fidelity of

ancient DNA sequences amplified from nuclear genes. Genetics, 172,

733–741.

Blankenberg D, Gordon A, Von Kuster G et al. (2010) Manipulation of

FASTQ data with Galaxy. Bioinformatics, 26, 1783–1785.

Brace S, Barnes I, Powell A et al. (2012) Population history of the Hispa-

niolan hutia Plagiodontia aedium (Rodentia: Capromyidae): testing the

model of ancient differentiation on a geotectonically complex

Caribbean island. Molecular Ecology, 21, 2239–2253.

Campos PF, Kristensen T, Orlando L et al. (2010a) Ancient DNA

sequences point to a large loss of mitochondrial genetic diversity in

the saiga antelope (Saiga tatarica) since the Pleistocene. Molecular Ecol-

ogy, 19, 4863–4875.

Campos PF, Willerslev E, Sher A et al. (2010b) Ancient DNA analyses

exclude humans as the driving force behind late Pleistocene musk

ox (Ovibos moschatus) population dynamics. Proceedings of the

National Academy of Sciences of the United States of America, 107,

5675–5680.

Castalanelli MA, Severtson DL, Brumley CJ et al. (2010) A rapid non-

destructive DNA extraction method for insects and other arthropods.

Journal of Asia-Pacific Entomology, 13, 243–248.

Coope GR (2004) Several million years of stability among insect species

because of, or in spite of, Ice Age climatic instability? Philosophical

Transactions of the Royal Society of London Series B-Biological Sciences,

359, 209–214.

Cooper A, Poinar HN (2000) Ancient DNA: do it right or not at all.

Science, 289, 1139.

Darcy JL, Lynch RC, King AJ, Robeson MS, Schmidt SK (2011) Global dis-

tribution of Polaromonas phylotypes–evidence for a highly successful

dispersal capacity. PLoS ONE, 6, e23742.

Dillon N, Austin AD, Bartowsky E (1996) Comparison of preservation

techniques for DNA extraction from hymenopterous insects. Insect

Molecular Biology, 5, 21–24.

Doughari HJ, Ndakidemi PA, Human IS, Benade S (2011) The ecology,

biology and pathogenesis of Acinetobacter spp.: an overview. Microbes

and Environments, 26, 101–112.

Duron O, Bouchon D, Boutin S et al. (2008) The diversity of reproductive

parasites among arthropods: Wolbachia do not walk alone. BMC Biol-

ogy, 6, 27.

Elias SA (1994) Quaternary Insects and Their Environments. Smithsonian

Institute, Washington, District of Columbia.

Elias SA (2010) Advances in Quaternary Entomology. Elsevier, Oxford, UK.

Elias SA, Berman D, Alfimov A (2000) Late Pleistocene beetle faunas of

Beringia: where east met west. Journal of Biogeography, 27, 1349–1363.

Faulds K, Smith WE, Graham D (2004) Evaluation of surface-enhanced

resonance Raman scattering for quantitative DNA analysis. Analytical

Chemistry, 76, 412–417.

Gibson CM, Kao RH, Blevins KK, Travers PD (2012) Integrative taxon-

omy for continental-scale terrestrial insect observations. PLoS ONE, 7,

e37528.

Gilbert MTP, Bandelt HJ, Hofreiter M, Barnes I (2005) Assessing ancient

DNA studies. Trends in Ecology and Evolution, 20, 541–544.

Gilbert MTP, Moore W, Melchior L, Worobey M (2007a) DNA extraction

from dry museum beetles without conferring external morphological

damage. PLoS ONE, 2, e272.

Gilbert MTP, Tomsho LP, Rendulic S et al. (2007b) Whole-genome shot-

gun sequencing of mitochondria from ancient hair shafts. Science, 317,

1927–1930.

Gilbert MTP, Kivisild T, Gronnow B et al. (2008) Paleo-Eskimo mtDNA

genome reveals matrilineal discontinuity in Greenland. Science, 320,

1787–1789.

Goecks J, Nekrutenko A, Taylor J, Team G (2010) Galaxy: a comprehen-

sive approach for supporting accessible, reproducible, and transparent

computational research in the life sciences. Genome Biology, 11, 1–13.

Goldstein PZ, Desalle R (2003) Calibrating phylogenetic species forma-

tion in a threatened insect using DNA from historical specimens.

Molecular Ecology, 12, 1993–1998.

Grove SJ, Stork NE (2000) An inordinate fondness for beetles. Invertebrate

Taxonomy, 14, 733–739.

Gullan PJ, Cranston PS (2010) The Insects: An Outline of Entomology, 4th

edn. Wiley-Blackwell, Chichester, UK.

Hansen AJ, Mitchell DL, Wiuf C et al. (2006) Crosslinks rather than

strand breaks determine access to ancient DNA sequences from frozen

sediments. Genetics, 173, 1175–1179.

Harper GL, Maclean N, Goulson D (2006) Analysis of museum speci-

mens suggests extreme genetic drift in the adonis blue butterfly (Poly-

ommatus bellargus). Biological Journal of the Linnean Society, 88, 447–452.

Hartley CJ, Newcomb RD, Russell RJ et al. (2006) Amplification of DNA

from preserved specimens shows blowflies were preadapted for the

rapid evolution of insecticide resistance. Proceedings of the National

Academy of Sciences of the United States of America, 103, 8757–8762.

Hofreiter M, Serre D, Poinar HN, Kuch M, Paabo S (2001) Ancient DNA.

Nature Reviews Genetics, 2, 353–359.

van Houdt JK, Breman FC, Virgilio M, de Meyer M (2010) Recovering full

DNA barcodes from natural history collections of Tephritid fruitflies

(Tephritidae, Diptera) using mini barcodes. Molecular Ecology

Resources, 10, 459–465.

Huson DH, Mitra S, Ruscheweyh HJ, Weber N, Schuster SC (2011) Inte-

grative analysis of environmental sequences using MEGAN4. Genome

Research, 21, 1552–1560.

Janssen PH (2006) Identifying the dominant soil bacterial taxa in libraries

of 16S rRNA and 16S rRNA genes. Applied and Environmental Microbiol-

ogy, 72, 1719–1728.

Keeling CI, Yuen MM, Liao NY et al. (2013) Draft genome of the moun-

tain pine beetle, Dendroctonus ponderosae Hopkins, a major forest pest.

Genome Biology, 14, R27.

King GA, Gilbert MTP, Willerslev E, Collins MJ, Kenward H (2009)

Recovery of DNA from archaeological insect remains: first results,

problems and potential. Journal of Archaeological Science, 36, 1179–1183.

Kircher M (2012) Analysis of high-throughput ancient DNA sequencing

data. In: Ancient DNA: Methods and Protocols, Methods in Molecular Biol-

ogy (eds Shapiro B & Hofreiter M), pp. 197–228. Springer, New York.

Klasson L, Westberg J, Sapountzis P et al. (2009) The mosaic genome

structure of the Wolbachia wRi strain infecting Drosophila simulans. Pro-

ceedings of the National Academy of Sciences of the United States of America,

106, 5725–5730.

Langmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie

2. Nature Methods, 9, 357–359.

Laub MT, Shapiro L, McAdams HH (2007) Systems biology of Caulobact-

er. Annual Review of Genetics, 41, 429–441.

Lindahl T (1993) Instability and Decay of the Primary Structure of DNA.

Nature, 362, 709–715.

Lindqvist C, Schuster SC, Sun Y et al. (2010) Complete mitochondrial

genome of a Pleistocene jawbone unveils the origin of polar bear. Pro-

ceedings of the National Academy of Sciences of the United States of America,

107, 5053–5057.

Mack L (2008) Investigating mitochondrial DNA phylogenies of Arctic and

European beetle taxa. MSc thesis, Mainz University.

Mandrioli M (2008) Insect collections and DNA analyses: how to manage

collections? Museum Management and Curatorship, 23, 193–199.

McKenna DD, Farrell BD (2009) Beetles (Coleoptera). In: The Timetree of

Life (eds Hedges SB & Kumar S), pp. 278–289. Oxford University Press,

Oxford, UK.



Meyer M, Kircher M (2010) Illumina sequencing library preparation for

highly multiplexed target capture and sequencing. Cold Spring Harbor

Protocols, 2010, pdb prot5448.

Miller W, Drautz DI, Ratan A et al. (2008) Sequencing the nuclear genome

of the extinct woolly mammoth. Nature, 456, 387–390.

Miller W, Drautz DI, Janecka JE et al. (2009) The mitochondrial genome

sequence of the Tasmanian tiger (Thylacinus cynocephalus). Genome

Research, 19, 213–220.

Milne I, Bayer M, Cardle L et al. (2010) Tablet–next generation sequence

assembly visualization. Bioinformatics, 26, 401–402.

New TR (2007) Beetles and conservation. Journal of Insect Conservation, 11,

1–4.

Philippot L, Hallin S, Schloter M (2007) Ecology of denitrifying prokary-

otes in agricultural soil. Advances in Agronomy, 96, 249–305.

Poinar HN, Schwarz C, Qi J et al. (2006) Metagenomics to paleogenomics:

large-scale sequencing of mammoth DNA. Science, 311, 392–394.

Prufer K, Stenzel U, Hofreiter M et al. (2010) Computational challenges in

the analysis of ancient DNA. Genome Biology, 11, R47.

Rasmussen M, Guo X, Wang Y et al. (2011) An Aboriginal Australian gen-

ome reveals separate human dispersals into Asia. Science, 334, 94–98.

Regassa LB, Gasparich GE (2006) Spiroplasmas: evolutionary relation-

ships and biodiversity. Frontiers in Bioscience, 11, 2983–3002.

Reiss RA (2006) Ancient DNA from ice age insects: proceed with caution.

Quaternary Science Reviews, 25, 1877–1893.

Reiss RA, Schwert DP, Ashworth AC (1995) Field preservation of Coleoptera

for molecular genetic analyses. Environmental Entomology, 24, 716–719.

Richards S, Gibbs RA, Weinstock GM et al. (2008) The genome of the

model beetle and pest Tribolium castaneum. Nature, 452, 949–955.

Rowe KC, Singhal S, Macmanes MD et al. (2011) Museum genomics: low-

cost and high-accuracy genetic data from historical specimens. Molecu-

lar Ecology Resources, 11, 1082–1092.

Rychlik W, Rychlik P (2005) Oligo: Primer Analysis Software. Molecular

Biology Insights Inc, Cascade, Colorado, USA.

Schaefer H, Heibl C, Renner SS (2009) Gourds afloat: a dated phylogeny

reveals an Asian origin of the gourd family (Cucurbitaceae) and

numerous oversea dispersal events. Proceedings of the Royal Society of

London Series B-Biological Sciences, 276, 843–851.

Schwarz C, Debruyne R, Kuch M et al. (2009) New insights from old

bones: DNA preservation and degradation in permafrost preserved

mammoth remains. Nucleic Acids Research, 37, 3215–3229.

Shapiro B, Cooper A (2003) Beringia as an Ice Age genetic museum. Qua-

ternary Research, 60, 94–100.

Shapiro B, Drummond AJ, Rambaut A et al. (2004) Rise and fall of the

Beringian steppe bison. Science, 306, 1561–1565.

Strange JP, Knoblett J, Griswold T (2009) DNA amplification from pin-

mounted bumble bees (Bombus) in a museum collection: effects of frag-

ment size and specimen age on successful PCR. Apidologie, 40, 134–139.

Thomsen PF, Elias S, Gilbert MT et al. (2009) Non-destructive sampling

of ancient insect DNA. PLoS ONE, 4, e5048.

Tsangarasa K, Greenwood AD (2012) Museums and disease: using tissue

archive and museum samples to study pathogens. Annals of Anatomy,

194, 58–73.

Ugelvig LV, Nielsen PS, Boomsma JJ, Nash DR (2011) Reconstructing eight

decades of genetic variation in an isolated Danish population of the

large blue butterflyMaculinea arion. BMC Evolutionary Biology, 11, 201.

Wandeler P, Hoeck PE, Keller LF (2007) Back to the future: museum

specimens in population genetics. Trends in Ecology and Evolution, 22,

634–642.

Watts PC, Thompson DJ, Allen KA, Kemp SJ (2007) How useful is DNA

extracted from the legs of archived insects for microsatellite-based

population genetic analyses? Journal of Insect Conservation, 11, 195–198.

Willerslev E, Cappellini E, Boomsma W et al. (2007) Ancient biomolecules

from deep ice cores reveal a forested Southern Greenland. Science, 317,

111–114.

P.D.H., I.B. and S.A.E. designed the research. I.B. and

S.A.E. oversaw the project. S.A.E. provided ancient

samples. P.D.H. and S.A.E. performed the research. K.M.

and K.P. performed the Illumina sequencing. P.D.H.

analysed the data. P.D.H. wrote the paper with input

from all other authors.

Data accessibility

Representative mitochondrial and nuclear DNA

sequences have been deposited in GenBank with Acces-

sion nos. KF695187-KF695189. BAM files and metage-

nomic data sets are in the DRYAD database at doi:10.

5061/dryad.0179t. Sample data can be found in the

online supplementary material.

Supporting Information

Additional Supporting Information may be found in the online

version of this article:

Data S1 Supplementary methods.

Data S2 Duplicate removal bias.

Fig. S1 Customised workflow for analysis of Illumina paired-

end DNA sequence data.

Table S1 Data on all Amara specimens used in this study.

Table S2 Locality data for ancient specimens of A. alpina.

Table S3 Primers used in this study.

Table S4 Details of the 15 reference sequences used to retrieve

A. alpina sequences from the merged read files.

Table S5 The proportion of reads per sample (a) aligned to the

15 reference sequences listed in Table S4, and (b) affected by the

duplicate removal bias toward nuclear DNA in modern and

museum samples.

Table S6 Mean fragment length differences between mitochon-

drial and nuclear DNA.

Table S7 Details of the contigs used for metagenomic assess-

ment, and the proportion that were taxonomically assigned,

including those for the Insecta.

Table S8 Major taxonomic components of the groups identified

in Fig 3, and their inferred origin.



mer_v14_i3_coverHeintzman et al. 2014

Documents

ISSN 1755-098X MOLECULAR ECOLOGY RESOURCES20191602062051... · 531 Historical and non-invasive samples: a study case of genotyping errors in newly isolated microsatellites for the