American Journal of Epidemiology

The Author 2008. Published by the Johns Hopkins Bloomberg School of Public Health.
All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

Vol. 167, No. 10

DOI: 10.1093/aje/kwn012
Advance Access publication March 7, 2008

Practice of Epidemiology
Buccal Swabs and Treated Cards: Methodological Considerations for Molecular
Epidemiologic Studies Examining Pediatric Populations

Sara M. Beckett1, Stephen J. Laughton2, Luciano Dalla Pozza3, Geoffrey B. McCowage3,

Glenn Marshall4, Richard J. Cohn4, Elizabeth Milne5, and Lesley J. Ashton1
1

Childrens Cancer Institute Australia for Medical Research, Randwick, New South Wales, Australia.
Department of Hematology and Oncology, Starship Childrens Hospital, Auckland, New Zealand.
3
Department of Oncology, The Childrens Hospital at Westmead, Westmead, New South Wales, Australia.
4
Center for Childrens Cancer and Blood Disorders, Sydney Childrens Hospital, Randwick, New South Wales, Australia.
5
Telethon Institute for Child Health Research, Center for Child Health Research, University of Western Australia, Perth, Western
Australia, Australia.
2

Self-collection of buccal cells provides a noninvasive method for obtaining biologic samples for genetic analyses
in pediatric studies. Nevertheless, low yields, microbial contamination, and degradation of buccal samples present
challenges for epidemiologic studies incorporating genetic investigations. The aims of this study were to compare
the quality and yield of DNA extracted from buccal specimens with BuccalAmp swabs (Epicenter BioTechnologies,
Madison, Wisconsin) or FTA cards (Whatman, Inc., Clifton, New Jersey) and to investigate the use of wholegenome amplication (WGA) for increasing DNA yields for single nucleotide polymorphism analyses. Buccal
specimens were collected from 55 children with acute lymphoblastic leukemia and 52 control children without
acute lymphoblastic leukemia in New South Wales, Australia, in 20032004. Real-time polymerase chain reaction
was used to evaluate polymorphisms in the genes encoding the cytochrome p450 enzyme CYP3A4 (CYP3A4
A392G, also known as CYP3A4*1B) and the steroid xenobiotic receptor (SXR C25385T). Results showed that
DNA could be isolated from buccal specimens collected by use of both methods and that yields could be substantially improved with WGA without introducing genotyping error. However, DNA quality was poorer in samples
collected by BuccalAmp swabs, and the presence of polymerase chain reaction inhibitors in these samples reduced
the sensitivity of quantitative real-time PCR analysis. These ndings show that different methods for collecting buccal
samples impact on the downstream success of genetic investigations and inuence DNA quality after WGA.
DNA; epidemiologic methods; epidemiology, molecular; genome; mouth mucosa; pediatrics

Abbreviations: MDA, multiple displacement amplication; PCR, polymerase chain reaction; SNP, single nucleotide
polymorphism; WGA, whole-genome amplication.

The collection of sufficient quantities of genomic DNA is

central to the success of genetic epidemiologic studies. Historically, many of these studies have relied on DNA isolated
from whole blood cells collected via venipuncture at designated collection sites or by investigators visiting the homes
of study participants (13). Although whole blood specimens

provide high quality DNA yields, venipuncture is an invasive

procedure that is not always practical for infants and young
babies; requires specialist collection, processing, and storage
facilities; and can necessitate the added expense and inconvenience of travel for families, often resulting in low participation rates (4).

Correspondence to Dr. Lesley Ashton, Molecular Epidemiology Group, Childrens Cancer Institute Australia for Medical Research, P.O. Box 81,
Randwick, NSW 2031, Australia (e-mail: lashton@ccia.unsw.edu.au).

1260

Am J Epidemiol 2008;167:12601267

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

Received for publication June 26, 2007; accepted for publication January 14, 2008.

Buccal Swabs and Treated Cards in Molecular Epidemiology

Am J Epidemiol 2008;167:12601267

cards and returned by regular postal service without the

provision of specialized packaging (14).
Despite overcoming the many limitations associated with
collecting blood by venipuncture, buccal specimens provide
substantially smaller quantities of DNA than do whole blood
specimens. The recent development of whole-genome amplification (WGA) provides an attractive solution to this
problem. Multiple displacement amplification (MDA) is
a WGA technique that uses the Phi29 DNA polymerase
enzyme and random exonuclease-resistant primers to uniformly amplify DNA products greater than 10 kilobases in
length (11, 29, 30). This technique has been successfully
used to extend genomic DNA from human blood (31), buccal samples (31, 32), and laser-dissected neurons (33). Concordance of single nucleotide polymorphism (SNP)
genotyping between samples before and after MDA has
been reported to be as high as 99.95 percent (34). However,
recent studies have suggested that accurate genotyping may
be dependent on the amount of DNA used in MDA reactions
(35) and that MDA may introduce allele amplification bias
in buccal-derived DNA specimens (36, 37).
Although previous investigators have shown that it is
possible to use WGA to amplify buccal samples collected
with swabs or brushes (11, 32, 38) and FTA cards (26), there
have been limited reports on the application of WGA to
amplify DNA from buccal samples collected from children.
The aims of this study were to compare DNA yields obtained from buccal specimens collected from children by
use of FTA cards or BuccalAmp swabs (Epicenter BioTechnologies, Madison, Wisconsin) both before and after WGA
and to measure the impact of these methods in real-time
PCR analyses.
MATERIALS AND METHODS
Study population

Cases included 55 children aged less than 16 years who

were diagnosed with acute lymphoblastic leukemia between
2001 and 2003 at the Sydney Childrens Hospital (n 31)
and the Childrens Hospital at Westmead (n 24) in Sydney, Australia. A control panel of 52 children without acute
lymphoblastic leukemia was recruited by random digit dialing from the state of New South Wales and matched for
age and sex. The study was approved by the appropriate
institutional ethics committees, and informed consent was
collected from the parents of all children who participated.
Buccal specimens were collected from cases and controls
over a 12-month period, commencing in February 2003.
Buccal DNA collection

FTA classic indicator cards (Whatman, Inc.) were used to

collect buccal samples from all children in the study. Additional samples were collected with BuccalAmp DNA extraction kits (Epicenter BioTechnologies) from the case
children at Sydney Childrens Hospital and the control children to assess DNA yields obtained by use of an alternative
method. Collection kits were mailed to residential addresses
with instructions to collect buccal specimens by both FTA

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

Obtaining human genomic DNA from buccal epithelial

cells overcomes many of the problems associated with venipuncture, such as the requirement for a trained phlebotomist to draw blood specimens and the need for study
participants to travel to collection sites, leading to improved
participation rates (5). Self-collection of buccal specimens
is also a more acceptable method for obtaining DNA samples from young children and has been shown to improve
recruitment in remote areas or widely dispersed populations
(6, 7).
Although collection of buccal cells provides a less invasive technique to obtain DNA samples from children, this
approach does present limitations. For example, the yield
and quantity of buccal DNA obtained from buccal specimens are lower than those obtained from whole blood cells
and can vary depending on how the cells were acquired,
stored, and transported (810). In addition, studies have also
shown that buccal samples obtained from children can have
lower DNA yields than those from adults (9, 11).
Various types of buccal swabs, mouthwash, or treated
cards have been used to obtain DNA. These include scraping
the inside cheek of the mouth with a tongue depressor (12),
cotton or synthetic swab (8, 13, 14), or a cytobrush (9, 10,
15). Buccal cells can also be obtained by gargling and using
a swish and spit method with either water or commercial
mouthwash (6, 1517) or by spitting saliva directly into
a collection vial (18).
Previous studies have shown that different methods used
to isolate buccal cells can impact on both DNA yield and
quality. In adults, collection of buccal cells via the swish
and spit protocol has been shown to consistently give the
highest yields of buccal DNA with the lowest rate of fragmentation (8, 16, 19). However, this method is not always
feasible in young children because of the alcohol content of
some commercial mouthwashes and the challenge of getting
young children to spit before swallowing. Similarly, collecting buccal cells from expectorated saliva can be difficult,
as infants and young babies may have trouble being directed
to supply the saliva sample. Therefore, in young children,
swabs or cytobrushes provide the most reliable and accepted
method for collection of buccal DNA (2023).
Delays in transportation of buccal swabs returned by surface mail have also been reported to result in DNA degradation and bacterial contamination (10, 15, 18), which can
lead to failed amplification of buccal DNA (7, 12) or decreased allelic discrimination in polymerase chain reaction
(PCR) analyses (8). Flinders Technology Associates chemically treated filter papers (FTA cards) provide a novel alternative to sending buccal swabs, brushes, or buccal
collection vials through the mail. FTA cards (Whatman,
Inc., Clifton, New Jersey) are chemically treated filter papers designed for the collection and room temperature storage of biologic samples for molecular analyses; these cards
protect DNA and are impregnated with denaturants that
guard against oxidation, nuclease and ultraviolet damage,
and both bacterial and fungal degradation. Recent studies
have shown that FTA cards provide a convenient and safe
method for the collection and storage of DNA from whole
blood, tissue, buccal, viral, and bacterial specimens (2428).
Moreover, buccal specimens can be self-collected with FTA

1261

1262 Beckett et al.

Multiplex PCR analysis

Multiplex PCR analysis was used to assess the quality of

DNA in a subset of buccal samples both before and after
WGA. DNA products of 100, 200, 300, 400, and 600 base
pairs were amplified from four human control genes as described elsewhere (37). The DNA template was either 1 ll of
BuccalAmp solution, one FTA hole-punch, or 1 ll of wholegenome amplified solution. All PCR reactions were analyzed
with a GeneAmp 9700 PCR system (Applied Biosystems,
Foster City, California). PCR products were visualized using
polyacrylamide gels stained with SYBR Safe DNA gel stain
(Invitrogen, Victoria, Australia).
Real-time PCR analyses

Human genomic DNA concentrations were quantified by

real-time PCR for the beta-actin housekeeping gene with the
following primer and probe sequences: forward 5#-TCACCCACACTGTGCCCATCTACGA-3#; reverse 5#-CAGCGGAACCGCTCATTGCCAATGG-3#; and 5#-VIC-ATGCCCT
CCCCCATGCCATCCTGCGT-TAMRA-3#. Standard dilution curves of human placental DNA (100 ng, 10 ng, 1 ng,
0.1 ng, and 0.01 ng) were generated and used to quantify the
DNA concentration in each sample. The lower detection
limit of the quantitative real-time PCR assay was 0.01 ng/ll.
SNPs in genes encoding the cytochrome p450 enzyme
CYP3A4 (CYP3A4 A392G, also known as CYP3A4*1B)
and the steroid xenobiotic receptor (SXR C25385T) were
determined by use of real-time PCR, in an ABI 7000
Real-Time system (Applied Biosystems). Sequences for
the CYP3A4 A392G primers and probes were obtained from

the SNP500 database (39). SNP specific primer sequences

used for SXR C25385T included the following: forward 5#AACTGTGGTCATTTTTTGGCAA-3# and reverse 5#-ACCACGATTGAGCAAACAGGT-3#, while the probes were
5#-FAM-CCCAGGTTCTCTTTTC-MGBNFQ-3# and 5#VIC-CCCAGGTTTTCTTTTC-MGBNFQ-3#. Primer concentrations varied, from 5 ng/ll (beta-actin) and 15 ng/ll
(SXR C25385T) to 90 ng/ll (CYP3A4 A392G). All samples
were analyzed in duplicate.
Whole-genome amplication

The GenomiPhi kit (Amersham Biosciences, Piscataway,

New Jersey) was used to amplify genomic DNA from
BuccalAmp DNA solutions and FTA hole-punches. Wholegenome amplification was performed with 1 ll of BuccalAmp
DNA solution or one FTA hole-punch as recommended by
the manufacturers. The WGA reaction mix was diluted 1:5
with DNase- and RNase-free distilled water (Invitrogen)
before DNA quantification.
DNA yields from all BuccalAmp specimens were quantified both before and after WGA by real-time PCR for the
beta-actin housekeeping gene. DNA from 1.2-mm FTA
hole-punches was not quantified before WGA, as real-time
quantification of DNA present on the same hole-punch prior
to WGA can reduce the sensitivity of the WGA reaction,
limit DNA yields, or result in allelic imbalance (failure to
amplify both alleles) (35). DNA quantification prior to
WGA on duplicate hole-punches was not feasible, as the
amount of DNA present on any two punches from the same
card can vary, because of the uneven distribution of DNA
immobilized on the card (14).
Quantication of PCR inhibition

A subset of DNA samples collected with BuccalAmp

swabs were examined for their potential inhibitory effect in
quantitative real-time PCR analysis. Observed DNA levels
quantified from a dilution series of BuccalAmp DNA were
compared with those expected by extrapolating from the
undiluted sample. PCR inhibition was further examined by
quantifying a previously determined amount of placental
DNA (50 ng) in the presence or absence of BuccalAmp DNA.
Statistical analysis

Wilcoxons rank-sum tests were used to compare median

DNA yields from cases and controls. Correlations between
DNA yields before and after WGA or age at DNA collection
were examined by use of Spearmans rank correlation tests.
All results were analyzed with STATA, version 8.2, statistical
software (StataCorp LP, College Station, Texas).
RESULTS
DNA yields before and after WGA

DNA yields were analyzed in samples from 31 cases from

Sydney Childrens Hospital and 52 controls who provided
buccal samples collected by both methods. On average,
Am J Epidemiol 2008;167:12601267

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

cards and BuccalAmp swabs based on the manufacturers

recommendations. All buccal specimens were collected on
the same day, as investigations in our laboratory had previously shown that two swabs from the same cheek on the
same day did not decrease the total DNA yields collected or
reduce the DNA yields obtained using either method (data
not shown).
Collection of buccal specimens was carried out by parents
either at home or during a hospital clinic visit. Samples
collected at home were posted in certified prepaid postage
bags, tracked electronically through the Australian postage
system, and shipped to the laboratory for processing within
24 hours, while samples collected or returned to staff during
a hospital visit were processed immediately. Neither method
was found to induce bleeding in the mouths of the children
sampled. Although it was not possible to accurately quantify
the total amount of DNA present on a FTA card, all FTA
cards were scored for sample coverage and defined as
having <25 percent, 2549 percent, 5074 percent, or 75
100 percent coverage.
BuccalAmp samples were processed according to the
manufacturers instructions and stored at
20
C. FTA cards
were stored at room temperature in sterile airtight containers
with silicon desiccant. A 1.2-mm disc was punched from
FTA cards by use of a Harris Micro-Punch (Whatman, Inc.)
and processed according to the manufacturers instructions
immediately before PCR analyses or WGA.

Buccal Swabs and Treated Cards in Molecular Epidemiology

1263

TABLE 1. DNA yields before and after whole-genome amplication in cases and
controls who returned both BuccalAmp swabs* and FTA cards*, New South Wales,
Australia, 20032004
Cases (n 31)
DNA sample

Median DNA
yields
(ng/100 ll)

Controls (n 52)

Interquartile
range

Median DNA
yields
(ng/100 ll)

Interquartile
range

p valuey

BuccalAmp solutions
Before WGAz
After WGA

34.1123.5

26.3

6.169.0

0.007

350

79.1

120.0596.5

78.8

8.8297.0

0.002

240

116.5424.5

134.3371.0

0.443

FTA cards
After WGA

257

* BuccalAmp swabs (Epicenter BioTechnologies, Madison, Wisconsin); FTA cards (Whatman,

Inc., Clifton, New Jersey).
y Wilcoxons rank-sum test p value.
z WGA, whole-genome amplication.
The WGA yield is from 1 ll of BuccalAmp solution or a 1.2-mm FTA card hole-punch.

analyses as they were considered to have an inadequate

template for the PCR reactions. Of the 79 BuccalAmp samples analyzed (30 cases and 49 controls), 1.3 percent failed
to produce a real-time PCR product before WGA in either
the CYP3A4 or SXR genotype analyses, while 6.3 percent of
all samples failed to produce a real-time PCR product after
WGA. No difference was observed in the failure rate between cases and controls, while 100 percent concordance
was observed in all samples successfully genotyped. Genotype analyses carried out on 55 cases and 52 controls who
supplied FTA cards showed that DNA collected with FTA
cards consistently produced genotype results with 100 percent concordance before and after WGA.

Genotyping concordance in buccal samples before and

after WGA

BuccalAmp specimens with detectable DNA (0.01 ng/

ll) were genotyped for CYP3A4 A392G and SXR C25385T
gene variants. One case and three controls with no detectable DNA prior to WGA were excluded from genotyping
Am J Epidemiol 2008;167:12601267

FIGURE 1. Correlation between BuccalAmp DNA yields obtained

before and after whole-genome amplication in 79 children, New
South Wales, Australia, 20032004. BuccalAmp (Epicenter BioTechnologies, Madison, Wisconsin).

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

returned FTA cards from both cases and controls showed

40 percent coverage, providing approximately 160 holepunches per card. Although the total DNA present on any
one FTA card could not be accurately determined, we estimate that the DNA present on 160 punches would collectively allow for 160 or more molecular applications
including multiplex PCR reactions or WGA. Table 1 shows
median DNA yields obtained before and after WGA for
samples collected with FTA and BuccalAmp techniques.
Overall, total BuccalAmp DNA yields ranged from levels
below detection (one case and three control samples) to 2.6
lg. Median DNA yields were significantly higher in BuccalAmp samples collected from cases compared with controls
both before (79.1 vs. 26.3 ng; p 0.007) and after (350.0 vs.
78.8 ng; p 0.002) WGA, while no significant difference
was observed for DNA yields obtained with FTA cards in
cases and controls after WGA (240.0 vs. 257.0 ng; p
0.443). Figure 1 shows the relation between the amount of
BuccalAmp DNA in a single WGA reaction and the total
DNA yield, where an increased DNA yield from WGA of
BuccalAmp swabs was strongly correlated to a higher concentration of input DNA (p < 0.001). Figure 2 illustrates the
absence of any significant correlation between age at collection and DNA yield from BuccalAmp swabs prior to
WGA (p 0.855).
Multiplex PCR analyses showed variations in the quality
of buccal DNA obtained with different methods. DNA from
FTA cards produced clearer, well-defined PCR products before and after WGA compared with BuccalAmp samples,
which appeared to yield more fragmented DNA (figure 3).

1264 Beckett et al.

Inhibition of real-time PCR analyses

Our investigations showed that quantitative real-time

PCR analysis for the beta-actin gene was inhibited by
DNA samples collected by use of BuccalAmp technology.
As seen in figure 4, part A, the observed DNA concentration
measured by real-time PCR analysis approached expected
levels after the BuccalAmp sample was diluted twofold or
more. Inhibition of real-time PCR analysis was also observed when a known quantity of placental DNA (50 ng)
was amplified in the presence of buccal DNA (figure 4, part
B). This investigation revealed that the amplification curve
of the quantitative real-time PCR reaction peaked at a later
cycle number in the presence of buccal DNA, compared
with when the sample was analyzed without buccal DNA.
DISCUSSION

Results from our study show that the overall quality of

buccal DNA collected with FTA cards was superior to that
collected with BuccalAmp swabs and that FTA cards overcame the problems presented by PCR inhibitors in buccal
specimens. Our investigations also demonstrated that yields
of buccal DNA collected from children could be substantially improved with WGA and that this process did not
generate genotyping errors in the real-time PCR assays examined. Nevertheless, successful PCR analyses both before
and after WGA appeared to be dependent on the quality and
quantity of buccal DNA.
Previous studies have reported variable yields of buccal
DNA obtained from children, ranging from 1.7 lg to 28.3
lg (7, 11). To some extent, these variations are likely to
reflect subtle differences in the methodologies used to collect and isolate DNA from buccal specimens, as well as the

FIGURE 3. Variation in DNA quality obtained before and after

whole-genome amplication (WGA) in samples obtained from the
same individuals by either BuccalAmp (A) or FTA card (B) technology,
New South Wales, Australia, 20032004. MWT, molecular weight
(base pairs). DNA yields from these samples are representative of
those observed in our study. BuccalAmp (Epicenter BioTechnologies,
Madison, Wisconsin); FTA cards (Whatman, Inc., Clifton, New
Jersey).

laboratory techniques used to quantify DNA. Although the

DNA yields isolated from BuccalAmp swabs were lower
than those previously reported, the use of quantitative
real-time PCR analyses in our study provided a sensitive
method to specifically measure human DNA compared with
alternative methods, such as spectrophotometry or fluorometry, which have limited capacity to discriminate between
human and bacterial DNA. Hence, the yields observed in
our study accurately reflect the quantity of human genomic
DNA collected.
Our investigations also showed that DNA yields from
BuccalAmp swabs were substantially higher for cases than
controls. This may have been because parents of children
with cancer were more motivated to adhere to the instructions provided and to return their buccal samples as soon as
possible compared with parents of control children. Alternatively, the delay between collection and processing of
BuccalAmp samples from control families who did not attend the clinic may have resulted in an increased bacterial
content or denatured DNA compared with the samples from
cases that were predominantly returned in the clinic and
processed within 12 hours of collection. Although it was
Am J Epidemiol 2008;167:12601267

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

FIGURE 2. Correlation between age at collection and the BuccalAmp DNA yields obtained from 79 children, New South Wales,
Australia, 20032004. BuccalAmp (Epicenter BioTechnologies,
Madison, Wisconsin).

Buccal Swabs and Treated Cards in Molecular Epidemiology

not possible to measure the exact length of time from when

the samples were collected and their arrival at the laboratory, five BuccalAmp swabs collected from control children
did show visible evidence of microbial contamination on
arrival in the laboratory, suggesting delayed transportation
after sample collection. Hence, DNA degradation appears to
be a more feasible explanation and is consistent with our
observation of poorer quality DNA in samples collected with
BuccalAmp swabs compared with FTA cards (figure 3).
Although it is difficult to accurately assess the total DNA
yield from FTA cards, we observed no visible difference in
the level of sample coverage on FTA indicator cards returned by cases and controls. Moreover, DNA yields from
FTA cards after WGA did not differ between cases and
controls (table 1). This observation is consistent with previous reports that DNA in buccal samples immobilized on
FTA cards remains stable for several years at room temperature (27, 40). Hence, DNA isolated from FTA cards was
Am J Epidemiol 2008;167:12601267

unlikely to be affected by any delay in returning samples to

the laboratory.
Previous studies have reported that inhibitory factors in
biologic specimens can influence DNA template amplification and PCR analyses with slight to moderate inhibition of
target DNA resulting in underestimates of DNA concentration, while severe inhibition can lead to false negative results or assay failure unless additional purification steps are
undertaken (7, 12, 41, 42). Over recent years, a wide range
of inhibitors have been reported. However, the identification
of specific inhibitors and their mode of action remain to be
fully determined (43). In our study, the lack of PCR inhibition when the placental DNA was analyzed in the absence of
buccal cells and saliva suggested that the observed PCR
inhibition was caused by an unidentified contaminant in
the BuccalAmp samples, possibly an unknown bacterial
product or salivary enzymes. We hypothesize that the antimicrobial pretreatment of the FTA cards and repeated
washes of each hole-punch prior to PCR amplification remove inhibitors from FTA cards. Despite the presence of
PCR inhibitors in buccal samples collected with BuccalAmp technology, we were able to minimize PCR inhibition
by diluting these samples prior to analyses (figure 4, part A).
However, in some cases, dilution of the buccal specimen
decreased the overall sensitivity of PCR assays by reducing
the target sequence to levels below the detectable limit of
the quantitative PCR assay.
Other investigators have shown that, when DNA template
concentrations are low, stochastic fluctuations in PCR
analyses can result in unequal sampling of heterozygous
loci, leading to false inference of a homozygous genotype
in samples with a low DNA template concentration (36,
4446). Although we observed a higher failure rate in
PCR analyses of samples collected with BuccalAmp swabs,
100 percent concordance was attained for all successful
genotyping of DNA samples from buccal specimens collected by use of either FTA cards or BuccalAmp swabs,
both before and after WGA.
Several studies have investigated the use of WGA for
amplifying human genomic DNA in different types of biospecimens and have shown that the genomic coverage of the
amplified products is generally good with a low rate of error
(29, 32, 47, 48). Although WGA amplification has been
reported to result in consistent DNA yields regardless of
the starting DNA amounts (29), our results showed a positive
correlation between increasing amounts of starting DNA
template and higher DNA yields obtained after WGA. This
finding is consistent with a previous report showing that
increasing amounts of DNA template used in the WGA reaction can influence the yields from WGA (35). Moreover,
trace amounts of bacterial contaminants or salivary enzymes
in DNA samples derived from BuccalAmp swabs may also
compete with the input template in WGA reaction, particularly when DNA template concentrations are low.
Prior investigations have also reported that the performance of WGA is dependent on the quality of DNA template used in the reaction and that poor DNA template can
decrease the accuracy of genotyping (35, 49). In our study,
multiplex PCR analyses showed that the quality of DNA
collected with BuccalAmp swabs was somewhat poorer

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

FIGURE 4. DNA sample inhibition of real-time polymerase chain

reaction in BuccalAmp samples from children, New South Wales,
Australia, 20032004. A, DNA quantication after dilution of BuccalAmp sample. Expected DNA yield was extrapolated from DNA yield
measured in the undiluted buccal specimen. B, real-time quantication of placental DNA in the presence or absence of the BuccalAmp
specimen. BuccalAmp (Epicenter BioTechnologies, Madison,
Wisconsin).

1265

1266 Beckett et al.

compared with the quality of DNA collected with FTA cards

both before and after WGA (figure 3). This observation was
further supported by the higher rate of genotype failure
observed in specimens collected by use of BuccalAmp
swabs compared with FTA cards. Nevertheless, this observation may have been limited to some extent by the small
number of samples and gene variants examined in our study.
Overall, our investigations show that both FTA cards and
BuccalAmp swabs provide useful tools for collecting buccal
specimens from children and that the low yields of genomic
DNA collected with these methods can be substantially improved after WGA. However, our findings show that FTA
cards present a more robust method for collecting DNA
samples for genetic epidemiologic studies, as they provide
higher quality DNA and reduce potential inhibition of
downstream investigations.

10.
11.

12.
13.

14.
15.

This project was partially supported by funding from the

National Health and Medical Research Council, Australia
(grant 350908). The Childrens Cancer Institute Australia
for Medical Research is affiliated with the University of
New South Wales and Sydney Childrens Hospital.
The authors would like to acknowledge the helpful
advice provided by Professors Michelle Haber and Murray
Norris on early versions of the manuscript and Stewart
Smith for his technical advice.
Conflict of interest: none declared.

16.
17.
18.

19.
REFERENCES
1. Austin MA, Ordovas JM, Eckfeldt JH, et al. Guidelines of the
National Heart, Lung, and Blood Institute Working Group on
Blood Drawing, Processing, and Storage for Genetic Studies.
Am J Epidemiol 1996;144:43741.
2. Steinberg K, Beck J, Nickerson D, et al. DNA banking for
epidemiologic studies: a review of current practices. Epidemiology 2002;13:24654.
3. Steinberg KK, Sanderlin KC, Ou CY, et al. DNA banking in
epidemiologic studies. Epidemiol Rev 1997;19:15662.
4. Dlugos DJ, Scattergood TM, Ferraro TN, et al. Recruitment
rates and fear of phlebotomy in pediatric patients in a genetic
study of epilepsy. Epilepsy Behav 2005;6:4446.
5. Harty LC, Shields PG, Winn DM, et al. Self-collection of oral
epithelial cell DNA under instruction from epidemiologic interviewers. Am J Epidemiol 2000;151:199205.
6. Le Marchand L, Lum-Jones A, Saltzman B, et al. Feasibility of
collecting buccal cell DNA by mail in a cohort study. Cancer
Epidemiol Biomarkers Prev 2001;10:7013.
7. Witso E, Stene LC, Paltiel L, et al. DNA extraction and HLA
genotyping using mailed mouth brushes from children. Pediatr
Diabetes 2002;3:8994.
8. Neuhaus T, Geisen G, Bolt HM, et al. Reliability of noninvasively acquired human genomic DNA as a substrate for
real-time PCR-assisted analysis of genetic polymorphisms.
Arch Toxicol 2004;78:3906.
9. Saftlas AF, Waldschmidt M, Logsden-Sackett N, et al. Optimizing buccal cell DNA yields in mothers and infants for

20.

21.
22.
23.
24.
25.

26.

27.

Am J Epidemiol 2008;167:12601267

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

ACKNOWLEDGMENTS

human leukocyte antigen genotyping. Am J Epidemiol

2004;160:7784.
Walker AH, Najarian D, White DL, et al. Collection of genomic
DNA by buccal swabs for polymerase chain reaction-based
biomarker assays. Environ Health Perspect 1999;107:51720.
Zheng S, Ma X, Buffler PA, et al. Whole genome amplification
increases the efficiency and validity of buccal cell genotyping
in pediatric populations. Cancer Epidemiol Biomarkers Prev
2001;10:697700.
Moore L, Wiencke JK, Eng C, et al. Evaluation of buccal
cell collection protocols for genetic susceptibility studies.
Biomarkers 2001;6:44854.
Freeman B, Smith N, Curtis C, et al. DNA from buccal swabs
recruited by mail: evaluation of storage effects on long-term
stability and suitability for multiplex polymerase chain
reaction genotyping. Behav Genet 2003;33:6772.
Milne E, van Bockxmeer FM, Robertson L, et al. Buccal DNA
collection: comparison of buccal swabs with FTA cards.
Cancer Epidemiol Biomarkers Prev 2006;15:81619.
Garcia-Closas M, Egan KM, Abruzzo J, et al. Collection of
genomic DNA from adults in epidemiological studies by
buccal cytobrush and mouthwash. Cancer Epidemiol
Biomarkers Prev 2001;10:68796.
Cozier YC, Palmer JR, Rosenberg L. Comparison of methods
for collection of DNA samples by mail in the Black Womens
Health Study. Ann Epidemiol 2004;14:11722.
London SJ, Xia J, Lehman TA, et al. Collection of buccal cell
DNA in seventh-grade children using water and a toothbrush.
Cancer Epidemiol Biomarkers Prev 2001;10:122730.
Rylander-Rudqvist T, Hakansson N, Tybring G, et al.
Quality and quantity of saliva DNA obtained from the selfadministrated oragene methoda pilot study on the cohort
of Swedish men. Cancer Epidemiol Biomarkers Prev 2006;
15:17425.
King IB, Satia-Abouta J, Thornquist MD, et al. Buccal cell
DNA yield, quality, and collection costs: comparison of
methods for large-scale studies. Cancer Epidemiol Biomarkers
Prev 2002;11:11303.
Bennett LC, Kraemer R, Liechti-Gallati S. Buccal cell DNA
analysis in premature and term neonates: screening for mutations
of the complete coding region for the cystic fibrosis transmembrane conductance regulator. Eur J Pediatr 2000;159:99102.
Gavriel G, Modi N, Stanier P, et al. Neonatal buccal cell
collection for DNA analysis. (Letter). Arch Dis Child
Fetal Neonatal Ed 2005;90:F187.
Parad RB. Buccal cell DNA mutation analysis for diagnosis
of cystic fibrosis in newborns and infants inaccessible to
sweat chloride measurement. Pediatrics 1998;101:8515.
Yoon PW, Rasmussen SA, Lynberg MC, et al. The National
Birth Defects Prevention Study. Public Health Rep 2001;
116(suppl 1):3240.
Devost NC, Choy FY. Mutation analysis of Gaucher disease
using dot-blood samples on FTA filter paper. Am J Med
Genet 2000;94:41720.
Galaal K, Meirovitz M, Hussain R, et al. The feasibility of
storing ovarian tumor cells on databasing paper: establishing
a library of ovarian cancer DNA. Int J Gynecol Cancer
2007;17:94100.
He H, Argiro L, Dessein H, et al. Improved technique that
allows the performance of large-scale SNP genotyping on
DNA immobilized by FTA technology. Infect Genet Evol
2007;7:12832.
Li CC, Beck IA, Seidel KD, et al. Persistence of human immunodeficiency virus type 1 subtype B DNA in dried-blood
samples on FTA filter paper. J Clin Microbiol 2004;42:38479.

Buccal Swabs and Treated Cards in Molecular Epidemiology

Am J Epidemiol 2008;167:12601267

38. Moore LE, Bergen AW, Haque KA, et al. Whole genome
amplification of buccal cytobrush DNA collected for molecular epidemiology studies. Biomarkers 2007;12:30312.
39. Packer BR, Yeager M, Burdett L, et al. SNP500Cancer:
a public resource for sequence validation, assay development,
and frequency analysis for genetic variation in candidate
genes. Nucleic Acids Res 2006;34:D61721.
40. Smith LM, Burgoyne LA. Collecting, archiving, and
processing DNA from wildlife samples using FTA databasing
paper. (Electronic article). BMC Ecol 2004;4:4.
41. de Lomas JG, Sunzeri FJ, Busch MP. False-negative results
by polymerase chain reaction due to contamination by glove
powder. Transfusion 1992;32:835.
42. Lienert K, Fowler JC. Analysis of mixed human/microbial
DNA samples: a validation study of two PCR AMP-FLP
typing methods. Biotechniques 1992;13:27681.
43. Wilson IG. Inhibition and facilitation of nucleic acid amplification. Appl Environ Microbiol 1997;63:374151.
44. Handt O, Krings M, Ward RH, et al. The retrieval of ancient
human DNA sequences. Am J Hum Genet 1996;59:36876.
45. Takahashi M, Kato Y, Mukoyama H, et al. Evaluation of five
polymorphic microsatellite markers for typing DNA from
decomposed human tissuescorrelation between the size of
the alleles and that of the template DNA. Forensic Sci Int
1997;90:19.
46. Kontanis EJ, Reed FA. Evaluation of real-time PCR amplification efficiencies to detect PCR inhibitors. J Forensic Sci
2006;51:795804.
47. Barker DL, Hansen MS, Faruqi AF, et al. Two methods of
whole-genome amplification enable accurate genotyping
across a 2320-SNP linkage panel. Genome Res 2004;14:
9017.
48. Lasken RS, Egholm M. Whole genome amplification: abundant supplies of DNA from precious samples or clinical
specimens. Trends Biotechnol 2003;21:5315.
49. Sun G, Kaushal R, Pal P, et al. Whole-genome amplification:
relative efficiencies of the current methods. Leg Med (Tokyo)
2005;7:27986.

Downloaded from http://aje.oxfordjournals.org/ by guest on March 6, 2016

28. Guio H, Okayama H, Ashino Y, et al. Method for efficient

storage and transportation of sputum specimens for
molecular testing of tuberculosis. Int J Tuberc Lung Dis
2006;10:90610.
29. Dean FB, Hosono S, Fang L, et al. Comprehensive human
genome amplification using multiple displacement amplification. Proc Natl Acad Sci U S A 2002;99:52616.
30. Haberstick BC, Smolen A. Genotyping of three single
nucleotide polymorphisms following whole genome preamplification of DNA collected from buccal cells. Behav Genet
2004;34:5417.
31. Holbrook JF, Stabley D, Sol-Church K. Exploring whole
genome amplification as a DNA recovery tool for molecular
genetic studies. J Biomol Tech 2005;16:12533.
32. Hosono S, Faruqi AF, Dean FB, et al. Unbiased whole-genome
amplification directly from clinical samples. Genome Res
2003;13:95464.
33. Pamphlett R, Heath PR, Holden H, et al. Detection of
mutations in whole genome-amplified DNA from lasermicrodissected neurons. J Neurosci Methods 2005;147:657.
34. Tzvetkov MV, Becker C, Kulle B, et al. Genome-wide singlenucleotide polymorphism arrays demonstrate high fidelity of
multiple displacement-based whole-genome amplification.
Electrophoresis 2005;26:71015.
35. Bergen AW, Qi Y, Haque KA, et al. Effects of DNA mass
on multiple displacement whole genome amplification and
genotyping performance. (Electronic article). BMC Biotechnol 2005;5:24.
36. Liang X, Trentham-Dietz A, Titus-Ernstoff L, et al. Wholegenome amplification of oral rinse self-collected DNA in
a population-based case-control study of breast cancer. Cancer
Epidemiol Biomarkers Prev 2007;16:161014.
37. van Dongen JJ, Langerak AW, Bruggemann M, et al. Design
and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene
recombinations in suspect lymphoproliferations: report of the
BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia
2003;17:2257317.

1267