Available online at www.sciencedirect.

com

Forensic Science International: Genetics 2 (2008) 226–230

www.elsevier.com/locate/fsig

Comparison of five DNA quantification methods

Karsten Nielsen a,*, Helle Smidt Mogensen a, Johannes Hedman b,
Harald Niederstätter c, Walther Parson c, Niels Morling a
a
Section of Forensic Genetics, Department of Forensic Medicine, Faculty of Health Sciences, University of Copenhagen, Denmark
b
Swedish National Laboratory of Forensic Science, Linköping, Sweden
c
Institute of Legal Medicine, Innsbruck Medical University, Innsbruck, Austria
Received 24 December 2007; received in revised form 11 February 2008; accepted 26 February 2008

Abstract
Six commercial preparations of human genomic DNA were quantified using five quantification methods: UV spectrometry, SYBR-Green dye
staining, slot blot hybridization with the probe D17Z1, QuantifilerTM Human DNA Quantification kit and RB1 rt-PCR. All methods measured
higher DNA concentrations than expected based on the information by the manufacturers. UV spectrometry, SYBR-Green dye staining, slot blot
and RB1 rt-PCR gave 39, 27, 11 and 12%, respectively, higher concentrations than expected based on the manufacturers’ information. The DNA
preparations were quantified using the QuantifilerTM Human DNA Quantification kit in two experiments. The measured DNA concentrations with
Quantifiler were 125 and 160% higher than expected based on the manufacturers’ information. When the QuantifilerTM human DNA standard (Raji
cell line) was replaced by the commercial human DNA preparation G147A (Promega) to generate the DNA standard curve in the QuantifilerTM
Human DNA Quantification kit, the DNA quantification results of the human DNA preparations were 31% higher than expected based on the
manufacturers’ information. The results indicate a calibration problem with the QuantifilerTM human DNA standard for its use with the
QuantifilerTM Human DNA Quantification kit. The possible reasons for the problem are discussed and a solution is suggested. The results
emphasise the need for standard reference DNA material and standard methods for DNA quantification.
# 2008 Published by Elsevier Ireland Ltd.

Keywords: Human DNA; Quantification; Quantifiler; Slot blot; SYBR-Green; UV spectrometry

1. Introduction PCR quantification method of nuclear and mitochondrial DNA

Reliable quantification of human DNA in a forensic genetic
sample makes it possible to adjust the concentration of template
DNA used for STR analysis in order to obtain optimal PCR
reactions and STR typing results with as small amounts of DNA
as possible. Various methods have been developed for
quantification of human DNA in forensic samples, e.g. a slot
blot method using a 40 bp long DNA-probe designed to match
the human specific alpha satellites that are present in 500–1000
copies on chromosome 17 (D17Z1) [1,2]. The AluQuant Human
Quantification System [3] includes an enzyme linked DNA probe
that is complementary to human specific Alu sequences, which
are repetitive elements with a consensus sequence of approxi-
mately 280 bp [4]. Alu sequences are found in 500,000–
1,000,000 copies per human genome. A human specific real-time
* Corresponding author. Tel.: +45 3532 6110; fax: +45 3532 6120.
E-mail address: karsten.nielsen@forensic.ku.dk (K. Nielsen).
based on the TaqMan assay validated for forensic samples was
reported by Andréasson et al. [5]. The nuclear target was the
human retinoblastoma susceptibility gene (RB1) which is a
single copy gene located on chromosome 13. Based on this assay,
Niederstätter et al. [6] developed a modular real-time (rt) PCR
system for absolute quantification of human nuclear (n) and/or
mitochondrial (mt) DNA with an optional internal amplification
positive control. Others have developed real-time PCR assays
based on multi copy number DNA targets like the Alu sequences
[7,8]. The advantage of using a multi copy number target DNA is
the potential of developing more sensitive assays, but the
variation of the DNA quantification results from person to person
may be higher due to variations in the numbers of copies of the
DNA targets. Applied Biosystems (Foster City, CA) introduced
the QuantifilerTM Human DNA Quantification kit for forensic
samples based on the single copy gene hTERT (human
telomerase reverse transcriptase) that is located on chromosome
5 [9]. This assay has an internal positive control to ascertain PCR
amplification.

1872-4973/$ – see front matter # 2008 Published by Elsevier Ireland Ltd.

doi:10.1016/j.fsigen.2008.02.008
 K. Nielsen et al. / Forensic Science International: Genetics 2 (2008) 226–230
The DNA quantification methods mentioned have in
common that that they are indirect measures of the DNA
concentration in a sample. A standard curve based on various
concentrations of known concentrations of reference DNA is
needed for the interpretation of test results. Standardized, high
quality reference DNA is needed to ensure reproducible DNA
quantification results within and between laboratories. UV
spectrometry directly detects the DNA and RNA in a sample.
The absorbance at 260 nm of 1 in a 1 cm path-length cuvette
approximately equals a DNA concentration of 50 ng/ml for
double stranded DNA (dsDNA) and 40 ng/ml for single
stranded DNA (ssDNA) and RNA [10]. UV spectrometry,
however, is not suitable for quantification of DNA in forensic
samples because UV spectrometry is not sensitive enough to
detect the low amounts of DNA in many crime case stains. The
DNA concentration must be at least 3 ng/ml in order to give
reliable results with UV spectrometry. In many forensic stain
samples, the DNA concentration is below 1 ng/ml. Forensic
stain DNA samples may be contaminated with UV-absorbing
contaminants like proteins and phenol, which may interfere
with the quantification results. Furthermore, UV spectrometry
measures all human and non-human DNA as well as RNA in a
sample, which reduces the predictive value for successful
DNA-profiling of human STRs.
A number of fluorescent dyes interact with DNA. The dye,
e.g. may on binding to DNA become fluorescent or its
emission spectrum can change from that in its unbound state.
Dyes can have different binding affinities to dsDNA, ssDNA
and RNA. PicoGreen, Hoechst dyes and SYBR-Green
preferentially detects dsDNA whereas OliGreen preferen-
tially detects ssDNA [11,12]. DNA quantification based on
fluorescent dyes like SYBR-Green is very sensitive.
However, fluorometric methods are not well suited for
DNA quantification if the origin of the DNA is uncertain,
because all kinds of DNA from animals, plants, bacterias,
etc. are detected in a sample. For a more detailed review of
methods used for quantification of DNA in forensic genetics,
see Nicklas and Buel [12].
Forensic laboratories certified under the EN/ISO/IEC 17025
standard [13] shall ensure that: ‘‘purchased manufacturers and
reagents and consumable materials that affect the quality of
tests and/or calibrations are not used until they have been
inspected or otherwise verified as complying with standard
specifications or requirements defined in the methods for the
tests and/or calibrations concerned’’. The results presented here
are part of the validation of the QuantifilerTM Human DNA
Quantification kit at the Section of Forensic Genetics,
Table 1
Commercial DNA preparations used in this study
DNA preparation Manufacturer Source
D3160 Sigma Placenta
D3035 Sigma Placenta
Human genomic DNA Roche Blood (buffy-coat)
G147A Promega Blood from multiple anonymous donors
G152A Promega Blood from multiple anonymous donors
QuantifilerTM DNA ABI Raji cell line
227
Department of Forensic Medicine, Faculty of Health Sciences,
University of Copenhagen, Denmark.
2. Materials and methods
2.1. UV spectrometry and SYBR-Green fluorometry
UV spectrometry was performed in a PerkinElmer UV/VIS
Lambda-16 spectrometer (PerkinElmer) with double beam. The
blank solution was distilled water.
SYBR-Green fluometry was performed in a Lightcycler
(Roche) with excitation: 494 nm and emission: 521 nm. A
standard curve was made from known amounts of calf thymus
DNA (Amersham Pharmacia Biotech).
Six commercial human DNA preparations (Table 1) were
subjected to agarose gel electrophoresis (1.2% (V/cm) 2 h) to
test for the presence of RNA and to test if the DNA preparations
were degraded. The DNA preparations were diluted to 3, 5 and
10 ng/ml before quantification with UV spectrometry and dye
staining with SYBR-Green (Table 2). Samples were scanned
from 200 to 350 nm using UV spectrometry to reveal UV
absorbing contaminants. The DNA concentration was calcu-
lated as: DNA concentration = absorbance at 260 nm  dilu-
dilution factor  50 ng/ml.
2.2. Comparison of slot blot and QuantifilerTM H
The DNA preparations were diluted from 50 to 0.023 ng/ml
before quantification with slot blot hybridization using the D17Z1
DNA probe (Table 2) and QuantifilerTM Human DNA
Quantification kit (QuantifilerTM H) following the manufacturer’s
instructions (Table 2). K562 DNA (Promega) was used as
reference DNA for the slot blot method. At the Section of Forensic
Genetics in Copenhagen, four replicates of each of the DNA
preparations were quantified with QuantifilerTM H using both the
QuantifilerTM DNA and G147A as reference DNA. Some of the
DNA preparations were quantified in the laboratories in Innsbruck
and Linköping. The DNA was diluted to 5 or 10 ng/ml before
shipping in PCR clean Safe-Lock 1.5 ml Eppendorf tubes by
ordinary mail. Three or four replicates of each DNA preparation
was quantified using RB1 rt-PCR with QuantifilerTM DNA as
reference DNA at the Innsbruck Institute of Legal Medicine
(GMI, Table 2). The DNA preparations G147A and G152A were
additionally quantified with RB1 rt-PCR using the GMI DNA
standard [6]. Two replicates of each D3160, D3035 and HG DNA
were quantified at the Swedish National Laboratory of Forensic
Science, Linköping, using the QuantifilerTM H kit (Table 2).
Source Sex DNA concentration (ng/ml)
Single Male 100
Single Female 140
Multiple Mix 200
Multiple Male 177
Multiple Female 109
Single Male 200
228 K. Nielsen et al. / Forensic Science International: Genetics 2 (2008) 226–230

Table 2
DNA quantification methods used in this study
Quantification method Instrument Principle Manufacturer References
a
UV spectrometry PerkinElmer Lambda 16 OD-260 – [10]
Dye staining Lightcycler (Roche) SYBR-Green I Bie & Berntsen [11]
Slot blot hybridization – D17Z1 DNA probe DNA-technology [1,2]
TaqMan real-time PCR ABI 7000 SDS, ABI 7300 SDS QuantifilerTM Human ABI [9]
TaqMan real-time PCR ABI 7700 SDS RB1 rt-PCR ABI [5,6]
a
Samples were scanned from 200 to 350 nm.

Table 3
UV spectrometry and SYBR-Green measurements
DNA preparation DNA stock (ng/ml)a OD-260 (ng/ml) SYBR-Green (ng/ml)
b
D3160 100 182  10 137  45
D3035 140 181  15 161  50
Human genomic DNA 200 446  18 300c
G147A 177 170  3 212
G152A 109 107  32 163
QuantifilerTM DNA 200 206  10 182  2
a
Manufacturers’ information.
b
Mean  Standard deviation.
c
Average of two measurements.

The data were log transformed and analyzed with ANOVA
using the free software package R (http://www.r-project.org).
Because of incomplete experimental design, the ANOVA was
performed as three pair wise comparisons of methods: (1)
QuantifilerTM H (Copenhagen) vs. QuantifilerTM H (G147A,
Copenhagen); (2) RB1 rt-PCR vs. QuantifilerTM H (G147A,
Copenhagen); and (3) QuantifilerTM H (Copenhagen) vs.
QuantifilerTM H (Linköping). Model: measured DNA = -
method + DNA preparation + method  DNA preparation + -
residual. Quantification results of QuantifilerTM DNA were not
included in the ANOVA.
2.3. Test of QuantifilerTM H batch variation
Two different batches of the QuantifilerTM H kit (L/N:
0404011 and 0411017) were tested using a 5 ng/ml dilution of
each of the six DNA preparations shown in Table 1. Each DNA
preparation was measured twice.
3. Results and discussion
degraded DNA, because single stranded DNA absorbs 20–30%
more UV light at 260 nm than double stranded DNA does. In
these three samples, contaminants that absorb UV light in the
range of 200–230 nm with a maximum of 6 units at
approximately 210 nm were observed (results not shown). The
presence of UVabsorbing contaminants may to some extent have
interfered with the measurements at 260 nm. UV-measurements
of DNA concentrations should not be relied upon unless one is
extremely confident of the quality and condition of the DNA. A
UV-measurement can be misleading unless the DNA is pure and
dissolved in a ‘UV-clean’ buffer solution. In general, DNA
quantification with SYBR-Green resulted in measurements of
15–50% higher DNA concentrations than expected based on the
declarations of the manufacturers, except for QuantifilerTM DNA
that was measured 9% lower than expected.
Fig. 2 shows a log–log scatter plot of the measured DNA
concentrations obtained with the slot blot and the QuantifilerTM
H (Raji cell reference DNA) quantification measurements of all
the DNA preparations (ordinate) compared to the expected

The six DNA preparations mainly contained DNA of high

molecular weight although D3160, D3035 and Human genomic
DNA (HG DNA) also contained a minor component of partly
degraded DNA (Fig. 1). There was no sign of RNA. The DNA
concentration measured by UV spectrometry of the DNA
preparations G147A, G152A and QuantifilerTM DNA were all
very similar (4%) and compared well to the expected DNA
concentrations declared by the manufacturers. The DNA
concentration of the DNA preparations D3160, D3035 and
HG DNA were 82, 29 and 123%, respectively, higher than the
expected DNA concentration (Table 3). The higher DNA Fig. 1. Agarose gel electrophoresis of DNA preparations. M1: ladder; 1:
concentration measurements for the samples D3160, D3035 and D3160; 2: 3035; 3: human genomic DNA; 4: G147A; 5: G152A; 6: quantifiler
HG DNA may to some extent be explained by the presence of DNA.
 K. Nielsen et al. / Forensic Science International: Genetics 2 (2008) 226–230
Fig. 2. Log–log scatter plot of DNA measurements using slot blot and
QuantifilerTM H. Regression lines were forced through (0,0) since all negative
DNA controls gave negative results for both methods. Expected DNA con-
centration: as informed by the manufacturers.
DNA concentration according to the declarations of the
manufacturers (abscissa). Both regression lines go through
(0,0) since the negative controls for both assays always gave
negative results. The slopes of the regression lines represent the
ratios between the measured and expected DNA concentra-
tions. In general, QuantifilerTM H measured the DNA
concentration 2.25 times higher than expected, while slot blot
in general measured 1.11 times higher than expected. The
regression coefficients of the QuantifilerTM H kit (R2 = 0.9486)
and the slot blot method (R2 = 0.7236) demonstrate that
QuantifilerTM H has less unexplained variation than the slot blot
method, i.e. QuantifilerTM H is more reproducible but less
accurate than slot blot. As supported by the supplementary
investigations presented below, the most likely explanation for
the inaccurate results obtained with QuantifilerTM H is the use
of the DNA from the Raji cell line as reference DNA.
In order to explore the unexpected measurements of DNA
concentrations with QuantifilerTM H, a collaborative exercise
was performed with the forensic genetic laboratories in
Innsbruck and Linköping. Fig. 3 shows the measured DNA
concentrations of the DNA preparations D3160, D3035, HG
DNA, G147A, G152A and QH DNA analyzed in three
laboratories using (1) QuantifilerTM H (Copenhagen and
Linköping), (2) QuantifilerTM H with G147A as reference
DNA (Copenhagen) and (3) RB1 rt-PCR with QuantifilerTM
DNA as reference DNA (Innsbruck). The method QuantifilerTM
H (Copenhagen) in average measured 100% higher than
QuantifilerTM H (G147A, Copenhagen). The result of the
ANOVA showed no interaction ( p = 0.5906) but highly
significant effects of method ( p < 0.001) and of DNA
preparation ( p < 0.001). In average, QuantifilerTM H (Linköp-
ing) measured 19% higher DNA concentrations than Quanti-
filerTM H (Copenhagen), and the result of the ANOVA showed
no interaction ( p = 0.6937) but significant effects of method
( p = 0.003) and of DNA preparation ( p = 0.002). When DNA
preparation G152A was excluded from the analysis, significant
interaction was observed ( p = 0.003), but no significant effect
of method ( p = 0.27) was observed. The quantification results
229
of G152A using RB1 rt-PCR were 56% lower than declared by
the manufacturers, and the results were significantly different
from those obtained with QuantifilerTM H with G147A as
reference. The reason for these results is not known. RB1 rt-
PCR quantifications of G147A and G152A were also performed
using a different DNA standard (GMI DNA [6]) and no
significant difference (t-test, p > 0.05) was observed between
the DNA concentrations measured using QuantifilerTM DNA
and GMI DNA as reference DNA (data not shown).
Gel electrophoresis and UV spectrometry (scanning from
200 to 350 nm) of the QuantifilerTM DNA showed that the DNA
was mainly of high molecular weight and that it contained
neither UV absorbing agents nor RNA in detectable amounts.
UV spectrometry (absorbance 260 nm), SYBR-Green fluoro-
metry, slot blot and RB1 rt-PCR measurements of DNA
confirmed that the DNA concentration was 200 ng/ml DNA as
declared by the manufacturer. In average, QuantifilerTM H
measured 100% higher DNA concentrations than QuantifilerTM
H with G147A as reference DNA demonstrating a calibration
problem when the QuantifilerTM DNA standard is used together
with the QuantifilerTM H kit. The hTERT gene encodes the
catalytic component of the human telomerase that is inactivated
in most normal cells, but is activated in 85% of all tested
immortalized cells [14]. This indicates that genetic differences
in the hTERT gene between wild type DNA and cell line DNA
exist. Mutations in the primer binding sites of one or both
hTERT alleles in the QuantifilerTM DNA standard might lower
the single-molecule PCR efficiency for affected gDNA target
molecules in early cycles, and result in the worst case in a total
loss of PCR product derived from the affected allele(s).
Sequence variation within the probe binding site could result in
a lowered number of fluorochrome molecules liberated per
PCR cycle causing a reduced ability (or even complete failure)
to detect the affected PCR products. Both possibilities would
Fig. 3. DNA quantification of samples with three methods in three forensic
genetic laboratories. The bars represent the means of two to four replicates and
the error bars represent the standard deviations. Quantifiler H (G147A): DNA
preparation G147A (Promega) was used as reference DNA. Using ANOVA
(methods: Quantifiler H and Quantifiler H (G147A); DNA preparations: D3160,
D3035, HG DNA, G147A and G152A) no interaction was observed
( p = 0.5906). Highly significant effects of both method ( p < 0.001) and
DNA preparation ( p < 0.001) were observed. In average, Quantifiler H mea-
sured 100% higher DNA concentrations than Quantifiler H (G147A).
230 K. Nielsen et al. / Forensic Science International: Genetics 2 (2008) 226–230
lead to an overestimation of the DNA content in the DNA
preparations tested.
In PCR assays for DNA quantification where genetic
differences in DNA template between wild type DNA and cell
line DNA can be expected, like hTERT, the use of cell line DNA
as reference DNA is not an optimal solution. However, the
QuantifilerTM cell line DNA standard works well with the RB1
rt-PCR kit, and the QuantifilerTM H kit works well with the
DNA preparation G147A (DNA extracted from blood samples
from 80–100 males). This demonstrates that cell line DNA can
be used as reference DNA and parts of potential oncogenes can
be used as DNA templates for quantification, but the
combination of cell line DNA and oncogene based DNA
quantification assays may create difficulties.
The method on which the QuantifilerTM H kit is based works
well. However, the measurements are inaccurate – although
reproducible – due to the fact that Raji cell DNA is used as
reference DNA. Many forensic laboratories still use the kit at
AB 7000 instruments. Therefore, it would be an improvement if
the Raji cell DNA were substituted with better suited DNA for
the purpose.
The results also demonstrate the need for internationally
agreed standard reference DNA materials and improvements of
the methods for DNA quantification [15,16].
Acknowledgment
We thank Poul Svante Eriksen, M.Sc., Department of
Mathematical Sciences, University of Aalborg, Denmark, for
help with the ANOVA analysis.
References
[1] P.S. Walsh, J. Varlaro, R. Reynolds, A rapid chemiluminiscent method for
quantification of human DNA, Nucleic Acids Res. 20 (1992) 5061–5065.
[2] J.S. Waye, H.F. Willard, Structure, organization, and sequence of alpha
satellite DNA from human chromosome 17: evidence for evolution by
unequal crossing-over and an ancestral pentamer repeat sharing with the
human X chromosome, Mol. Cell Biol. 6 (1986) 3156–3165.
[3] M.N. Mandrekar, A.M. Erickson, K. Kopp, B.E. Krenke, P.V. Mandrekar,
R. Nelson, K. Peterson, J. Shultz, A. Tereba, N. Westphal, Development of
a human DNA quantification system, Croat. Med. J. 442 (2001) 336–339.
[4] A.J. Mighell, A.F. Markham, P.A. Robinson, Alu sequences, FEBS Lett.
417 (1997) 1–5.
[5] H. Andréasson, U. Gyllensten, M. Allen, Real-time DNA quantification of
nuclear and mitochondrial DNA in forensic samples, Biotechniques 33
(2002) 402–411.
[6] H. Niederstätter, S. Köchl, P. Grubwieser, M. Pavlic, M. Steinlechner, W.
Parson, A modular real-time PCR concept for determining the quantity
and quality of human nuclear and mitochondrial DNA, Forensic Sci. Int.:
Genet. 1 (2007) 29–34.
[7] M.E. Sifis, K. Both, Burgoyne LA, A more sensitive method for quanti-
fication of genomic DNA by Alu amplification, J. Forensic Sci. 47 (2002)
589–592.
[8] J.A. Nicklas, E. Buel, Development of an Alu-based, real-time PCR
method for quantification of human DNA in forensic samples, J. Forensic
Sci. 48 (2003) 936–944.
[9] Quantifiler kits user manual, Applied Biosystems, Foster City, US, 2003.
[10] J. Sambrook, D.W. Russell, Molecular Cloning: A Laboratory Manual,
3rd ed., Cold Spring Harbor Laboratory Press, Cold Spring Habor, NY,
USA, 2001.
[11] K. Rengarajan, S.M. Cristol, M. Mehta, J.M. Nickerson, Quantifying DNA
concentrations using fluorometry: a comparison of fluorophores, Mol. Vis.
8 (2002) 416–421.
[12] J.A. Nicklas, E. Buel, Quantification of DNA in forensic samples, Anal.
Bioanal. Chem. 376 (2003) 1160–1167.
[13] General requirements for the competence of testing and calibrating
laboratories (ISO/IEC 17025:2005), CEN Management Centre, rue de
Strassart 36, B-1050 Brussels, Belgium.
[14] A. Zhang, et al., Frequent amplification of the telomerase reverse tran-
scriptase gene in human tumors, Cancer Res. 60 (2000) 6230–6235.
[15] M.C. Kline, D.L. Duewer, J.W. Redman, J.M. Butler, Results from the
NIST 2004 quantification study, J. Forensic Sci. 50 (2005) 570–578.
[16] P.M. Vallone, M.C. Kline, D.L. Duewer, A.E. Decker, J.W. Redman, J.C.
Travis, M.V. Smith, J.M. Butler, Development and usage of a NIST
Standard Reference Materials for real-time PCR quantitation of human
DNA, Forensic Sci. Int.: Genet., in press.