Int J Legal Med (2006) 120: 42–48
Ulrike Schmidt.Sabine Lutz-Bonengel.
Hans-Joachim Weisser.Timo Sänger.Stefan Pollak.
Ulrike Schön.Thomas Zacher.Wolfgang Mann
Low-volume amplification on chemically structured chips using
the PowerPlex16 DNA amplification kit
Received: 2 May 2005 / Accepted: 3 August 2005 / Published online: 18 October 2005
# Springer-Verlag 2005
Abstract In forensic DNA analysis, improvement of DNA
typing technologies has always been an issue. It has been
shown that DNA amplification in low volumes is a suitable
way to enhance the sensitivity and efficiency of amplifica-
tion. In this study, DNA amplification was performed on a
flat, chemically structured glass slide in 1-μl reaction
volumes from cell line DNA contents between 1,000 and
4 pg. On-chip DNA amplification reproducibly yielded full
allelic profiles from as little as 32 pg of template DNA.
Applicability on the simultaneous amplification of 15 short
tandem repeats and of a segment of the Amelogenin gene,
which are routinely used in forensic DNA analysis, is
shown. The results are compared to conventional in-tube
amplification carried out in 25-μl reaction volumes.
Keywords STR.On-chip amplification.DNA typing.
Low-volume amplification.Fragment length analysis
DNA analysis has become an indispensable tool in almost
every forensic laboratory and is an integral part of main
forensic tasks (i.e., paternity testing, human and nonhuman
identification, and examination of forensic stains) [1–6].
The application of polymerase-based amplification of
DNA—and especially of autosomal short tandem repeats
(STRs)—to forensic molecular genetics has been path-
breaking [7–10]. Meanwhile, there is a considerable panel
of sufficiently validated STR loci that is available in the
form of manufactured kits that can be employed to con-
veniently process the majority of samples that come up in
forensic casework [e.g., 11].
However, there are occasions when conventional analy-
ses of autosomal STRs fail. To deal with an ever-increasing
amount of samples containing either small amounts of
template or degraded DNA, several alternative approaches
have been applied to forensic DNA typing, depending on
analysis of mitochondrial (mtDNA) and X-chromosome
markers can be of major interest [12, 13]. Due to a higher
copy number, analysis of mtDNA is promising in degraded
samples and in samples lacking great amounts of nuclear
redesigned primers yielding reduced-size STR amplicons
. Y-chromosome STRs are applied to analyses of male/
female mixed stains .
Other than changing the target of analysis, there is a
possibility to enhance the sensitivity of DNA amplification
and to improve the detection of amplification products in
order to generate reliable allelic information from the
smallest amounts of template DNA. This objective brought
about many attempts to improve existing methods of DNA
analysis or to develop alternative strategies (e.g., based on
mass spectrometry, denaturing high-performance liquid
chromatography, microfluidic electrophoresis, or micro-
array technologies) [18–24].
Miniaturization of amplification reactions as well as
reduction of reaction volumes have been identified as pos-
sible methods to enhance the sensitivity of amplification
reactions and to simplify and speed up DNA analyses [24,
25]. Initial developments have been made of lab-on-chip
solutions that combine sample preparation, DNA amplifi-
cation and detection, and determination and visualization
of results . Various approaches in (micro)miniaturiza-
tion of PCR reactions, using microchips, capillaries, and
surfaces of microarrays, exist [18, 20, 27–29]. Electronic
U. Schmidt and S. Lutz-Bonengel contributed equally to this work.
U. Schmidt (*).S. Lutz-Bonengel.H.-J. Weisser.T. Sänger.
Institute of Legal Medicine,
Albert Ludwig University Freiburg,
79104 Freiburg, Germany
U. Schön.T. Zacher.W. Mann
95326 Kulmbach, Germany
microarrays also integrate the amplification and detection
of multiple samples . These methods are aimed at
automated high-throughput processing of samples and
have been introduced to forensic single nucleotide poly-
morphism typing, rather than to STRtyping . However,
many chip-based approaches require considerable expan-
sion or even change of laboratory equipment, especially
when analyzing STRs [27, 32, 33].
In this paper, the technical feasibility of efficient DNA
amplification in minimal reaction volumes, using a chem-
ically structured chip that can be integrated into existing
laboratory environments concerned with detection, is
Materials and methods
Human female DNA 9947A, supplied with the commer-
cially available PCR amplification kit PowerPlex16
(Promega, Mannheim, Germany), was diluted in sterile
water into DNA contents of 1,000, 500, 250, 125, 63, 32,
16, 8, and 4 pg, respectively.
DNA amplification on chemically structured chips
The chips were supplied by Alopex GmbH (Kulmbach,
Germany). The basic format is a standard microscope glass
slide, which was originally developed for single cell anal-
ysis and quantification of single genome equivalents .
Each chip exhibits a flat hydrophobic surface, which
contains 60 hydrophilic reaction compartments (“anchor
spots,” 1.6 mm in diameter) arranged in 12 rows of five
spots. Each of these anchor spots can be loaded with a 1-μl
amplification reaction volume, which is then covered with
oil in order to prevent evaporation and contamination
(Fig. 1a). The actual pipetting scheme with positioning of
negative controls and arrangement of samples is given in
In the experiments described here, reagents from the PCR
amplification kit PowerPlex16 (Promega) were used. Each
PCR reaction contained 0.1 μl of Gold Star 10× buffer,
0.1 μl of PowerPlex16 Primer Pair Mix, 0.5 μl of diluted
Fig. 1 a Preparation of the amplification chip. Application of 0.5 μl
of aqueous template solution (I), droplet of template solution
confined to a hydrophilic reaction compartment (II), addition of
0.5 μl of PCR reaction mix (III), complete reaction droplet confined
to a hydrophilic reaction compartment (IV), addition of 5 μl of oil
(V), finished preparation of one 1-μl amplification reaction covered
with 5 μl of oil (VI). b Example of a pipetting scheme. Am-
plification chip with randomly distributed negative controls. Black
spots indicate the positions of the negative controls. Samples were
loaded onto 40 of 60 anchor spots and were distributed as follows:
1,000 pg, reaction compartments 1, 3, and 4; 500 pg, reaction
compartments 6, 8, and 9; 250 pg, reaction compartments 11, 12,
and 14; 125 pg, reaction compartments 16, 17, and 18; 63 pg,
reaction compartments 22, 23, and 24; 32 pg, reaction compartments
27, 28, and 29; 16 pg, reaction compartments 31, 32, and 34; 8 pg,
reaction compartments 36, 38, and 39; 4 pg, reaction compartments
41, 42, and 43
DNA template, and 0.5 U of Taq DNA Polymerase
(Promega), in a total volume of 1 μl. Cycling was
performed using an Eppendorf Mastercycler with in situ
adapter (Eppendorf AG, Hamburg, Germany). Cycling
conditions were set according to the manufacturer’s rec-
ommendations. Negative controls were performed ran-
domly on different positions on the chip using the same
reagent solutions without DNA.
Fig. 2 Polyacrylamide gel electrophoresis analysis of 40 Promega
PowerPlex16 amplification products from a single chip. The amount
of template is indicated. nc Negative control, s size standard, DNA
molecular weight marker VIII (Roche, Mannheim, Germany),
including fragments of (19, 26, 34, 34, and 37 bp), 67, 110, 124,
147, 190, 242, 320, 404, 489, 501, 692, and 900 bp
Preparation of the chips
Four chips were prepared for cycling, with each chip
containing 27 samples with varying DNA contents and 13
negative controls. Individual anchor spots were supplied
with 0.5 μl of DNA template and 0.5 μl of reagent solution.
On each chip, 40 of 60 anchor spots (ten rows offour spots)
were prepared (see Fig. 1b). Finally, each anchor spot was
covered with 5 μl of mineral oil.
Analysis of PCR products
After amplification, all PCR products were transferred into
0.2-ml reaction tubes and mixed with 5 μl of H2O bidest,
and 1 μl of each diluted sample was analyzed on a 6%
polyacrylamide (PAA) gel (10 min, 62 W, and 400 mA).
Gels were stained with 0.1% AgNO3and developed with
(1:3) 0.3 M NaOH/36.5% formaldehyde solution. The re-
maining5μlwasdissolved in10μl offormamide. Samples
were centrifuged to completely separate the mineral oil
from the aqueous sample. Ten microliters of each diluted
sample was then analyzed on an ABI Prism 3100 Avant
Genetic Analyzer using GeneMapper ID Software Version
3.1 (Applied Biosystems, Darmstadt, Germany).
DNA amplification in PCR reaction tubes
Each PCR reaction contained 2.5 μl of Gold Star 10×
buffer, 2.5 μl of PowerPlex16 Primer Pair Mix, 1 μl of
diluted DNA template, and 4 U of Taq DNA Polymerase
(Promega), in a total volume of 25 μl. Cycling was per-
formed using an Eppendorf Mastercycler (Eppendorf AG).
Cycling conditions were set according to the man-
ufacturer’s recommendations. Negative controls were
Analysis of PCR products
From each amplification product, 5 μl was analyzed on a
2.5% agarose gel prior to capillary gel electrophoresis on
an ABI Prism 3100 Avant Genetic Analyzer. For the
analysis, GeneMapper ID Software Version 3.1 (Applied
Biosystems) was used.
Altogether, 108 DNA samples and 52 negative controls
were amplified. Figure 2 shows a silver-stained PAA gel of
40 amplification products obtained from one of four chips.
The samples include 27 amplification products resulting
from the amplification of different contents of template
DNA (1,000, 500, 250, 125, 63, 32, 16, 8, and 4 pg) and 13
Capillary gel electrophoresis demonstrated that all 52
negative controls contained no amplification products. A
summary of results on the content of DNA template is
shown in Table 1. Altogether, 62 complete allelic profiles
could be obtained from DNA contents from 1,000 to 32 pg.
Figure 3a exemplarily shows an electropherogram of am-
plification products obtained from 32 pg of template DNA.
Unsuccessful amplification occurred in 17 of 108 samples.
This phenomenon affected samples with DNA template
contents of 63 pg and less, as well as 1 of 12 samples with
homozygosity) as well as dropout of individual STR
systems were found in 29 of 108 PCR products. All of
these PCR products had been generated from template
contents of 16 pg or less. Figure 3b shows an example of an
allelic dropout simulating homozygosity in the STR system
Penta E. Allelic dropout was observed in all STR systems.
For in-tube PCR, 60 amplification reactions were per-
formed, including six negative controls (Table 1), and 25
samples generated from DNA contents between 1,000 and
63 pg yielded complete allelic profiles. Allelic dropout was
observed in samples with template amounts of 250 pg and
less. Unsuccessful amplification occurred in samples with
DNA contents of 16 pg (two of six) and 8 pg (five of six).
The amplification of 4 pg of template failed.
Table 1 Results of on-chip and in-tube PCR reactions in relation to
the amount of DNA template
PCR reactions Amount of template (pg)
1,000500 250125 6332 1684
000002 116 10
Total1212 1212 12 121212 12
The actual numbers of samples are given
As shown in this preliminary study, low-volume amplifi-
cation on chemically structured chips has the potential to
efficiently amplify even a large panel of forensically used
STRs in a very small reaction volume.
Preparation, handling, and manipulation of the chip are
very simple due to its chemical structure. Aqueous so-
lutions (reaction mixes and template DNA) are attracted to
the defined hydrophilic anchor spots.
Control of PCR yields is required prior to fragment
length analysis. However, the small reaction volume of the
chip PCR is challenging. Good impression of PCR results
could be obtained with a small portion of the diluted
samples on a PAA gel, with electrophoresis parameters as
Initial testing showed that the close local arrangement of
the anchor spots did not lead to any contamination of
negative controls. However, 4 of 108 samples with tem-
plate concentrations of 32, 16, and 4 pg, respectively,
showed additional alleles in 1 of 16 analyzed loci. Most
likely, these additional signals were PCR artifacts resulting
from a minor activity of the enzyme used during chip
preparation, which was performed at room temperature.
Allelic dropout was observed in all STR systems. A
correlation between allelic dropout and length of missing
alleles or size of missing STR systems could not be de-
tected. Allelic dropout, as a consequence of a higher
dilution factor, was randomly distributed. Unsuccessful
amplification reaction, as observed in 17 samples, was
more likely due to the geometry of the droplets than to the
inhibition from too much template. Further studies on this
topic are necessary.
Allelic dropout simulating homozygosity was observed
in all samples from DNA contents of 32 pg and less. A few
of these samples did not show any dropout of complete
Fig. 3 a PowerPlex16 electropherogram of a sample generated
from 32 pg of DNA template showing a complete allelic profile. b
Electropherogram of fluorescein-labeled STRs of PowerPlex16
showing an allelic dropout simulating homozygosity in Penta E.
This sample was amplified from 16 pg of DNA template