Demonstration of rapid multiplex PCR amplification involving 16 genetic loci§
Peter M. Vallone*, Carolyn R. Hill, John M. Butler
National Institute of Standards and Technology, Biochemical Science Division, 100 Bureau Drive, Mail Stop 8311, Gaithersburg, MD 20899-8311, United States
The forensic DNA typing community has settled on a core set of
short tandem repeat (STR) markers that are widely used to
generate DNA profiles used in database and casework applications
[1,2]. Multiplex amplification using the polymerase chain reaction
(PCR) copies these STR regions to detectable levels and labels the
PCR products with different colored fluorescent dyes. This multi-
plex PCR is commonly performed with 16plex STR typing kits, such
as Identifiler (Applied Biosystems, Foster City, CA) and PowerPlex
16 (Promega Corporation, Madison, WI), that simultaneously
amplify 15 STRs and the amelogenin sex-typing marker [3,4].
Standardthermal cycling for theseSTRkitstypicallytakes2.5–
3 h, which contributes to a significant portion of the approxi-
mately 8–10 h required to generate a DNA profile. Increasing the
improve the throughput of a DNA typing laboratory (commercial,
academic or governmental), but would also help speed up the
applications, such as analysis of individuals at a point of interest
like an airport or a country border.
While a great deal of effort has been invested over the past
decade in developing portable, miniature and integrated DNA
typing devices [5–7], very little focus appears to have been spent
on rapid PCR protocols in terms of getting multiple STR loci
amplified in a robust manner [8,9]. Most of the rapid PCR work to-
date is for single marker targets [5,6] that do not have to worry
about locus-to-locus and heterozygote intra-locus imbalance or
incomplete adenylation that can impact STR data interpretation
The performance and potential of rapid multiplex PCR
By maximizing the speed of the widely used GeneAmp 9700
thermal cycler (Applied Biosystems) and evaluating two rapid PCR
enzymes, a simple protocol for rapid amplification with a
commonly used STR typing kit is demonstrated.
2. Materials and methods
2.1. DNA samples and STR typing kits
Sixty samples from a U.S. population set  were used for
testing rapid cycling protocols. The samples were previously
genotyped  using the Identifiler (ID) STR typing kit with the
[3,4]. The primer mix from the ID kit was used without any further
alteration as described in the PCR conditions.
Forensic Science International: Genetics 3 (2008) 42–45
A R T I C L EI N F O
Received 4 August 2008
Received in revised form 10 September 2008
Accepted 14 September 2008
A B S T R A C T
Current forensic DNA typing is conducted in approximately 8–10 h. Steps include DNA extraction,
quantification, polymerase chain reaction (PCR) amplification of multiple short tandem repeat (STR) loci,
capillary electrophoresis separation with fluorescence detection, data analysis and DNA profile
interpretation. The PCR amplification portion of the workflow typically takes approximately 3 h with
standard thermal cycling protocols. Here we demonstrate a rapid cycling protocol that amplifies 15 STR
loci and the sex-typing marker amelogenin from the Identifiler STR typing kit in less than 36 min. This
rapidprotocolemployscommerciallyavailable polymerasesand thewidelyusedGeneAmp 9700thermal
cycler.Complete concordance of STR allele calls (for 60 samples) between the rapid and standard thermal
cycling protocols were observed although there was incomplete adenylation at several of the loci
examined and some PCR artifacts were detected. Using less than 750 pg of template DNA and 28 cycles,
STR peaks for all loci were above a 150 relative fluorescent unit (RFU) detection threshold with fully
adequate inter-locus balance and heterozygote peak height ratios of greater than 0.84.
Published by Elsevier B.V.
§Certain commercial equipment, instruments and materials are identified in
order to specify experimental procedures as completely as possible. In no case does
such identification imply a recommendation or endorsement by the National
Institute of Standards and Technology nor does it imply that any of the materials,
instruments or equipment identified are necessarily the best available for the
* Corresponding author. Tel.: +1 301 975 4872; fax: +1 301 975 8505.
E-mail address: firstname.lastname@example.org (P.M. Vallone).
Contents lists available at ScienceDirect
Forensic Science International: Genetics
journal homepage: www.elsevier.com/locate/fsig
1872-4973/$ – see front matter. Published by Elsevier B.V.
2.2. Rapid polymerases
The two polymerases utilized in this study were PyroStart
(Fermentas, Glen Burnie, MD) and SpeedSTAR (Takara Bio USA,
Madison, WI). PyroStart is shipped as a 2? master mix that
product literature. The SpeedSTAR enzyme is shipped separately
from buffer and reagent components. The SpeedSTAR enzyme
(approximately 1 unit) was added to the PyroStart mastermix to
increase PCR efficiency and to improve adenylation. The reported
rates of nucleotide incorporation for PyroStart and SpeedSTAR
based on product literature are ?40 and ?100 nucleotides/s,
2.3. Rapid thermal cycling
PCR was carried out in 10 mL reaction volumes. A typical
reaction consisted of 5 mL PyroStart 2? master mix, 0.25 mL
SpeedSTAR enzyme (5 units/mL), 2 mL of the ID STR primer mix,
1.25 mL water and 1.5 mL of template DNA (0.5 ng/mL).
All thermal cycling experiments were performed on a Gene
Amp 9700 (Applied Biosystems) using a maximum ramp rate of
4 8C/s. Amplification conditions consisted of 95 8C for 1 min
followed by 28 cycles of 95 8C 5 s, 58 8C 10 s, 72 8C 10 s, followed
by a 1 min incubation at 72 8C to aid adenylation, and then 25 8C
until removed from the thermal cycler.
2.4. Data collection and analysis
Following PCR, 1 mL of the amplified products was diluted into
8.7 mL of Hi–Di formamide (Applied Biosystems) and 0.3 mL of LIZ
GS500 internal size standard (Applied Biosystems). Samples were
electrokinetically injected at 3 kV for 10 s and separated on a
3130xl Genetic Analyzer (Applied Biosystems) using POP-6
polymer (Applied Biosystems) on a 36 cm capillary array (Applied
Biosystems). After data collection, genotyping was performed in
GeneMapperID v3.2 (Applied Biosystems) using manufacturer
provided bins and panels.
2.5. Sensitivity study
One nanogram of a pristine DNA template was serially diluted
down to 50 pg (1000, 750, 500, 400, 300, 200, 100, 50 pg). The DNA
template was amplified in duplicate as described above. A volume
of 1 mL of the DNA template and 1.75mL of water were used in the
rapid amplification protocol. Allele calls were performed using a
threshold of 50 RFU to evaluate the sensitivity limitations of the
3. Results and discussion
3.1. Rapid thermal cycling protocol and time savings
Rapid cycling parameters such as ramp rate and dwell times
were altered to reduce the time required for completion of 28
cycles. A general comparison is illustrated in Table 1 contrasting
the thermal cycling times for the rapid cycling protocol versus a
hypothetical standard cycling protocol. A direct comparison
indicates that the majority of the time is saved in the final soak
step (41.2%) and in the combined time saved resulting from the
reduced cycling times (?50%). Time is also saved by using a
polymerase other than AmpliTaq Gold that does not have a 10 min
activation time. When compared on the same time scale, the 28
rapid cycles are completed before the end of the fourth cycle under
standard thermal cycling conditions.
Initial rapid multiplex PCR experiments performed poorly
without the additional SpeedSTAR enzyme. This was evidenced by
locus drop out, poor locus-to-locus balance and low heterozygote
peak height ratios (data not shown). The improvement from the
additional enzyme was drastic (full profiles, higher peak height
ratios, improved adenylation). Few adjustments were made to the
thermal cycling protocol indicating that the optimal fast enzyme
cocktail/combination was essential for rapid multiplex PCR
success. Previous attempts at rapid multiplex PCR using the
relatively slower activating hot start AmpliTaq Gold 
commonly used in the forensic community may have been a
limiting factor in developing a successful rapid amplification
3.3. Rapid STR typing results
Complete 16 locus amplification profiles were successfully
obtained in less than 36 min using primers from the ID STR typing
kit. A visual inspection of the rapid amplification of the ID kit
suggests that there is adequate balance between all 16 loci (Fig. 1).
However, there were some consistent low intensity PCR artifacts
observed in the VIC dye channel at ?107, ?168, ?287 bp and
incomplete adenylation was observed for several loci including
D8S1179, D7S820, D3S1358, TH01, vWA, TPOX, and D5S818.
D2S1338 and CSF1PO were also poorly adenylated, but the ?A and
+A peaks were not fully resolved (see Fig. 1). The level of
incomplete adenylation was consistent for all the rapid ID PCR
experiments. It should be noted that other minor non-specific PCR
artifacts were observed, but under the described rapid amplifica-
tion protocol these did not interfere with accurate genotyping. The
D19S433 and D21S11 loci exhibited lower signal intensity on
average across the samples examined for this study using 750 pg
input DNA (?850 and 1720 RFUs, respectively). This lower
D19S433 and D21S11 signal was consistently observed in all 60
samples examined. The average peak height ratio (PHR) was
determined from the heterozygous samples at each locus. On
average, PHR ranged from 0.84 (vWA) to 0.92 (D8S1179). Our
observed inter-locus performance was similar to what is to be
DNA (PHR > 0.80) [3,4].
3.4. PCR amplicon size
By using an annealing temperature of 58 8C in our rapid cycling
protocol it was evident that the PCR primers were binding to their
specific template targets and the elongation time was sufficient
(10 s) for efficient production of a full length amplicon of greater
than 300 bp. The larger molecular weight loci, CSF1PO, D2S1338
and D18S51, all produced strong signal intensities.
At the concentration of input DNA investigated (750 pg), a
strong correlation between amplicon size and PCR efficiency is not
Comparison of thermal cycling times.
Parameter Standard RapidDifference (min) % difference
Hot start (min)
Ramp rate (8/s)
P.M. Vallone et al./Forensic Science International: Genetics 3 (2008) 42–45
evident. This can be further illustrated by the relatively low signal
of D19S433 (amplicon size < 140 bp). A ‘ski-slope’ pattern with a
lower signal for larger amplicons that one might have predicted
was not evident for any of the electropherograms suggesting that
rapid PCR efficiency is independent of amplicon size. This is in
contrast to previous rapid PCR results obtained with a 3 locus
multiplex system . Belgrader et al. observed drop out for the
largest locus FGA (?240 bp) when using 4 units of AmpliTaq Gold,
10 ng of template DNA and 30 cycles while attempting to reduce
cycling time to less than 1 h .
3.5. Incomplete adenylation
Various loci exhibited products with incomplete adenylation.
This can be an issue for profile interpretation. However, the results
are promising considering that the final soak time was reduced
from 60 min down to a single minute. Further work in this area
could include the selection and characterization of primer
sequences that promote more efficient adenylation under rapid
PCR conditions similar to what was done previously with
conventional thermal cycling . Alternatively, additional loci
(autosomal, Y chromosome, mitochondrial) could be specifically
selected with properties that exhibit success with rapid cycling
conditions such as complete adenylation, good amplification
efficiency and peak balance.
3.6. Genotyping concordance
The genotyping results from rapid cycling experiments were
fully concordant with those from the standard vendor protocol.
A total of 960 correct genotype calls were made using the 16
locus multiplex (60 samples ? 16 loci = 960). In our limited
sample cohort, allele drop out was not observed for any of the
3.7. Sensitivity study
An evaluation of the minimal amount of DNA sample that can
reliably be amplified with the rapid amplification protocol was
performed. A highly characterized sample was amplified in
duplicate for 28 cycles. The DNA template concentrations tested
were; 1000, 750, 500, 400, 300, 200, 100 and 50 pg. Signal intensity
for D21S11 and D19S433 was consistently lower for all the
concentration tested. Allele drop out was observed between 200
and 300 pg ofDNA template whereas conventionalthermal cycling
with 28 cycles could routinely produce results with DNA template
in the 100–200 pg concentration range (data not shown). This
lower efficiency may be a result of the extremely short hold times
(5 and 10 s) employed in the rapid amplification protocol.
The initial work presented here indicates the great potential of
reduced thermal cycling times for large STR multiplexes using
enzymes other than the standard AmpliTaq Gold DNA polymerase
provided with the commercial STR kits. It should be stated that the
ID primer mix performed surprisingly well under PCR conditions
quite different than prescribed by the manufacturer. Results
obtained from a commercial multiplex provide insight on the
needs to be addressed for further rapid PCR protocol optimization.
used for successful amplification of 16 STR loci is very promising
for the developers of screening and integrated portable devices.
Portable or integrated devices often use smaller PCR volumes and a
Fig. 1. An Identifiler result utilizing the rapid PCR protocol demonstrating that all 16 loci are amplified. Several amplification artifacts that were consistently observed are
outlined with solid boxes. Dotted boxes indicate loci with incomplete adenylation.
P.M. Vallone et al./Forensic Science International: Genetics 3 (2008) 42–45
custom thermal cycling platform. Results presented here provide
the potential for optimizing a larger STR PCR multiplex on a
PCR may alsobe useful for typingreference samples inforensicand
paternity testing labs.
The goal of this work is to provide a starting point for
investigating and further optimizing rapid PCR protocols. This
36 min using commercially available primer sets and enzymes and
a thermal cycler that is used in most forensic laboratories. Even if
some artifacts arise during rapid PCR protocols, such as incomplete
adenylation or a few non-specific products, the STR profile
information can still be valuable for general screening and
informational purposes. In simple screening situations there
should be sufficient quantities of single-source high-quality
Until recently, success with amplifying large (>4 loci) multi-
plexes in less than 1 h has been limited [9,12]. The STR
performance parameters of locus-to-locus balance, locus or allelic
drop out, incomplete adenylation, low peak height ratio and
general robustness must be evaluated when developing a rapid
PCR protocol. Some of these problems were addressed with
additional enzyme or altering cycling times. But others are specific
to a locus, primer pair sequence or primer pair concentration
present in a commercial primer mix. From the results presented
with a commercial kit there is now a basis for understanding the
limitations of using a ‘fixed’ (in terms of primer sequence and
concentration) primer mix. It also allows us to direct future focus
on specific aspects of primer design to resolve these issues when
developing new rapid multiplex PCR assays.
The PyroStart/SpeedSTAR polymerase combination will add
slightly to the enzyme cost ($1.15) of a rapid cycling protocol per
reaction compared to 2 units of AmpliTaq Gold ($0.86). However,
the additional cost of a rapid cycling protocol should be balanced
against the time savings and potential for increased throughput.
There are a number of additional experiments to perform
including evaluating more rapid thermal cycling instrumentation
capable of faster temperature ramp rates, investigating the impact
of further variations in PCR volumes, further optimization of
annealing temperatures, and determining levels of sensitivity. We
also plan to examine additional non-kit STR loci  where primer
concentrations can be modified to adjust locus-to-locus balance
and primers ends can be modified to aid adenylation . While
our initial results are promising, there is still much to do to further
the understanding of performance characteristics with rapid
This work is funded in part by the National Institute of Justice
of Law Enforcement Standards. Points of view in this document are
those of the authors and do not necessarily represent the official
position or policies of the US Department of Justice.
 J.M. Butler, Forensic DNA Typing, 2nd edition, Elsevier, New York, 2005.
 J.M. Butler, Genetics and genomics ofcore STR loci used inhuman identity testing,
J. Forensic Sci. 51 (2006) 253–265.
 B.E. Krenke, A. Tereba, S.J. Anderson, E. Buel, S. Culhane, C.J. Finis, C.S. Tomsey, J.M.
Zachetti, A. Masibay, D.R. Rabbach, E.A. Amiott, C.J. Sprecher, Validation of a 16-
locus fluorescent multiplex system, J. Forensic Sci. 47 (2002) 773–785.
 P.J. Collins, L.K. Hennessy, C.S. Leibelt, R.K. Roby, D.J. Reeder, P.A. Foxall, Devel-
opmental validation of a single-tube amplification of the 13 CODIS STR loci,
D2S1338, D19S433, and amelogenin: the AmpFlSTR Identifiler PCR Amplification
Kit, J. Forensic Sci. 49 (2004) 1265–1277.
 P. Belgrader, W. Benett, D. Hadley, J. Richards, P. Stratton, R. Mariella Jr., F.
Milanovich, PCR detection of bacteria in seven minutes, Science 284 (1999)
 C.J. Easley, J.M. Karlinsey, J.M. Bienvenue, L.A. Legendre, M.G. Roper, S.H. Feldman,
M.A. Hughes, E.L. Hewlett, T.J. Merkel, J.P. Ferrance, J.P. Landers, A fully integrated
microfluidic genetic analysis system with sample-in-answer-out capability, Proc.
Natl. Acad. Sci. U.S.A. 103 (2006) 19272–19277.
 P. Liu, T.S. Seo, N. Beyor, K.J. Shin, J.R. Scherer, R.A. Mathies, Integrated portable
polymerase chain reaction-capillary electrophoresis microsystem for rapid for-
ensic short tandem repeat typing, Anal. Chem. 79 (2007) 1881–1889.
 M.G. Roper, C.J. Easley, J.P. Landers, Advances in polymerase chain reaction on
microfluidic chips, Anal. Chem. 77 (2005) 3887–3893.
 K.M. Horsman, J.M. Bienvenue, K.R. Blasier, J.P. Landers, Forensic DNA analysis on
microfluidic devices: a review, J. Forensic Sci. 52 (2007) 784–799.
 J.M. Butler, R. Schoske, P.M. Vallone, J.W. Redman, M.C. Kline, Allele frequencies
for 15 autosomal STR loci on U.S. Caucasian, African American, and Hispanic
populations, J. Forensic Sci. 48 (2003) 908–911.
Concordance study between the AmpFlSTR MiniFiler PCR Amplification Kit and
conventional STR typing kits, J. Forensic Sci. 52 (2007) 870–873.
 P.Belgrader, J.K. Smith,V.W.Weedn, M.A.Northrup,RapidPCR foridentity testing
using a battery-powered miniature thermal cycler, J. Forensic Sci. 43 (1998) 315–
 M.J. Brownstein, J.D. Carpten, J.R. Smith, Modulation of non-templated nucleotide
BioTechniques 20 (1996) 1004–1010.
 C.R. Hill, M.D. Coble, J.M. Butler, Characterization of 26 miniSTR loci for improved
analysis of degraded DNA samples, J. Forensic Sci. 53 (2008) 73–80.
P.M. Vallone et al./Forensic Science International: Genetics 3 (2008) 42–45