WI

\

HMIIHLHIHI V

W

Ill!‘

WW“

I”

 

fl
is

620255”
73;?" 7/6 A?

LIBRARIES
MICHIGAN STATE UNIVERSITY
EAST LANSING, MICH 48824-1048

This is to certify that the
thesis entitled

THE UTILITY OF WHOLE GENOME AMPLIFICATION IN FORENSIC
DNA ANALYSIS

presented by

AMY LEIGH BARBER

has been accepted towards fulfillment
of the requirements for the

M. S. degree in Forensic Science

 

 

,~.—.—-—\_

Major Professor's Signature
3/ / //o 7‘
Date

 

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 cz/cficmmeoueinwms

 

THE UTILITY OF WHOLE GENOME AMPLIFICATION IN FORENSIC DNA
ANALYSIS

By

Amy Leigh Barber

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Criminal Justice

2005

ABSTRACT

THE UTILITY OF WHOLE GENOME AMPLIFICATION IN FORENSIC DNA
ANALYSIS

By
Amy Leigh Barber
Forensic biologists are ofien confronted with issues of limited DNA which may
restrict their ability to generate an identification profile and reserve sample for future
analysis. Although PCR is commonly used to amplify specific segments of DNA, whole
genome amplification (WGA) is a process which is capable of replicating an entire
sample; reducing the likelihood of entirely consuming evidence. WGA of low quantity
clinical samples has been studied thoroughly, but there is nothing in the literature
describing its use in forensic biology. The objective of this project was to characterize
Improved Primer Extension Preamplification (I-PEP) and Multiple Displacement
Amplification (MDA), two methods of WGA, on samples commonly seen in forensic
laboratories by testing both low quality and quantity DNA. Control DNA, artificially
degraded DNA, and DNA from fresh and aged blood, hair, and bone before and after
WGA were PCR amplified to test maximum amplicon lengths and to examine product
yields. No difference was seen in the maximum PCR product length of DNA from fresh
blood, longer product was generated from aged blood, and there was a reduction in
amplicon length from hair and aged bone after WGA. Product yield was increased 20 to
2000 fold by I-PEP and 1000 to 10,000 fold by MDA. All MDA tests on low quality
samples were unsuccessful and often resulted in extensive non-target DNA amplification.
Overall, I-PEP and MDA increased the product yield of high quality DNA, but they were

not successful on highly degraded samples.

To my mother, Kay, and my grandparents, Frederick and Irene——

my three biggest fans

iii

ACKNOWLEDGEMENTS

First and foremost, I would like to thank Dr. David Foran for all his patience and
assistance with the writing of this thesis, not to mention all the guidance he has provided
over the last two years. I would also like to thank Dr. Kim Scribner and Dr. Mahesh
Nalla for taking time out of their schedules to be on my committee, your comments were
greatly appreciated.

Next, I would like to acknowledge Elizabeth Graffy, Lisa Misner and Stephanie
Rennick for their previous work with DNA from hair and bone, and Andrea Halverson
for establishing the blood samples. I can not thank you all enough for allowing me to use
your samples and to reproduce your results.

Two companies were kind enough to provide me with free samples for this
project. Roche Applied Sciences donated an Expand High Fidelity PCR System, and
Amersham Pharmacia donated a GenomiPhi kit.

What would you do without family? I would like to thank Grammy and Grampy
Wellman, all my parents (Kay and Eric, Russell and Mary), siblings (Karin and Nathan),
and nieces (Madison and Morgan) for always being here for me, even if only in spirit.
And finally, I need to thank Vincent Annese for motivating me, supporting me, and for

keeping me laughing.

iv

 

TABLE OF CONTENTS

LIST OF TABLES .. .vii
LISTOFFIGURES V111
Overview l
MethodsofWGA 2
ProblemsAssociatedwithWGA 6
Forensic Samples Used to Characterize I-PEP and MDA ....8
Preparation ofArtificiallyDegradedDNA . . .. . ...10
DigestionofFreshBlood ...ll
DigestionongedBlood 12
DNA from Aged Skeletal Material and Hair . ...12
Organic Extraction ofDNA ......12
Whole Genome Amplification . 13
Improved Primer Extension Preamplification .. 13
MultipleDisplacementAmplification ...........l4
Purification of Whole Genome Amplified Sample . 14
PCRAmplification . .......15
STR Analysis of Blood and Artificially Degraded DNA .. ........17
Mitochondrial DNA Sequencing of Bone and Hair Samples .. 18
RESULTS . ....20
Quality of I-PEP and MDA Product from Artificially Degraded DNA ....21
Quality of I-PEP and MDA Product using DNA from Fresh and Aged Blood ..... 27
Quality of I-PEP and MDA Product using DNA from Hair Shafts .............. 28
Quality of I-PEP and MDA Product from Aged Skeletal Material . ....28
DNA Yield Comparisons using Serial Dilutions . ..30
Quantity of I-PEP and MDA Product using DNA fi'om Fresh and Aged 131668.30
Quantity of I-PEP and MDA Product using DNA from Hair ...32

STR Analysis of Artificially Degraded DNA and Fresh
andAgedBloodSamples .............33
Sequencing of mtDNA from Hair Shaft and Aged Skeletal MateriaI ...40

 

 

TABLE OF CONTENTS CONTINUED

DISCUSSION

Advantages and Disadvantages of I-PEP and MDA

General WGA Results

Product Yield from I-PEP and MDA

Downstream Analysis of Whole Genome Amplified Product ..

Conclusions

REFERENCES

vi

.. ...42
...42
.°.'.:.'.'..47
....50

51

.......56

 

LIST OF TABLES

NuclearDNAPrimers 15
MitochondrialDNAPrimers l6
PrimerPairsUsedforPCRAmplificationandSequencing 16
Nuclear and mtDNA Amplification of Artificially Degraded DNA ...................... 23
Amplification of Nuclear and Mitochondrial DNA from Fresh and Aged Blood .27
Amplification of Mitochondrial and NuclearDNA from Hair Shafts 28
Amplification of Mitochondrial DNA from Aged Skeletal Material 29

PCR Amplification of Nuclear DNA (496bp product) from Fresh and Aged Blood .31

Amplification of Serial Diluted mtDNA from Hair Shafis ................33
STR Profiles of Artificially Degraded DNA (Trial 5, Single Source DNA) .35
STR Profiles of DNA from Fresh and Aged Blood with and without WGA ...36

Mitochondrial Sequence Comparisons of Untreated and [PEP DNA from Hair Shafts. . . 4]

vii

LIST OF FIGURES

Schematic Illustrating the MDA Process
Image of I-PEP and MDA Whole Genome Amplified Product
DNase I Digested Male DNA

220bp mtDNA Product from MDA and I-PEP Treated Voegtly Cemetery
Bone Samples

496bp Nuclear PCR Product of I-PEP, MDA and Untreated DNA Diluted from Fresh
496bp Nuclear PCR Product of I-PEP and MDA DNA Diluted from Aged Blood
203bp mtDNA PCR Product of I-PEP Diluted DNA from Hair Shafts
Electropherogram of Untreated DNA from Fresh Blood

Electropherogram of DNA from Fresh Blood after I-PEP WGA

Electropherogram of DNA from Fresh Blood after MDA WGA

viii

...20

...21

...29

...31

...32

...33

...37

...38

...39

INTRODUCTION

Overview

DNA analysis has revolutionized forensic identification. Even before DNA was
considered in forensic casework, new techniques were being developed and optimized to
make DNA analysis easier, faster, and more informative. With advances in molecular
biology developing quickly, the use of DNA as a means of individual identification
became an obvious choice for the forensic community. The advent of the polymerase
chain reaction (PCR) has further revolutionized forensic identification by allowing
laboratories to obtain results with ever smaller amounts of DNA. However, while
amplification techniques require a relatively small amount of input DNA template, the
initial quantity and quality of the extracted DNA can still be a limiting factor. In many
instances, only a minute amount of evidence containing DNA is deposited at a crime
scene. Likewise, adverse interim storage conditions may compromise the quality of the
sample. In these or other cases, obtaining sufficient amounts of DNA to determine a
sample’s identification profile may be problematic. Moreover, the preservation of a
sample for analysis by the defense must be considered; failure to do so may raise further
legal issues.

Concerns associated with limited starting quantities of DNA in forensic samples
could be alleviated by increasing the overall amount of DNA prior to beginning standard
laboratory procedures. Whole genome amplification (WGA) is a procedure used to
duplicate DNA, but unlike PCR, which generally amplifies small sections of the genome,

WGA is capable of replicating the genome in its entirety. Currently, a handfirl of WGA

 

techniques have been developed, including Primer Extension Preamplification (PEP),
Improved Primer Extension Preamplification (I-PEP), Multiple Displacement
Amplification (MDA), OmniPlex, and Degenerative Oligonucleotide Primed Polymerase
Chain Reaction (DOP-PCR). These methods are all backed by authors who claim
product yield is increased 1 to 80ug per 100ul reaction with great fidelity (Barker et al.,
2004; Dietmaier et al., 1999; Hosono et al., 2003; Kuivaniemi et al., 2002; Lovmar et al.,
2003). These studies involved the use of WGA on clinical samples that are typically in
good condition but are limited in quantity. Currently, no data have been published
characterizing WGA for use with forensic samples in which DNA can be limiting in both
quality and quantity.

The goal of this project was to characterize WGA methods for use on samples
commonly encountered in forensic casework. This included determining the usefulness of
WGA when very small amounts of DNA were present, and when DNA quality was poor
(degraded DNA). Two WGA methods, I-PEP and MDA, were tested in detail, as
previous work showed they increase DNA yields while requiring less input DNA
(Dietrnaier et al., 1999; Dean et al., 2002). The three remaining methods were not
evaluated as they had characteristics and requirements that made them inappropriate for

forensic samples (see below).

Methods of WGA

The WGA technique DOP-PCR (Telenius et al., 1992) uses a 22-base primer that
has six random bases inserted between specific sequences on the 5’ and 3’ ends (5’-
CCGACTCGAGNNNNNNATGTGG-3’). This degenerative Oligonucleotide is added to

the WGA mixture and cycled through an amplification reaction of 95°C for 5min,

followed by five cycles of 94°C for 1min, 30°C for 90sec, ramping to 72°C at a rate of
3.5°C/153ec and 72°C for 3min, then 35 cycles of 94°C for 1min, 62°C forlmin and 72°C
for 2min (with a 14sec increase per cycle) and a final extension of 72°C for 7min.
Although it has been shown that DOP-PCR has the capability of amplifying greater
amounts of DNA than some other methods (Dean et al., 2002), it was not characterized
for use with forensic samples as the required amount of input DNA is relatively high,
with a minimum of DNA from 10 to 100 cells being necessary for complete amplification
(Dietmaier et al., 1999). Further, Grant et al. (2002) showed that preferential
amplification of certain regions of the genome may occur due to the partially random
nature of the primer. These authors also found that DOP-PCR whole genome amplified
product may not be stable even when stored at —20°C, although no explanation for this
instability was given.

OmniPlex (Langmore, 2002) is a relatively new method of WGA, marketed by
Rubicon Genomics. With this technique, a library of the sample is made from the DNA
and then WGA is performed. Very little information exists in the primary literature on
this proprietary method, and therefore it was not used in this study. Future investigations
will be necessary to determine its usability in forensic casework.

PEP was first introduced by Zhang et al. in 1992, and was one of the earlier
methods developed for WGA. PEP is completed under conditions similar to that of
typical PCR, except that random 15 base oligonucleotides are utilized. The PEP
amplification procedure involves 50 cycles of denaturation of DNA at 95°C for 1 minute,
annealing of primers at 37°C for 2 minutes, a 10 sec /degree ramping from 37°C to 55°C,

followed by a 4 minute extension step at 55°C. Sun et a]. (1999) showed that PEP WGA

 

 

 

can result in a 30 fold increase in DNA. However, due to improvements made to this
technique, it was not investigated for use with forensic samples.

I-PEP (Dietmaier et al., 1999) incorporates three modifications to PEP to allow
for more complete DNA amplification. First, the I-PEP method utilizes a particular lysis
buffer, (1X Expand High Fidelity PCR buffer (no 3), 4mg/ml proteinase K and 0.5%
Tween 20) during the initial DNA extraction phase to release DNA from a sample more
efficiently. Second, I-PEP includes an additional elongation step at the end of each cycle;
the modified settings are 95°C for 1 minute, 37°C for 2 minutes, a lOsec/degree ramping
from 37°C to 55°C, 55°C for 4 minutes, and 68°C for 30 seconds, all repeated 50 times.
The final adjustment made by Dietrnaier et a1. (1999) is the incorporation of both Taq
polymerase and a proofreading enzyme in the reaction mixture. The researchers used a
product by Boehring Mannheim called Expand High Fidelity PCR System which
contained Pwo DNA polymerase (the enzyme has since been substituted by the
manufacturer for the enzyme Tgo DNA polymerase, though there is no functional
difference between the two enzymes, Roche, 2003). Like Taq, Pwo is a heat stable, 5’ to
3’ polymerase, which also exhibits a 3’ to 5’ exonuclease activity which allows the
enzyme to remove incorrectly incorporated nucleotides. This proofreading ability results
in an error rate 18-fold lower than that of Taq polymerase alone (Roche, 2003). DNA
polymerases present in a PEP and I-PEP WGA reactions are, in theory, capable of
replicating the genome in an unbiased manner as the random 15-mer oligonucleotides can
locate a complimentary sequence anywhere in the sample DNA. Like PEP, I-PEP WGA

has been shown to result in a 30 fold increase in product yield (Dietmaier et al., 1999).

MDA, was first described by Dean et a1. (2002; see Figure l for an illustration of
the MDA process) and utilizes the DNA polymerase Phi 29, which has an optimal
working temperature of 30°C. This polymerase is very processive, with an average
product length of approximately lOkb before becoming dislodged from the template
strand, however it is capable of synthesizing up to 70 kilobases (Blanco et al., 1989). In
addition, Phi 29 has a proofreading 3’ to 5’ exonuclease activity and is capable of
separating double stranded DNA encountered during synthesis. Separation of the DNA
helix in front of the enzyme not only allows continuous access to the template strand, but
it also frees the non-template strand of DNA to which more random primers can anneal,
allowing for more Phi29 enzyme extension. Despite its recent introduction, this method
has been studied extensively. Hosono et a]. (2004) determined that up to a 10,000 fold
increase in DNA yield can be achieved with MDA.

Two companies offer kits that can simplify the WA reaction, including the one
used in this project called GenomiPhi, manufactured by GE Healthcare
Systems/Amersham Pharmacia. The MDA reaction involves denaturating the DNA
template in the provided sample buffer, which contains random hexamer primers. The
mixture is then chilled, in order to protect the heat sensitive Phi 29 DNA Polymerase.
Next, the reaction buffer (containing dNTP’s) and the enzyme mix (composed of the Phi
29 DNA polymerase, additional random hexamer primers, and a proprietary buffer) is

added, and the reaction mixture is incubated at 30°C for 16 to 18 hours.

Figure 1: Schematic Illustrating the MDA Process

3
Polymorilatmn
continues

Primers band
to template

4 _
Strand displacement New primers bind to
n uwly formed
DNA

6
Polymvnzation from
new strands

 

Dark grey dots represent the Phi29 DNA polymerase, white bars represent the random hexamer primers,
solid gray lines represent template DNA, lines transitioning fiom white to gray represent recently amplified
DNA. (Image from Amersham Biosciences, 2003)

Problems Associated with WGA

The premise of WGA is that product yield can be increased from limited starting
material. Authors have demonstrated that DNA from as little as 1 cell can be amplified
(Dietmaier et al., 1999, Lasken et al., 2003, Zhang et al., 1992), but such minute starting
material raises concern regarding efficiency rates, loss of heterozygosity and sequence
bias. Dietrnaier et a1. (1999) found that when using the I-PEP procedure, post-WGA
testing (sequencing and RF LP-PCR) had a 100% efficiency rate (based on the percentage
of successful amplifications) when the DNA from as little as 5 cells was used. This was
in comparison to a 10 to 50% efficiency rate for PEP, and a 0 to 30% efficiency rate for

DOP-PCR with the same amount of starting DNA. However, they determined that when

DNA was extracted from formalin fixed, paraffm embedded tissues, (a treatment that is
expected to adversely affect the quality of the sample) a minimum of 30 cells was needed
to achieve an efficiency rate of 100% from I-PEP samples. Dean et al. (2002) claimed
that less than 10 cells are necessary to accurately amplify the genome with MDA, but did
not indicate the necessary starting quality.

Another concern surrounding I-PEP is the loss of heterozygosity when a minute
DNA sample is subjected to WGA. This loss is likely due to stochastic effects resulting
from so few copies of each allele being present in the starting material, such that one
allele is amplified by chance far more than the other, which could be problematic for any
method of WGA. Dietrnaier et al. (1999) showed that when a known number of cells (10
to 100) was added to PEP and I-PEP reactions, fewer cells were necessary for I-PEP to
accurately amplify both alleles at 6 loci.

Finally, Dean et al. (2002) found that some sequence bias towards certain regions
of the genome may occur during PEP, DOP-PCR and MDA. The likelihood of any one
region of the genome being preferentially amplified over another can be attributed to
priming efficiency, GC content, and proximity to the ends of the chromosome (Dean et
al. 2002). In studies investigating the utility of MDA (Pan et al., 2004; Tzvetkov et al.,
2005), direct sequencing and SNP analysis were used to determine the degree of
amplification bias. These authors showed that MDA often under-represents certain
regions of the genome, typically near the ends of a chromosome, but that approximately
99% of the genome is efficiently replicated. Dean el al. (2002) showed that sequence
bias experienced during MDA is reduced in comparison to the PEP and DOP-PCR

methods.

Forensic Samples Used to Characterize I-PEP and WA

Understanding the practical limitations of I-PEP and MDA is necessary to
recognize when WGA may accurately amplify a forensic sample and when it may be a
waste of time, supplies, and most importantly, limited evidentiary material. To recognize
each method’s benefits and limitations in terms of DNA quantity and quality, high
molecular weight DNA was first used to optimize each method and then to test
progressively smaller amounts of whole genome amplified DNA. Next, a stock of male
DNA was digested and fractionated through gel electrophoresis to evaluate the practical
limitations of WGA on degraded DNA.

The use of DNA from human hair, skeletal material, and aged blood made it
possible to analyze WGA on samples common to forensic biology laboratories. Head hair
samples were those previously extracted by Graffy (2003), using the standard grinding
protocol described by the Armed Forces DNA Identification Laboratory (AF DIL)
(http://www.afip.org/Departments/oafine/dna/ afdil/protocols.html) followed by an
organic extraction. DNA from aged skeletal material was previously extracted by Misner
(2004) and Rennick (personal communication, February 18, 2005). One source of skeletal
material originated from a 3000 year old burial mound in Kamenica, Albania. Three
Albanian samples were utilized; 1 originating from a clavicle and 2 from femur. Rennick
(personal communication, February 18, 2005) had previously amplified a 266bp mtDNA
product from each of the bones. The second group of samples was extracted from bones
(5 femur samples), that had been buried for 140 to 170 years at the Voegtly Cemetery in
Pittsburgh Pennsylvania (Ubelaker and Jones, 2003). It had been previously determined

that up to 220bp of mtDNA could be amplified from these samples (Misner, 2004). The

DNAs from the Voegtly Cemetery and Albania burials were isolated using a standard
organic extraction which included grinding or drilling of the bone, an overnight
incubation of bone particles at 55°C in digestion buffer (20mMTris, 100mMEDTA, 0.1%

SDS) and proteinase K, followed by an organic extraction.

MATERIALS AND METHODS

Preparation of Artificially Degraded DNA

DNA was artificially degraded by using the endonuclease DNase I (Promega) to
obtain DNA of varying lengths. The enzyme randomly nicks DNA leaving a
phosphorylated 5’ end, and a hydroxylated 3’ end. To determine the length of time
needed to digest DNA, 5000ng of male DNA was added to lOul of 10X digestion buffer
(Promega), 5 units of DNase I and 55.5111 of water. Immediately, 5p] of the mixture was
removed and placed into a microcentrifuge tube containing 5111 of 20mM EDTA to stop
the enzymatic activity; which served as the control, undigested sample. The reaction
mixture was incubated at 17°C, where after 1, 2, 5, 10 and 30 minutes 15111 was
transferred to a new microcentrifuge tube containing 5111 of 20mM EDTA. Digested DNA
was denatured (95°C for 5 minutes), and separated through a 2% agarose gel, using a
100bp ladder (Promega) as a size reference. Based on the outcome of these timed
experiments, it was determined that the combination of a 1 and 2 minute digestion would
produce degraded DNA ranging from approximately 100 bases to DNA greater than 1500
bases in length.

The artificial degradation of DNA for WGA analysis was completed by mixing
750ng of male DNA, 1.5 ul of 10X digestion buffer, 0.75 units of DNase I and water for a
total reaction volume of 15111. The DNA was incubated at 17°C for 1 and 2 minute time
increments after which 7.5 ul of the reaction mixture was transferred to a new
microcentrifuge tube containing 2.51.11 of 20mM EDTA. For later trials of DNA digestion,

a 30 second incubation was also included; in these, 1125ng of male DNA was added to

10

 

 

 

 

2.25 pl of 10X digestion buffer and 1 unit of DNase I for a total reaction volume of
22.5 pl.

The degraded DNA was pooled into one microcentrifuge tube and denatured for 5
minutes at 95°C then chilled on ice. The entire volume was loaded onto a 2% low
melting point agarose gel. An 8cm long gel was used to separate the digested DNA from
trials 1 and 2. Greater separation was attempted for trials 3, 4 and 5 by using a 12cm gel.
For this longer gel, 3 2% gel of standard agarose was poured first and allowed to solidify
to act as a supporting gel. Then the middle section of the standard agarose was removed
to allow the low melting point agarose to be poured. DNA was separated in a cold room
at a constant amperage of 60mAmps, and stained with ethidiurn bromide. The DNA
smear was cut into six fiactionated sections of <200, 200 to 400, 400 to 600, 600 to 800,
800 to 1000 and >1000 nucleotides. Each section of gel was placed into labeled micro-
centrifuge tubes and heated to 65°C for approximately 5 minutes to liquefy the agarose.

See Organic Extraction of DNA below for the extraction of DNA from gel.

Digestion of Fresh Blood

Fresh blood samples of 1, 2, 5, and lOpl were placed into microcentrifuge tubes
and allowed to dry overnight. These were suspended in 300111 of digestion buffer (20mM
Tris, 100mM EDTA, 0.1% SDS) along with 3 pl of 20mg/ml Proteinase K and incubated
for 1 hour at 55°C. An organic extraction was then completed on each sample; see

Organic Extraction of DNA below.

11

Digestion of A ged Blood

In October of 2003, a sample of human blood was allowed to dry on a sterile Petri
dish and left at room temperature to age in a dark, dry environment. To characterize the
efficiency of the WGA methods on aged DNA, a flake of blood approximately 1mm by
1mm was taken from the 16 month old sample, suspended in 300111 of digestion buffer
and 3 pl of proteinase K, and incubated for 1 hour at 55°C. An organic extraction was

then completed; see Organic Extraction of DNA below.

DNA from Aged Skeletal Material and Hair

To characterize the effectiveness of WGA on aged skeletal material, five Voegtly
DNA samples (burials 27F, 114F, 249F, 448F, and 704F) and three Albanian samples
(burial 76F, 198F and 198C) were tested. Five hair samples (24, 27, 28, 29 and 36) were

also used for WGA analysis, see Introduction.

Organic Extraction of DNA

Artificially degraded DNA and DNA from blood were organically extracted. To
the digestion buffer (blood) or melted agarose (artificially degraded DNA), an equal
volume of phenol was added, vortexed for approximately 10 seconds, then centrifuged
for 5 minutes at 12,000rpm. A second phenol extraction was completed if the sample
appeared reddish or cloudy. The aqueous portion was transferred to a new
microcentrifuge tube, where an equal volume of chloroform was added, vortexed for 10
seconds then centrifuged for 5 minutes at 12,000rpm. The aqueous portion was
transferred to a tube where DNA was precipitated using two volumes of cold 95%

ethanol and 10% 3M sodium acetate for at least one hour at -20°C. The samples were

12

centrifuged at 12,000 rpm, 4°C for 15 minutes, ethanol was removed, and pellets were
washed with 200 pl of cold 70% ethanol, followed by a final centrifugation (5 minutes at
4°C, 12,000rpm). Ethanol was removed and the pellet was vacuum dried for 10 to 15
minutes. Artificially degraded samples were resuspended in TE (lOmM Tris, lmM
EDTA) bringing the final estimated DNA concentration to 10ng/ul, based on the amount
of DNA added to the initial digestion and the assumption that DNA was evenly
distributed when separated into fi'actions. Aged and fresh blood samples were

resuspended in 15111 of TE.

Whole Genome Amplification

Artificially degraded DNA was whole genome amplified using an estimated 2ng
of input DNA. One microliter of DNA from blood (fresh and aged) was whole genome
amplified. DNA from hair was whole genome amplified using 1111 of a 20 fold dilution.
One microliter of DNA from the Albanian skeletal material was whole genome
amplified. DNA fiom the Voegtly samples was whole genome amplified using 1 ul of a
20 fold dilution. The hair and bone dilutions were based on the optimal amounts of DNA
described by Graf’fy (2003), Rennick (personal communication, February, 18, 2005) and
Misner (2004), respectively. WGA reactions with no input DNA were processed as

negative controls.

Improved Primer Extension Preamplification
I-PEP WGA was performed as described by Dietrnaier et al. (1999), with the
exceptions of digestion buffers; see Digestion of Fresh Blood above. Using an Expand

High Fidelity PCR system (Roche Applied Sciences) each reaction contained sample

13

DNA, 20uM random 15 base primer, 0.2mM dNT P, 1X Expand High Fidelity Buffer
without magnesium, 2.5mM MgClz, and 0.75 units of Expand High Fidelity Enzyme mix,
in a total reaction volume of 10p]. When WGA was performed on DNA from hair or
bone, 1 ul of BSA (3 rig/pl stock) was included in the reaction mixture. The amplification
process was performed in an ABI Gene Amp PCR system 2400. Amplification included
an initial DNA denaturation for 1 minute at 94°C, followed by 50 cycles of 94°C for one
minute, 37°C for 2 minutes, 0.1°C per second temperature ramp from 37°C to 55°C, 4

minutes at 55°C, and 68°C for 30 seconds.

Multiple Displacement Amplification

MDA was performed using a GenomiPhi DNA amplification kit (Amersham
Biosciences). DNA was added to 4.5111 of Sample Buffer and then denatured at 95°C for 3
minutes. Samples were immediately cooled on ice, then 4.5111 of Reaction Buffer and
0.5 ul of enzyme was added to the denatured sample and incubated at 30°C for 18 hours in
an Eppendorf Mastercycler Gradient thermocycler. The enzyme was inactivated by

heating the sample to 65°C for 10 minutes.

Purification of Whole Genome Amplified Sample

Excess primer and unincorporated nucleotides were removed fi'om whole genome
amplified samples using a Microcon YM-100 column (Millipore). Each sample was
transferred to a column along with 300p] of TE and centrifuged for 12 minutes at 500x g.
A total of three washes, (3 00111 of TE per wash), were performed. The final volume was

returned to lOpl, the original WGA volume

14

PCR Amplification

PCR amplification was undertaken on untreated samples and whole genome
amplified product, including negative controls, using both mitochondrial and nuclear
DNA primers (see Tables 1 and 2). The mtDNA primers are commonly used in forensic
DNA analysis, and were designed to amplify sections of the mtDNA control region. The
nuclear specific primers were designed to amplify a section of the amelogenin gene——a
single copy gene. The first PCR reaction was performed using primers designed to
produce a small amplicon. If the amplification was successful additional PCR
amplifications were attempted using different primers designed to produce larger product
lengths. Tests comparing the quality and quantity of DNA from hair and bone were
performed using the same primer pairs utilized by Graffy (2003), Misner (2004), or
Rennick (personal communication, February 18, 2005). PCR reactions consisted of 1X
Hot Start PCR buffer (Eppendorf), 0.2mM dNTP, 2 pmol of each primer, and 1 unit
HotMaster Taq (Eppendorf) to a final reaction volume of 20ul for sequencing reactions or
10111 for all other reactions. Amplification reactions of DNA from bone also included 2 pl

of3ug/ul BSA.

Table 1: Nuclear DNA Primers

    
  

   

5,

  
 

Reverse
Reverse

Column name.
the PCR when used with ‘Forward
(Nakahori et al., 1991; Rosen and Skaletsky, 2000).

                 

reverse name to of
. Column 2 indicates the primer’s sequence

  
   

Table 2: Mitochondrial DNA Primers

 

 

 

 

 

 

 

 

 

 

Primer Name Primer Sequence

Forward 15 5’ CACCCTATTAACCACTCACG 3’
Forward 82 5’ ATAGCATTGCGAGACGCTGG 3’
Forward 155 5’ TATTTATCGCACCTACGTTC 3’
Reverse 285 5’ GTTATGATGTCTGTGTGGAA 3’
Reverse 485 5’ TGAGATTAGTAGTATGGGAG 3’
Forward 16190 5’ CCCCATGCTTACAAGCAAGT 3’
Forward 15989 5’ CCCAAAGCTAAGATTCTAAT 3’
Reverse 16144 5’ TGACCACCTGTAGTACATAA 3’
Reverse 16410 5’ GAGGATGGTGGTCAAGGGGAC 3’

 

 

 

 

Column 1 displays the primer name and location on the human mitochondrial genome where the primer’s
5’ end anneals (Anderson et al., 1981). Column 2 gives the primcr’s sequence, (AFDIL,
http://www.afip.org/ Departments/oafine/dna/afdil/protocols.html).

Table 3: Primer Pairs Used for PCR Amplification and Sequencing

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Artificially Degraded DNA

Forward Amel/ Reverse Amel 200 213bp Nuclear

Forward Amel/ Reverse Amel 496 496bp Nuclear

Forward 16190/ Reverse 16410 220bp Mitochondrial

Forward 15/ Reverse 285 270bp Mitochondrial

Forward 15989/ Reverse 16410 421b Mitochondrial

Forward 16190/ Reverse 484 863bp Mitochondrial

Fresh Blood

Forward Amel/ Reverse Amel 200 213 bgp Nuclear

Forward Amel/ Reverse Amel 496 496bp Nuclear

Forward 15989/ Reverse 285 1064bp Mitochondrial

A ed Blood

Forward Amel/ Reverse Amel 200 213bp Nuclear

Forward Amel/ Reverse Amel 496 496bp Nuclear
Aged Skeletal materiaL—Voefly Cemetery

Forward 16190/ Reverse16410* 220bp Mitochondrial

Forward 155/ Reverse 484 329bp Mitochondrial

A ed Skeletal Material—Albanian Burial Mound

Forward 16144/Reverse16410* 1 266bp I Mitochondrial

Hair

Forward 82/ Reverse285* 203bp Mitochondrial

Forward Amel/ Reverse Ame1200 213bp Nuclear

 

 

 

 

Column 1 includes the primer pair used to amplify the sample. Column 2 displays the length of the PCR
product. Column 3 is the type of DNA amplified by the primer pair. "' indicates the primer pair was used in
previous studies. (Graffy, 2003; Misner, 2004; Rennick, unpublished)

l6

Whole genome amplified DNA was serial diluted from 10 to 1,000,000 fold to
determine the maximum dilution possible while still generating a positive result—an
indication of starting DNA quantity. Amplification dilutions which were determined to
produce the brightest band (greatest amount of PCR product) were used as a guideline for
further analyses, (maximum PCR amplicon length, DNA sequencing, or STR analysis),
however if the amplification using these dilution guidelines failed, a second amplification
was attempted with more DNA. The quality of whole genome amplified product was
tested using a series of primers which are designed to produce progressively longer PCR
products. Cycling parameters for nuclear single locus amplifications were 94°C for 2
minutes, followed by 38 cycles of 94°C for 1 minute, 58°C for 45 seconds, and 72°C for
45 seconds. Cycling parameters for mtDNAs were 94°C for 2minutes, followed by 38
amplification cycles of 94°C for 30 seconds, 56°C for 45 seconds and 72°C for 45
seconds. A final extension of 72°C for 5 minutes was added to the end of all PCR
programs. Five microliters of PCR product was loaded onto an agarose gel; for PCR
products of 400 base pairs or less a 2% agarose gel was utilized while a 1.5% agarose gel

was utilized for larger PCR products.

ST R Analysis of Blood and Artificially Degraded DNA

STR analysis was conducted using an Applied Biosystems Identifiler kit.
Reactions consisted of 4111 AmpFlSTR PCR reaction mix, 2111 AmpFlSTR Identifiler
Primer set, 1 unit AmpliTaq Gold, and DNA in a final volume of 10111. For artificially
degraded DNA size classes, 1111 of untreated DNA diluted 10 fold, and 11.11 of undiluted
WGA product was PCR amplified. For the PCR amplification of STR loci from fi'esh

blood, 1111 of untreated DNA diluted 10 fold, 1 111 of neat IPEP product, and 1111 of MDA

l7

product diluted 1000 fold was amplified. STR amplification from aged blood was started
with 1111 of neat untreated DNA, and 1111 of WGA product diluted 10 fold. The PCR
parameters were 95°C for llminutes, 35 amplification cycles of 94°C for 1 minute, 59°C
for 1 minute and 72°C for 1 minute, followed by a final extension step of 60°C for 60
minutes. Five microliters of PCR product was loaded onto a 2% agarose gel for
amplification confirmation. For analysis, 1.511] of amplified product, 24.5111 of
formamide and 0.5111 of ABI GeneScan 500 Liz size standard were heat denatured (95°C
for 3 minutes) then chilled on ice. STR profiles were generated on an ABI Prism 310
Genetic Analyzer (GS STR POP4 (lml) G5.md5 module, 5 second injection, 15kV
injection, 15kV run voltage, 28 minute run time) using ABI 310 Genetic Analyzer Data
Collection Software version 3.0.0. ABI GeneMapper ID, version 3.1 software was used

to analyze data.

Mitochondrial DNA Sequencing of Bone and Hair Samples

Amplified mtDNA was purified using a Montage PCR Centrifugal Filter Device
(Millipore) (one 400111 rinse of TE, 1000x g). DNAs were sequenced using a CEQ DTCS
Quick Start Kit (Beckman Coulter). Sequencing reactions consisted of 4111 DTCS Quick
Start Master Mix, 211M primer and 50 to 100fmol PCR product, with a final volume of
10111. The sequencing parameters were 30 cycles of 95°C for 20 seconds, 50°C for 20
seconds, and 60°C for 4 minutes. Each reaction was stopped with 2.5111 of a stop solution
(1.2M sodium acetate, 20mM EDTA, 4mg/mL glycogen). The DNA was precipitated in
60111 of 95% ethanol then centrifuged immediately at 14,000rpm for 15 minutes. Ethanol
was removed, and pellets were rinsed twice with 200111 of 70% ethanol followed by

centrifugation at 14,000rpm for 2 minutes. DNA pellets were vacuum dried for 15

18

minutes then resuspended in 40111 of deionized formamide. DNA sequences were
obtained on a Beckman Coulter CEQ 8000 Genetic Analysis System (120 second
denaturation at 90°C, 15 second injection, 20kV injection voltage, 4.2KV run voltage, 60
minute separation time), using CEQ 8000 Genetic Analysis System software, version 8.0.
Sequences were aligned using Bio Edit software (Hall, 1999) and the Anderson reference
sequence (Anderson et al., 1981). These results were ultimately compared to sequences

previously obtained by the original researcher.

19

RESULTS

The characterization of I-PEP and MDA was completed using various forensic
sample types. Whole genome amplified product was compared to untreated DNA in
terms of quality, (whether there was a change in the maximum PCR product length),
quantity (which dilution factors successfiilly amplified DNA), and downstream analysis,
including DNA sequencing and STR testing. Figure 2 presents an image of an agarose
gel of whole genome amplified product. Notice that a smear of DNA is seen in each lane,
even when no DNA was added to the WGA reaction. Also, the MDA lanes are brighter

than the I-PEP lanes, and the positive lanes are slightly brighter than the reagent blanks.

Figure 2: Image of I-PEP and MDA Whole Genome Amplified Product

 

Lane 1 is a MDA reagent blank. Lane 2 is a MDA reaction with DNA added. Lane 3 is an I-PEP reagent
blank. Lane 4 is an I-PEP reaction with DNA added.

20

Quality of l—PEP and MDA Product from Artificially Degraded DNA

The maximum nuclear and mitochondrial PCR amplicon that could be obtained
from artificially degraded fiactions was determined. DNA was degraded with DNase I
five times, four with DNA from a commercial stock (which was created from a mixture
of sources; Trials 1 through 4) and once from a single source of DNA (Trial 5). Figure 3
displays an image of DNase I digested DNA separated by gel electrophoresis. Notice the
digested DNA is visible below the 100bp marker and extends to the top of the image,
near the 1500bp band, and that the intensity of the DNA becomes progressively weaker
as the length of the digested DNA becomes longer. Table 4 displays results for maximum
PCR product length experiments (nuclear or mtDNA) between fractionated DNA and

fractionated DNA following both methods of WGA.

Figure 3: DNase I Digested Male DNA

 

Lane 1 is the ladder. Lane 2 is digested DNA.

21

When testing untreated DNA from trial 1, a 496bp nuclear product was generated
from all size classes except the <200 and >1000 fractions. A 496bp nuclear product was
only generated from the >1000 size class from trial 2 and the 600 — 800, and 800 — 1000
sized classes from trial 4. The 496bp amplification was unsuccessful from all trial 3 size
classes and was not attempted in trial 5. A 213bp nuclear product was amplified from
untreated DNA in all but six reactions throughout all trials and size fractions. The
amplification success of the 213bp nuclear product across the five trials of DNA does not
seem directly related to size class; the 200 - 400, 400 — 600 and >1000 size amplified 5
out of 5 times, the 600 - 800 size amplified 4 out of 5 times, the 800 - 1000 size
amplified 3 out of 5 times, and the <200 size amplified 2 out of 5 times. An 865bp
mtDNA product was generated in all size classes from trials 1 and 2 but not from
subsequent trials. A 270bp mtDNA product was generated fi'om four of the trial 3 size
classes while DNA from the <200 and the 600 — 800 size class had a maximum product
of 220bp. All size classes from trial 4 had a maximum mtDNA product of 220bp. From
trial 5, 270bp was the maximum mtDNA product length generated from all size classes.
There does not seem to be a direct relationship between size class and amplification
success of a 270bp mtDNA product across the 5 digestion trials; the 200 — 400 class
amplified 4 out of 5 times, the 400 — 600, 800 — 1000 and >1000 sizes amplified 3 out of
5 times, the 600 — 800 size amplified 2 out of 5 times, and the <200 size amplified 1 out
of 5 times.

When testing I-PEP treated DNA, a 496bp nuclear product was successfully
amplified from all trial 1 size classes. A 496bp nuclear product was only generated fi'om

the >1000 fraction from trial 2, and the 600 - 800, and 800 — 1000 size classes from trial

22

Table 4: Nuclear and mtDNA Amplification of Artificially Degraded DNA

Fraction sizes are located at the beginning of each row, followed by the attempted PCR product. 'Nuc'
indicates primers were nuclear specific, ‘mt’ indicates primers were mtDNA specific. Numbers following
the DNA type indicate the PCR product size. Treatments (untreated, I-PEP or MDA) and degradation trial
numbers are specified across the top of columns. + indicates successful amplification, - indicates PCR
failure. [-] indicates amplification was not successful, but multiple nonspecific bands were also present. [+]
indicates amplification not successful, but multiple nonspecific bands were also present. No data available
is indicated as n.d., largest mt indicated maximum mtDNA amplicon obtained

23

3/5
5/5 3/5

5/5 4/5 4/5
4/5 5/5 3/5
3/5 5/5 3/5

2/5 3/5

VDM

VDM

   
   

(13111 '1

VDM

(ll'ld'l

VD M

(Ii'IrI'I

VDM

VDM

002> 007'002 009'009 008‘009 000! '008 000|<

24

4 (no 496bp nuclear amplifications were attempted from trial 5). A 213bp nuclear
product was amplified from I-PEP DNA in all but three reactions throughout all trials and
size fiactions. The three reactions which did not amplify were from fractions less than
400 nucleotides in length. The ability to replicate nuclear PCR amplification results
using I-PEP treated samples from all trials was related to the fraction size. DNA from all
fractions 400 bases and larger were successfully amplified from each 213bp nuclear DNA
attempt. Successful PCR amplification of a 213bp product fi'om the <200 and 200 - 400
nucleotide size classes occurred 3 and 4 times out of 5, respectively.

An 865bp mtDNA product was generated from all trial 1 and 2 size classes alter
I-PEP. Size fractions greater than 800 nucleotides from trial 3 had a maximum mtDNA
product of 270bp; all other reactions from this trial had a maximum product length of
220bp. Fractions from trial 4 greater than 400 nucleotides had a maximum mtDNA
product of 270bp, while all others had a maximum product of 220bp. All size classes
from trial 5 had a maximum mtDNA product length of 270bp. The replication of mtDNA
PCR amplification results using I-PEP treated samples from all trials was related to the
fraction size. All attempted amplifications of a 270bp mtDNA amplicon were successful
from the 400 — 600, 800 — 1000 and >1000 size classes. Successful PCR amplification of
a 270bp product fi'om the 200 - 400 and 600 — 800 size classes occurred 4 times out of 5
times, and amplifications from the <200 class were successful 3 out of 5 times.

The amplification of a 496bp nuclear product fi'om MDA treated DNA was
successful from the <200, 200 — 400, 600 - 800, >1000 size classes from trial 1, and the
>1000 size class from trial 2. A 213bp nuclear product was generated from all size

classes from trial 1 and 2. A maximum nuclear product length of 213bp was generated

25

from the 200 — 400, 400 — 600, 600 — 800 size classes from trial 3, and the 200 - 400, 400
- 600, 800 - 1000, and >1000 size classes from trial 4. No nuclear product was generated
from the DNA digested in trial 5. There was no direct relationship between WA size
classes and amplification success of a 213bp nuclear product across the 5 trials; 3 213bp
nuclear DNA product was amplified from the 200 — 400 and 400 — 600 size classes 4 out
of 5 times, the <200, 600 — 800, 800 — 1000, and >1000 MDA nucleotide fractions was
successful 3 out of 5 attempts. Multiple non-target DNAs were amplified in many of the
PCR reactions of MDA product; these bands were not observed in reagent blank PCR
reactions.

MDA treated DNA fiom trial 1 was not tested for an 865bp mtDNA product due
to multiple bands in the 213bp nuclear PCR reactions, the largest mtDNA product tested
was 270bp. Trial 2 had a maximum mtDNA product length of 865bp in all size classes,
while the maximum mtDNA product from trials 3 and 4 was 270bp. No mtDNA
reactions were attempted from trial 5 because no nuclear product was obtained. The
ability to reproduce mtDNA PCR amplification results using MDA treated samples from
all trials was not directly related to the fi'action size; each size class was successfully
amplified 2 out of 4 times.

One inconsistency observed across trials of digestion was the larger PCR products
generated from trials 1 and 2 in comparison to trials 3, 4, and 5. A 496bp nuclear product
was amplified in all size classes from trial 1, and as seen in the ‘largest mt’ rows of Table
4, an 865bp mtDNA was generated in all size classes in untreated and I-PEP samples
from trial 1 and 2, and in all size classes of MDA samples fi'om trial 2. DNA digested in

trials 3, 4 and 5 had a maximum mtDNA product of either 220 or 270bp.

26

Quality of I-PEP and WA Product using DNA fi'om Fresh and Aged Blood

Table 5 displays the results of PCR amplifications from fresh and aged blood
samples with and without WGA. A 496bp nuclear product was PCR amplified from
untreated and whole genome amplified DNA extracted from 1, 2, 5, and 10111 of fresh
blood. As amplification was successful from all blood volumes, all further tests were
completed on the 1111 sample. All amplifications using DNA from fresh blood were
successful, testing a maximum nuclear DNA product of 496bp and a maximum mtDNA
product of 1065bp. The longest nuclear and mtDNA PCR product obtainable from
untreated aged blood was 496bp and 865bp, respectively. After I-PEP and MDA, a
496bp nuclear, and 1064bp mtDNA amplicon was amplified. Multiple bands were

observed in the MDA nuclear DNA amplification reactions.

Table 5: Amplification of Nuclear and Mitochondrial DNA fiom Fresh and Aged Blood

 

 

 

 

 

 

 

 

 

 

O- Q. a
i s 5“ a s
N v- 3 c O
8 8 3 1'3 :
Z Z 8 a e
NOWGA + + nd nd +
a m. + + ...,d
E
,U NOWGA - + ‘ _ + _

<

The top row displays the attempted nuclear (nuc) or mitochondrial (mt) PCR product lengths. + indicates
successful amplification, - indicates PCR failure. [+] indicates amplification was successful, but multiple
nonspecific bands were also present. ‘nd’ indicates that no data is available since the amplification was not
attempted. White rows are results from DNA without WGA, light gray rows are results from l-PEP whole
genome amplified DNA, dark gray rows are results from MDA whole genome amplified DNA

27

Quality of I-PEP and MDA Product using DNA fiom Hair Shafts

Table 6 displays the results of attempted DNA amplifications from hair shaft
samples with and without WGA. No nuclear DNA could be amplified from the untreated
or whole genome amplified hair samples. According to Graffy (2003), the five DNA
samples extracted from hair had a maximum mtDNA amplicon size of 865bp. After I-
PEP, maximum mtDNA PCR product length was reduced for all five samples; DNA fi'om
two samples (number 24 and 27) had a maximum amplicon length of 664bp, while three
had a maximum product length of 203bp (number 28, 29, and 36). The amplification of
mtDNA from all MDA samples was unsuccessful but multiple non-target DNAs were

amplified.

Table 6: Amplification of Mitochondrial and Nuclear DNA from Hair Shafts

mtDNA nucDNA

     
   

  
 
 

#
24
27
28
29
36 + + +

DNA type is indicated at the top of each column, followed by DNA treatment, the third row displays the
attempted nuclear (nuc) or mitochondrial (mt) product lengths, + indicates successful amplification, -
indicates amplification failure. [-] indicates amplification was unsuccessful, but multiple non-target bands
were generated. Sample ID and columns marked with a " indicates data are reproduced from Graffy (2003)

 
      
 

  
     

      

Quality of I-PEP and WA Product fiom Aged Skeletal Material
Given the age and highly degraded nature of the skeletal samples, nuclear DNA

was not tested. Attempted PCR amplifications of mtDNA from aged skeletal material are

28

shown in Table 7. Misner (2004) determined that up to a 220bp mtDNA product could
be amplified from the Voegtly bones; Rennick (personal communication) successfully
amplified a 266bp mtDNA product using nested PCR from the Albanian samples; these
results were used to compare maximum product lengths generated from whole genome
amplified DNA. Of the eight I-PEP samples, only one was successfirlly PCR amplified
(Voegtly bone sample 704; Table 7, column 4; Figure 4). MDA product was determined

to be unusable due to the high degree of non-target bands after PCR amplification.

Table 7: Amplification of Mitochondrial DNA from Aged Skeletal Material

    

Treatment type on followed by the attempted mtDNA PCR product
length“ indicates data were obtained from Misner (2004), ” data obtained from Rennick (personal
communication), + indicates successful arnplificatiom- indicates amplification failure, N indicates nested
PCR was completed to obtain a positive result, [+] indicates sample amplification was successful, but
multiple nonspecific bands were present.

Figure 4: 220bp mtDNA Product fi'om MDA and I-PEP Treated Voegtly Cemetery Bone
Samples

irl‘tl VHNI Will — ' ‘V‘ 371"! H"!

'\ /

 

Lanes 1 throughlZ are PCR amplified aged bone samples which underwent MDA. Lanes 12 through 24
are PCR amplified I-PEP aged bone samples. Each sample is present twice, the first was PCR amplified at
10 fold dilution, the second at a 100 fold dilution. Arrows denote positive amplifications.

29

DNA Yield Comparisons using Serial Dilutions

Relative DNA concentrations before and after WGA were examined by
conducting 10 fold dilutions until PCR amplifications were no longer successful. Dilution
factors presented below are in reference to 1111 of neat (undiluted) untreated DNA. Since
WGA was completed in a 10111 reaction (where 1111 of template was added to 9111 of
reaction mixture), 1111 of whole genome amplified product flom fresh and aged blood
began 10 fold more dilute than 1111 of untreated DNA. DNA flom hair that was to be
treated with I-PEP was first diluted 20 fold, based on Graffy (2003). Taking this and the
10 fold dilution from the WGA procedure into account, I-PEP hair samples ended up
being 200 times more dilute than the original hair DNA. No yield comparisons were
completed on MDA product flom hair because quality experiments were unsuccessful.
No yield comparisons were conducted on WGA product flom aged skeletal material since

only one sample amplified.

Quantity of I-PEP and WA Product using DNA fiom Fresh and Aged Blood

Table 8 displays the amplification results of untreated, I-PEP and MDA samples
flom flesh and aged blood after serial dilutions. PCR amplification of untreated DNA
flom flesh blood was successful when input PCR DNA was diluted 10 fold. The PCR
amplification of I-PEP DNA flom flesh blood was successful when DNA was 1000 times
more dilute than untreated DNA, while MDA DNA 10,000 times more dilute than
untreated DNA was successfully amplified. Figure 5 displays an image of a 496bp

nuclear product flom serial diluted untreated, I-PEP and WA DNA.

30

Figure 5: 496bp Nuclear PCR Product of I-PEP, MDA and Untreated DNA Diluted flom
Fresh Blood

1 2 3 4 5 6 7 8 9 10 11

  

Lanes 1, 2, 3 are I-PEP samples flom flesh blood at 1000, 10,000, 100,000 fold dilutions, respectively.
Lanes 4, 5, 6, 7 are MDA samples flom flesh blood at 1000, 10,000, 100,000, 1,000,000 fold dilutions,
respectively, Lanes 8 and 9 are untreated DNA at 10 and 100 dilutions, respectively, Lane 10 is a positive
control, Lane 11 is a negative control.

Without WGA on DNA flom aged blood, a 496bp nuclear DNA product was
amplified using 1111 of sample. After WGA, PCR amplification of I-PEP product 100
times and MDA product 1,000 times more dilute than untreated DNA was successful. As
seen in Figure 6, diluted whole genome amplified products were successfully amplified,
but extra bands were generated in the MDA reactions.

Table 8: PCR Amplification of Nuclear DNA (496bp product) flom Fresh and Aged
Blood

 

 

 

 

 

 

 

 

 

 

 

 

o § §.
.. o 8 § § 8“ 8.
i S s s s s s
[No WGA + + - - - - -
fi I-PEP .- + + + - ..
11..
No WGA + -
'3 IoPEP . -- + I+“ s
2:“

The top row displays the attempted dilutions, sample types are displayed in the first column, + indicates
successful amplification, - indicates amplification failure. [+] indicates amplification was successful, but
multiple nonspecific bands were also present. - -indicates the reaction is not possible due to the 10 fold
dilution flom the WGA procedure. White rows are results florn DNA without WGA, light gray rows are
results flom I-PEP DNA, dark gray rows are results from MDA DNA.

31

Figure 6: 496bp Nuclear PCR Product of I-PEP and MDA DNA Diluted from Aged
Blood

 

Lane 1: positive control; lane 2: I-PEP reagent blank; Lanes 3 to 5: I-PEP DNA from aged blood at 1/10,
1/ 100, 1/1000 fold dilutions, respectively; Lane 6: MDA reagent blank; Lanes 7 to 9: MDA DNA from
aged blood at 1/ 10, l/ 100, 1/1000 fold dilutions, respectively.

Quantity ofl-PEP and WA Product using DNA from Hair

Yield experiments on DNA flom hair were performed on mtDNA since it is the
only type of DNA that can be regularly amplified flom hair shafts. Experiments testing
the quantity of I-PEP product were compared to dilutions attempted by Graffy (2003);
results are displayed in Table 9. MDA quantity experiments were not attempted since
multiple bands were observed in initial PCR reactions. Without WGA, PCR product was
obtained from two samples diluted 100 fold, (numbers 24, and 29) the remaining three
samples (number 27, 28 and 36) were amplified with a 10 fold dilution. Alter I-PEP, PCR
product flom one sample (number 29) was amplified flom starting DNA 20 times more
dilute than untreated DNA PCR product flom two hair samples, (numbers 24 and 36)
could be obtained flom starting DNA 200 times more dilute, and two samples (numbers
27 and 28) 2000 times more dilute than the same sample without WGA As indicated in
the I-PEP columns of Table 9 and in Figure 7, some PCR reactions were inhibited in the

more concentrated reactions.

32

Table 9: Amplification of Serial Diluted mtDNA flom Hair Shafts

   

The top row of each column successful amplification, -
indicates amplification failure. I indicates amplification was inhibited. Sample ID and data marked with a “
were reproduced flom Graffy (2003).

Figure 7: 203bp mtDNA PCR Product of I-PEP Diluted DNA flom Hair Shafts

 

Lanes 1 through 3 are sample 24, diluted 1/10, 1/ 100, 1/1000 respectively. Lane 4 through 6 are sample 27
diluted, 1/10, 1/100,1/1000 respectively. Lane 7 through 9 are sample 28, diluted 1/10, 1/100, 1/1000
respectively. Note the reactions which contain no product also show weak or no primer bands. The arrow
points to positive bands.

ST R analysis of Artificially Degraded DNA and Fresh and Aged Blood Samples

The ABI Identifiler kit is a multiplex of primers designed to amplify 16 loci (l3
CODIS STR loci, 1 gender marker, and 2 additional STR loci) ranging in length flom
100bp to 360bp. STR analysis was completed on artificially degraded DNA, flesh and
aged blood samples using successful dilutions flom initial experiments as DNA input
guidelines. However, in all cases a second amplification was needed in order to add

more DNA into these multiple loci amplification reactions. STR testing on artificially

33

degraded DNA was completed using product from the fifth trial of degradation (Table
10). Un-degraded DNA from this source was used as the reference sequence.

More loci were amplified fiom untreated artificially degraded fractions than
whole genome amplified fi'actions. The untreated artificially degraded DNA fiom the
400 — 600, 600 - 800, and 800 — 1000 fractions matched the known profile at 14, 16 and
15 loci, respectively. Complete amplification of both alleles from the 200 — 400 size
class occurred at 7 loci, partial profiles were obtained for 3 loci, and 4 loci were not typed
due to multiple peaks per locus. No loci amplified from the untreated >lOOO nucleotide
fiaction. I-PEP DNA from the >1000 fraction, matched the known profile at four loci,
and an incomplete profile was generated at seven additional loci. MDA DNA from the
>1000 fraction matched at nine loci, and a partial profile was obtained from 2 loci. The
alleles fi'om whole genome amplified DNA which matched the known profile were fiom
loci of shorter product length in the multiplex; the longest loci in the multiplex are
CSFlPO, DZSI338, and D18851. No profiles were generated from artificially degraded

DNA fractions below 1000 nucleotides after WGA.

34

Table 10: STR Profiles of DNA Source

    

 

Profile

 

type across top
name of each locus. Alleles highlighted in gray matched the known profile. AMEL is an abbreviation for
amelogenin, the sex marker.

Five of the thirteen STR loci from untreated fresh blood, diluted 10 fold, matched
the known sample (Table 11). Nine additional loci displayed false homozygous peaks
resulting from allele drop-out, and no results were obtained at 2 loci (Figure 8). There
was no relationship between amplification success and locus length. STR tests of I-PEP
DNA fi'om fresh blood, using 1 pl of WGA product, matched the known profile at all 16
loci (Figure 9). MDA product from fi‘esh blood, diluted 1000 fold, matched at 14 of the
16 loci, (Figure 10). False homozygous peaks occurred at two loci, D3Sl358 (a small
locus) and TPOX (a medium sized locus).

No STR profile was generated from untreated aged blood samples even when 2111
of sample (double the amount from prior experiments) was added to the amplification
reaction. STR analysis of I-PEP DNA from aged blood matched at 2 loci, D19S433 and
Amelogenin, which both have a short PCR product length. Partial profiles for 2
additional loci (D881 179 and D5 8818; both short loci) matched the reference profile.
Aged blood DNA which underwent MDA matched the known sample profile at one

allele (D881 179).

35

Table 11: STR Profiles of DNA fiom Fresh and Aged Blood with and without WGA

      

      

name treatment across top
column, followed by the input dilution for amplification. Alleles highlighted in gray matched the known
profile. AMEL is an abbreviation for amelogenin, the sex marker. Number not highlighted indicates a small
peak was present, but no allele was called.

36

Figure 8: Electropherogram of Untreated DNA from Fresh Blood

 

:r
I ‘j

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H ‘k ‘&* A L

Locus name is located in the gray box above corresponding peaks. Number of repeats per allele is
displayed below each peak. The amelogenin locus is indicated as ‘ . . .’ PCR product length increases from
left to right across each row. Note the number of loci which experienced allele drop out, as indicated by
arrows. Also note that product length was unrelated to PCR amplification success.

 

37

Figure 9: Electropherogram of DNA from Fresh Blood afier I-PEP WGA

 

ix 3:1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100 200 300

l

 

Locus name is located in the gray box above corresponding peaks. Number of repeats per allele is
displayed below each peak. The amelogenin locus is indicated as ‘ . . . ’ PCR product length increases from
left to right across each row. Note an extra peak at D19S433, denoted by arrow.

38

Figure 10: Electropherogram of DNA from Fresh Blood after MDA WGA

 

 

 

 

mm! E1311 1M“... 3
290 A 390

A
‘r

ir

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Locus name is located in the gray box above corresponding peaks. Number of repeats per allele is
displayed below each peak. The amelogenin locus is indicated as ‘ . . .’. PCR product length increases from
left to right across each row. Notice that at locus D3Sl358, only one allele (15) matches the reference
profile, the 15.2 peak is possible pull up due to spectral overlap of primer dyes. Also, notice allele dropout
at TPOX. Both partial matches are denoted by arrows.

39

Sequencing of mtDNA fi'om Hair Shafi and Aged Skeletal Material

Table12 displays polymorphisms in the I-PEP whole genome amplified hair
samples in comparison to a reference sequence (Anderson et al., 1981) and to the
sequences reported by Graffy, (2003). PCR amplification was successful for 4 of the 5
hair samples after I-PEP. DNA sequences were obtained from these 4 I-PEP hair
samples. Polymorphisms in the I-PEP samples were consistent with those reported in
Graffy (2003), except that at three locations the I-PEP sample showed a dual peak and
Graffy (2003) observed a single peak. In one sample (sample 29), Graffy (2003) reported

a possible difference; after WGA no difference was noted from the reference sequence.

40

Table 12: Mitochondrial Sequence Comparisons of Untreated and IPEP DNA from Hair
Shafts

 

 

 

 

 

 

Sample ID No WGA" I-PEP
24 97-285 94-279
185A 185A/G
188G 188G
288A 288A/G
263G 263G
27 85-284 141-249
263G n.d.
28 90-274 88-184
-- l75C/A
185G
188G
195N
228A
263G
29 84-284 82-286
146C 146C
152C 152C
195C 195C
209N --
249G 249G
263G 263G

 

 

 

 

Sample ID and the column marked with a "' reproduced from Graffy (2003). The first line of data in bold
for each sample is the range of the mtDNA genome sequenced, as numbered by Anderson et al. (1981).
Differences from the reference sequence are indicated afier the DNA location. ‘n.d.’ indicates that no
polymorphisms occurred in the range sequenced, -- shows that there were no differences from the reference
sequence.

Misner (2004) sequenced the five Voegtly samples, and Rennick (unpublished
data) sequenced the three Albanian samples (Table 7). PCR amplification of 7 out of 8
aged bone samples were unsuccessful after I-PEP and MDA. One aged skeletal sample
(Voegtly sample 704) was amplified afler I-PEP. Sequencing of this sample showed no
polymorphisms in this sample’s sequence, which was consistent with results reported by

Misner (2004).

41

DISCUSSION

Limited DNA can be problematic for forensic biologists. WGA, a process which
randomly amplifies an entire DNA sample, has been studied in great detail on clinical
biopsy material, yet there is a lack of information regarding the usefulness of WGA on
forensic samples. This project characterized two methods of WGA on DNA isolated fiom
tissues commonly seen in forensic casework. I-PEP and MDA were tested on high
quality DNA from fresh blood, DNA that was artificially degraded to obtain controlled

fragment lengths, and DNA from hair shafts, aged blood, and aged bone.

Advantages and Disadvantages of I-PEP and WA

I-PEP follows a straightforward procedure which uses a reaction mixture similar
in composition to PCR, and once set up, the amplification process takes no additional
hands-on time. The cycling process itself is time consuming however, due to the slow
0.1°C per second temperature ramp included in each of the 50 cycles. In a field which is
already backlogged, this 12 — 13 hour step seems counterproductive, but when the
amplification was allowed to proceed overnight, traditional protocols could be resumed
the following morning. Currently, no kits are available to simplify the I-PEP setup, but
buffer, dNTPs and random primers could easily be packaged into one tube to reduce
hands on time even further.

MDA is a simple reaction that requires little hands on time as kits, like
GenomiPhi, have combined all necessary reaction components into 3 tubes. Although the
setup process has been simplified, the 18 hour incubation was troublesome as a sample

essentially had to be extracted and set up for WGA a full day in advance of any PCR

42

amplification. A second concern is the need to reopen sample tubes after denaturing the
DNA at 95°C in order to add the Phi 29 enzyme; this could result in contamination.

Exogenous DNA is a particular concern with WGA. Any source of
contaminating DNA, whether it is bacterial, fungal, or fi'om a crime scene or laboratory
technician, could be preferentially amplified, especially since the source is likely to be
higher quality than that of the forensic material.

Another problem encountered with any WGA method that relies on random
primers annealing throughout the genome is the shortening of product. For illustrative
purposes, consider a sample of degraded DNA that is 800bp in length. This DNA has a
maximum amplicon length of 800bp, assuming the random primers bind at the very ends
of the template. In the vast majority of instances the random primers will anneal at other
locations on the 800 base fragment, and a shortened product will be produced. In
subsequent rounds, product will continue to be shortened. Similarly, consider an 800bp
piece of DNA which contains primer binding sites for a 400bp PCR product located
directly in the middle of the template, and 200bp of flanking sequence on either side.
Only if a WGA primer anneals in the 200bp 5’ of the PCR primer site will both PCR
primer locations be replicated. If the WGA primer anneals within the 400bp target region,
the 5’ PCR primer site will be lost, and the WGA product will be useless. If the WGA
primer anneals in the 3’ flanking region, neither PCR primer site will be replicated. Thus
in WGA, 3 out of 4 amplifications of this 800bp DNA will be useless, and as the
shortening described above occurs, this ratio will only worsen. Again, as the WGA
process continues and more product is amplified, trimming will worsen from one round

of amplification to the next as the primers bind short of the strand ends.

43

General WGA Results

When WGA product was separated on an agarose gel, a vertical smear of DNA
extended the length of the gel and occurred with or without input (reagent blank) DNA
(Figure 2). The intensity of both positive and reagent blank smears from MDA product
was greater than that from I-PEP. Further there was apparent high molecular weight DNA
in the I-PEP and MDA reactions, even from reagent blanks. One possible cause of the
smear, especially from samples such as aged bone which was buried in soil, was the
amplification of microbial DNA. However this is not the only potential cause since the
smear was observed in reagent blanks and positive reactions of all sample types, thus the
smear seems to be a result of the WGA procedure in general, while the difference in
smear intensity is due to a difference in the way each method amplifies product. In an I-
PEP reaction, the random primers could overlap one another and act as template for the
polymerases, but this product must be beat denatured in order for template to become
accessible to new primers. This need to denature limits the overall yield increase since
additional priming cannot occur until the next cycle. On the other hand, MDA utilizes the
Phi 29 polymerase, which is capable of displacing double stranded DNA. As
amplification occurs, primers anneal to template and the enzyme extends. If the template
is high molecular weight DNA, the highly processive nature of the enzyme will continue
to amplify sample until the DNA end is reached. In a reaction which has no input DNA,
the primers (and any extraneous DNA) could overlap and act as template for the enzyme,
progressively synthesizing longer product. As the enzyme extends the template strand,
its displacement properties release a new strand of DNA, and amplification of this new

strand can either be primed with random hexamers or the single stranded template could

44

potentially fold over and form a hairpin to act as its own primer. In either situation, any
DNA added to the reaction will be continuously amplified by the Phi 29 enzyme.

PCR amplification of MDA treated DNA from artificially degraded sample, aged
blood, hair and aged bone resulted in multiple non-target DNAs (Figure 3 and 6). The
lengths of the extra products were different among samples, but if the same sample was
amplified in duplicate, identical sized bands were generated. When product from an
MDA reagent blank was amplified, extra bands were not produced. A simple explanation
for these extra bands is that residual random primers were not washed completely from
the MDA product which then acted as primers in a PCR reaction. This explanation is
unlikely since the residual primers would most likely produce random products,
visualized as more smearing, and not multiple bands. A second explanation is that there
is random amplification of small pieces of DNA that contain the target PCR binding site,
and these get incorporated into the larger pieces of DNA that result in the smears. This
would account for the lack of bands in negative controls, even though these too contain
smears. A final explanation involves the production of MDA artifacts related to the Phi
29 amplification process. As described above, MDA may amplify and lengthen any input
DNA. If the single stranded WGA DNA were to form a hairpin loop and then extended
on itself, product length could be doubled (or even tripled with a second hairpin). The
resulting product would consist of tandem forward and reverse complement sequences of
the original template, and could potentially contain PCR primer binding sites. Since the
lengths of the non-target bands differed among samples, but were the same in serial

dilutions of each WGA product, the extra PCR bands are not locus-specific (e.g., a region

45

of mtDNA does not generate a specific artifact). Regardless of the mechanism however,
production of non-target bands will have a negative impact on subsequent analyses.

Dean et a1. (2002) briefly mentioned the production of artifacts, or extra non-
target DNA, from PCR based WGA methods, such as I-PEP and DOP-PCR. They
suggested that the production of artifacts was due to the repeated heating of template that
degraded the sample. However, they did not include any results displaying the production
of non-target amplifications, did not note that the PCR process itself repeatedly heats
DNA, and failed to mention similar issues with their MDA method. They did reference a
paper by Cheung and Nelson, (1996) who evaluated DOP-PCR and discussed the
production of non-genomic product, possibly due to non-specific priming. These authors
reported that the production of non-specific DNAs influenced the success of PCR
amplification in up to two-thirds of attempts. It is possible that artifacts associated with
the MDA process may not have been observed by Dean et al. (2002) since they were not
investigating the utility of WGA on degraded DNA.

Yet another problem encountered with I-PEP and MDA product was that PCR
amplification of some whole genome amplified samples was inhibited at a dilution that
was successful for untreated DNA (Table 9, Figure 7). This inhibition was generally
eliminated when the WGA sample was diluted 10 to 1000 fold prior to PCR. One
explanation for this PCR inhibition is that input DNA was too concentrated following
WGA. It is possible that the amount of some components, such as primers and dNTPs,
became limiting before the PCR process was capable of generating the desired amplicon.
The amplification of hair shaft samples was a good example of this inhibition. As seen in

Figure 7, hair samples diluted 10 to 100 fold (number 27 and 28) were not amplified, but

46

with a 1000 fold dilution, amplification was successful. A second explanation for this
inhibition is that WGA artifacts were somehow acting as inhibitory agents, and as the

sample is diluted so are these artifacts.

Product Quality of Untreated, I-PEP, and WA Samples

Quality assessments of untreated and whole genome amplified DNA were
completed by determining maximum PCR product lengths (nuclear or mtDNA) from
artificially degraded DNA, fiesh blood, aged blood, hair shafts and aged bone. These
results were compared to one another to determine the usefulness of WGA on samples
which have varying levels of degradation.

The effects of WGA on degraded DNA were investigated using DNA of a
controlled length (Table 4). There is an apparent relationship between amplification
success of I-PEP treated DNA and size class, where a 213bp nuclear product was always
amplifiable from the fractions greater than 400 nucleotides, while the smaller <200 and
200 — 400 size classes had decreased PCR success, amplifying 3 and 4 times out of 5,
respectively (Summary Data fi'om Table 4). The relationship between size class and
amplification success of I-PEP product is reinforced by mtDNA amplification results
where larger size classes were again more successful than the small size classes. In
contrast there is no apparent relationship between MDA treated size classes and
amplification success of a 213bp nuclear product; the <200, 600 — 800, 800 — 1000, and
>1000 MDA nucleotide fi'actions successfully amplified in 3 out of 5 attempts, while the
200 — 400 and 400 — 600 size classes successfully amplified 4 out of 5 times. A 270bp
mtDNA amplicon amplified in 2 out of 4 attempts for all MDA size classes. The

difference in amplification trends between I-PEP and MDA could be due to WGA

47

shortening. If this was the case, shortening of I-PEP product was only noticeable in the
smaller size classes, while it appears that all MDA treated size classes were affected
equally.

When comparing PCR amplification results of untreated, I-PEP and MDA treated
samples from the 5 trials of degradation, 3 of the 6 I-PEP size classes had greater nuclear
amplification success than untreated and 4 of the 6 had greater success than MDA treated
DNA. Five out of 6 of the I-PEP size classes had a higher mtDNA amplification success
than untreated size classes, and all 6 I-PEP size classes had a higher success than MDA
samples. Results from untreated and MDA product indicate that PCR amplification fiom
untreated DNA is more successful than MDA, where 4 of 6 untreated size classes had
greater amplification success, 2 size classes had equal amplifications between the two
treatments, and 1 untreated size class was less successful than MDA product. Clearly, I-
PEP treated DNA was amplified more fiequently than other treatments, across
degradation trials and size classes. This is consistent with the findings noted above in
which I-PEP shortening affected the larger size classes less than shorter size classes,
which in turn would result in greater PCR amplification results, while shortening of
MDA product affected all size classes leading to a reduction in amplification success.
Further, untreated DNA was amplified more often than MDA product. The reduction of
amplification success after WA is a concern for forensic biologists and could prevent
the method from being implemented in casework unless a satisfactory explanation and
remedy could be obtained.

There were discrepancies in amplified PCR product lengths across trials of

artificial degradation; longer product was amplified from trials 1 and 2 (496bp nuclear,

48

864bp mtDNA), than fiom trials 3 — 5 (213bp nuclear, 220 to 270bp mtDNA). This
difference in product length is likely due to a change in the separation protocol, which in
later trials was accomplished through the use of a longer gel. The modified procedure
clearly improved separation of the digested DNA, since the amplification of an 865bp
product should not be possible from a <200 nucleotide size class. Moreover, the data
from the last 3 trials were more consistent with each other, indicating these data are more
reliable and the modified technique was more successful at separating the DNA pieces.

Maximum amplicon length obtained after I-PEP and MDA fi'om fresh blood,
(496bp nuclear, 1064bp mtDNA) did not differ from untreated DNA (Table 5). There
was also no difference in the maximum nuclear DNA product (496bp) between untreated
and whole genome amplified DNA from aged blood. However, untreated aged blood had
a reduced maximum mtDNA amplicon length of 865bp, while a 1064bp product was
obtained fiom I-PEP and MDA treated DNA. Presumably the untreated DNA sample
contained mtDNA of at least 1064bp, but the quantity was not great enough for
successfirl PCR amplification. After WGA, the amount of longer DNA strands was
increased beyond the minimum threshold required for PCR. Had the sample simply been
PCR amplified, the evidentiary value of the DNA would have been diminished.
Ironically, when tests were attempted to verify the difference in product lengths, not
enough untreated DNA remained, illustrating the usefulness of WGA.

In contrast, the length of amplifiable PCR product from hair and skeletal material
was clearly reduced after I-PEP. An 865bp mtDNA product was amplified from
untreated DNA from hair, but after I-PEP, the maximum product length of 2 samples was

reduced to 664bp and three more had a maximum amplicon length of 203bp. Without

49

WGA, a 220bp mtDNA product was amplified from the 5 Voegtly bone samples. Only
one bone sample could be amplified after I-PEP; no product was generated from the 4
others. None of Albanian samples could be amplified following I-PEP, even though a
266bp mtDNA product was generated from them prior to WGA. The MDA method was
even less successful on degraded DNA. All nuclear and mitochondrial amplification
attempts from MDA treated aged bone and hair failed. In general, by comparing the PCR
amplification results of all sample types, the hypothesis that the WGA process shortens
product, leading to the loss of primer binding sites, is exemplified since maximum PCR
product length fi'om high quality DNA did not differ, but when testing degraded samples,

there was a clear reduction in maximum amplicon length after WGA.

Product Yield fi'om I-PEP and MDA

WGA product yield was compared to the quantity of untreated DNA by PCR
amplifying nuclear (blood), or mtDNA (hair) product with serial diluted input DNA. In
all cases where PCR amplification was successfirl, the yield of a 496bp nuclear or 203 bp
mtDNA product was increased by WGA. I-PEP product yield was increased 1000 fold
for fi'esh blood, 100 fold for aged blood and 20 to 2000 fold for DNA from hair shafts, in
comparison to untreated DNA. MDA increased product yield from fresh blood 10,000
fold and aged blood 1000 fold (Table 8 and Figure 5).

The difference in product yield between flesh and aged blood is again consistent
with the idea that whole genome amplified product presumably will be subjected to
shortening and PCR primer binding site loss is more likely with short template.
Additionally, the inconsistent yield increases among hair samples may be due to the

quality of the mtDNA available in each sample.

50

Downstream Analysis of Whole Genome Amplified Product

STR analysis of untreated and whole genome amplified samples was performed
on artificially degraded DNA from trial 5 (Table 10). The untreated 400 — 600, 600 —
800, 800 - 1000 size classes matched the reference sequence at 14, 16, and 15 loci,
respectively. Of the three loci which were not typed, 2 have longer PCR products, and
one locus could not be typed because 3 peaks were present. Amplification of the >1000
size class for STR analysis was unsuccessful. It is unclear why this size class did not
amplify, particularly since a 213bp nuclear product was amplified fi'om this size class in
earlier experiments, but it may simply be due to PCR amplification failure unrelated to
WGA.

STR amplification of the artificially degraded >1000 size class after I-PEP was
less successful than untreated DNA, where 4 I-PEP loci completely matches and 7 loci
were partial matches. STR amplification from the MDA treated artificially degraded
>1000 fraction was also reduced in comparison to untreated sample, with complete
matches at 9 loci, and partial matches at 2 loci. The unsuccessful alleles fi'om both WGA
procedures were from the longest loci in the multiplex. Therefore, it is likely that WGA
product shortening is again responsible for the loss of primer binding sites from longer
loci. The overall superior STR amplification results of untreated artificially degraded
DNA over I-PEP treated DNA was inconsistent with results obtained from single locus
testing, where I-PEP product was amplified more fi'equently. In earlier tests, a 213bp
nuclear product was attempted; STR loci lengths are between 100 and 360bp. A
reduction in the amplification success of I-PEP product may have been observed if earlier

tests had been performed using primers designed to amplify a 300bp product, and it is

51

possible the 213bp amplicon represents the upper limit of what can be amplified from the
I-PEP product.

Interestingly, STR amplification of the MDA treated fraction was more successful
than I-PEP, even though amplification of the I-PEP fractions was more successful in
single locus PCR reactions among all trial of degradation. This seeming inconsistency
may be explained by a difference in the amount of available template. MDA is expected
to increase product yield more than I-PEP, and since the same amount of sample was
added to the STR amplification reaction, it is likely that MDA product had a greater
amount of DNA available.

STR analysis was performed on untreated and whole genome amplified DNA
from the luL sample of fresh blood (Table 11). STR amplification of untreated DNA
(Figure 8) was successful at 5 loci, and incomplete matches, due to allele drop-out, were
obtained at 9 loci. The locus length was not related to amplification success; 2 incomplete
matches occurred at short length loci, and 5 incomplete matches occurred at medium
length loci, while 1 long locus completely matched. These results were unexpected since
the single locus amplification of a 496bp nuclear product was successful and the longest
locus length in the multiplex is 360bp. In addition, equivalent amounts of DNA and
similar PCR cycle numbers were used (38 for single locus, 35 for STR amplification).
Therefore, the loss of heterozygosity for untreated samples may be explainable by an
inadequate quantity of starting DNA.

STR amplification of the I-PEP treated fi'esh blood sample was successfill at all
16 loci, while amplification of the MDA sample from fresh blood was successful at 14

loci, and incomplete matches were obtained from 2. There was no relationship between

52

locus length and amplification failure, however, more input MDA sample would likely
result in a more complete profile since two peaks were observed at the TPOX locus, but
one was below the minimum intensity threshold. Also, STR loci were amplified fi‘om
MDA product when diluted 100 fold more than I-PEP or untreated DNA. Overall, STR
analysis was more successful from whole genome amplified sample. Prior to WGA there
was not enough DNA in the sample to amplify all loci; upon completion of I-PEP and
MDA, more loci were amplified. These results clearly illustrate the usefulness of WGA
on high quality samples that are limited in quantity.

STR testing of untreated and whole genome amplified DNA from aged blood was
largely unsuccessful. No profile was generated from untreated DNA, even though
additional amplifications were attempted with an increased amount of input DNA.
Neither WGA method generated a complete profile, but after I-PEP 7 alleles, and after
MDA 1 allele could be matched to the known profile. Alleles which matched the known
profile were all from loci of short product length. In comparison to the single nuclear
locus testing, where no difference in maximum product length was observed, it is
surprising that there was a difference between the number of amplified STR alleles fiom
untreated DNA and WGA product, especially since earlier PCR amplifications of a
496bp nuclear product was successful, (thus the largest STR locus, at 360bp, should be
amplifiable). However, this discrepancy is consistent with the maximum length mtDNA
results fi'om the same samples, where the difference in PCR product length between
untreated and treated samples was likely due to an increase in available template after

WGA. In short, the ability to amplify STR alleles from untreated DNA was limited by

53

starting quantity, after WGA the increase in DNA quantity made STR amplification
possible.

MtDNA sequencing for identification of forensic samples was also completed
with WGA product. Amplification of mtDNA from five hair and eight bone samples
without WGA treatment was successfirl. After WGA, amplification was possible from 4
of 5 hair samples and 1 bone sample. As mentioned above, differences in the
amplification success of these samples before and after WGA, is likely due to the loss of
primer binding sites as a result of product shortening. Sequencing of these amplicons was
successfill and results were consistent with known profiles, with the caveat that more
heteroplasmy was observed in the WGA sequences (Table 12). It is problematic that
sequence differences from two whole genome amplified hair samples resulted in dual
peaks. However, as seen in sample 29, an ambiguity in untreated DNA was resolved after
WGA; more experiments will be needed to fully understand the cause of these sequence
differences.

Conclusions

Both methods of WGA were easy to perform, but I-PEP was preferred since non-
target DNAs are not generated following PCR and valuable DNA was not exposed to
potential contamination by reopening the sample tube. Further, the entire I-PEP
procedure is less time consuming and thus advantageous in a field where casework is
already backlogged. I-PEP and MDA are valid methods of increasing product yield for
forensic samples which are brought into a crime lab in very limited quantity but not

quality, such as small amounts of relatively fresh blood. Overall, MDA increased

54

product yield 1000 to 10,000 times, and I-PEP increased product yield between 20 and
2000 times.

Based on the variety of experiments described here, including those on controlled
(artificially manipulated) DNAs, as well as forensic samples, it is unlikely that highly
degraded DNA will benefit from WGA. Results of PCR amplification from aged bone
and hair shaft samples were not improved after WGA; in fact, most samples had better
amplification success without WGA than with. In addition, multiple non-target products
were PCR amplified from highly degraded DNA which underwent MDA. Furthermore,
STR analysis of untreated artificially degraded DNA generated more complete profiles
than those amplified after I-PEP or MDA. The reduced amplification success is most
likely due to a shortening of the whole genome amplified product, which resulted in the
loss of primer binding sites. Forensic identification information could be lost if WGA
were attempted on highly degraded DNA.

In conclusion, MDA was capable of increasing yield more proficiently, but PCR
amplification of I-PEP product was more successful. While the principle objective of
WGA is to increase product yield, the reliable amplification of forensic samples in
downstream reactions is more important than large product increases. It is this regularity

which makes the I-PEP method a better choice for use with forensic samples.

55

REFERENCES

56

REFERENCES

Amersham Biosciences. GenomiPhi DNA Amplification Kit: More from Less. [Brochure].
2003-2004

Anderson S, Bankier AT, Barrel] BG, de Bruijn MH, Coulson AR, Drouin J, Eperon IC,
Nierlich DP, Roe BA, Sanger F, Schreier PH, Smith A], Staden R, Young IG.
Sequence and organization of the human mitochondrial genome. Nature. 1981
Apr 9;290:457-465.

Barker DL, Hansen MST, F aruqi AF, Giannola D, Irsula OR, Lasken RS, Latterich M,
Makarov V, Oliphant A, Pinter JH, Shen R, Sleptsova I, Ziehler W, Lai E. Two
Methods of Whole-Genome Amplification Enable Accurate Genotyping Across a
2320-SP Linkage Panel. Genome Res. 2004 May;14(5):901-907.

Blanco L, Bemad A, Lazaro JM, Marin G, Gannendia C, Salas M. Highly efficient DNA
synthesis by the phage phi 29 DNA polymerase. Symmetrical mode of DNA
replication. J Biol Chem. 1989 May 25;264(15):8935-8940.

Cheung VG, Nelson SF. Whole genome amplification using a degenerate Oligonucleotide
primer allows hundreds of genotypes to be performed on less than one nanogram
of genomic DNA. Proc Natl Acad Sci U S A. 1996 Dec 10;93(25): 14676-14679.

Dean FB, Hosono S, Fang L, Wu X, Faruqi AF, Bray-Ward P, Sun Z, Zong Q, Du Y, Du
J, Driscoll M, Song W, Kingsmore SF, Egholm M, Lasken RS. Comprehensive
human genome amplification using multiple displacement amplification. Proc
Natl Acad Sci U S A. 2002 Apr l6;99(8):5261-5266.

Dietrnaier W, Hartrnann A, Wallinger S, Heinmoller E, Kerner T, Endl E, J auch KW,
Hofstadter F, Ruschoff J. Multiple mutation analyses in single tumor cells with
improved whole genome amplification. Am J Pathol. 1999 Jan;154(l):83-95.

Esteban JA, Salas M, Blanco L. Fidelity of phi 29 DNA polymerase. Comparison
between protein-primed initiation and DNA polymerization. J Biol Chem. 1993
Feb 5;268(4):2719-2726.

Garmendia C, Bemad A, Esteban JA, Blanco L, Salas M. Garmendia C, Bemad A,
Esteban JA, Blanco L, Salas M. The bacteriophage phi 29 DNA polymerase, a
proofreading enzyme. J Biol Chem. 1992 Feb 5;267(4):2594-2599.

Graffy, E A. Development and Validation of an Alkaline Extraction Method for Isolating

Mitochondrial DNA from Human Hair Shafts. East Lansing, MI. Michigan State
University. 2003 (MS-thesis.)

57

Grant SF, Steinlicht S, Nentwich U, Kern R, Burwinkel B, Tolle R. SNP genotyping on a
genome-wide amplified DOP-PCR template. Nucleic Acids Res. 2002 Nov
15;30(22):e125.

Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis
program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41 :95-98.

Hosono S, Faruqi AF, Dean FB, Du Y, Sun Z, Wu X, Du J, Kingsmore SF, Egholm M,
Lasken RS. Unbiased whole-genome amplification directly from clinical samples.
Genome Res. 2003 May;13(5):954-964. Epub 2003 Apr 14.

Kuivaniemi H, Yoon S, Shibamura H, Skunca M, Vongpunsawad S, Tromp G. Primer-
extension preamplified DNA is a reliable template for genotyping. Clin Chem.
2002 Sep;48(9): 1601-1604.

Langmore JP. Rubicon Genomics, Inc. Pharmacogenomics. 2002 Jul;3(4):557-60.

Lasken RS, Egholm M. Whole genome amplification: abundant supplies of DNA from
precious samples or clinical specimens. Trends Biotechnol. 2003 Dec;21(12):531-
525.

Mi sner L. Predicting mtDNA Quality Based on Bone Weathering and Type. East Lansing,
MI. Michigan State University. 2004 (MS-thesis.)

Nakahori Y, Takenaka O, Nakagome Y. A human X-Y homologous region encodes
"amelogenin". Genomics. 1991 Feb;9(2):264-9.

Paez JG, Lin M, Beroukhim R, Lee JC, Zhao X, Richter DJ, Gabriel S, Herman P, Sasaki
H, Altshuler D, Li C, Meyerson M, Sellers WR. Genome coverage and sequence
fidelity of ph129 polymerase-based multiple strand displacement whole genome
amplification. Nucleic Acids Res. 2004 May 18;32(9):e71.

Roche Applied Science. Elliminate Errors. <http://www.roche-applied-
sciencecom/fstlamplification.htm?/sis/amplification/pifs/enzymes/pwo.htm>
(2003). Accessed March 14, 2005.

Roche Applied Sciences. Important Note: Expand High Fidelity PCR System. [Product
Update]. 2003.

Rozen S, Skaletsky H. Primer3 on the W for general users and for biologist
programmers. In: Krawetz S, Misener S (eds) Bioinformatics Methods and
Protocols: Methods in Molecular Biology. (2000) Humana Press, Totowa, NJ, pp
365-3 86

Telenius H, Carter NP, Bebb CE, Nordenskjold M, Ponder BA, Tunnacliffe A.

Degenerate Oligonucleotide-primed PCR: general amplification of target DNA by
a single degenerate primer. Genomics. 1992 J ul;13(3):718-25.

58

Tzvetkov MV, Becker C, Kulle B, Numberg P, Brockmoller J, Wojnowski L. Genome-
wide single-nucleotide polymorphism arrays demonstrate high fidelity of multiple

displacement-based whole-genome amplification. Electrophoresis. 2005
Feb;26(3):710-715.

Ubelaker, D.H., E.B.. Jones, Editors. Diane Beynon Landers, Associate Editor for
Archaeology. Human Remains fi'om Voegtly Cemetery, Pittsburgh, Pennsylvania.
Smithsonian Contributions to Anthropology Series, Number 46, Smithsonian
Institution Press, Washington DC. 2003

Zhang L, Cui X, Schmitt K, Hubert R, Navidi W, Amheim N. Whole genome

amplification from a single cell: implications for genetic analysis. Proc Natl Acad
Sci U S A. 1992 Jul l;89(13):5847-5851.

59

   

I"Elllljjijlfljljlfll14ml