Page 1 of 63

OPTIMISING SILICA-BASED SOLID PHASE DNA EXTRACTION

METHODS FOR LOW CONCENTRATION FORENSIC SAMPLES

By

Katherine Dilley

A thesis submitted in fulfilment of the requirements for the degree of

Master of Forensic Science (Professional Practice)

in

The School of Veterinary and Life Sciences

Murdoch University

Supervisors:

Brendan Chapman (Murdoch University)

Felicity Pagan (Australian Federal Police)

Sam Cornwell (PathWest Laboratory Medicine WA)

Marie Rye (PathWest Laboratory Medicine WA)

Semester 1, 2019

Page 2 of 63

Declaration

I declare that this thesis does not contain any material submitted previously for the award

of any other degree or diploma at any university or other tertiary institution.

Furthermore, to the best of my knowledge, it does not contain any material previously

published or written by another individual, except where due reference has been made in

the text. Finally, I declare that all reported experimentations performed in this research

were carried out by myself, except that any contribution by others, with whom I have

worked is explicitly acknowledged.

Signed: Date: 1/7/2019

Page 3 of 63

Acknowledgements

I would like to whole-heartedly thank my supervisor Brendan Chapman for his continuous

support and guidance throughout my entire Master’s degree. I greatly appreciate your

patience and knowledge, which have been invaluable during this research project.

To my family, friends and fellow Masters students—thank you for enduring my highs and

lows, providing endless encouragement and rendering assistance at all hours.

Page 4 of 63

Table of Contents

Title Page .............................................................................................................................................................. i

Declaration ......................................................................................................................................................... ii

Acknowledgements ....................................................................................................................................... iii

Part One

Literature Review ............................................................................................................... 6-39

Part Two

Manuscript ......................................................................................................................... 41-63

Page 5 of 63

Blank Page – not numbered

Page 6 of 63

Part One

Literature Review

OPTIMISING DNA EXTRACTION METHODS FOR LOW CONCENTRATION FORENSIC SAMPLES: A REVIEW

Page 7 of 63

ABSTRACT

Trace DNA analysis from minute biological samples has become commonplace in modern

forensic laboratories due to increased sensitivities in genotyping systems and improved

extraction chemistries. However, analysis still remains a challenge as no single protocol

exists that will isolate DNA in both sufficient quantity and quality for downstream

applications. Extraction is the most crucial step for maximising recovery of DNA, and

thorough optimisation of procedures is needed to ensure informative genetic profiles can

be generated. This review will investigate the efficiency of different methods available for

isolating trace quantities of DNA from forensic samples, discussing their advantages and

limitations. It will explore improvements to the extraction methodology, including

optimisation of the elution volume and methods of post-extraction purification. Lastly,

centrifugal filters will be debated for their concentrating properties and ability to improve

the recovery of trace DNA.

Page 8 of 63

TABLE OF CONTENTS

ABSTRACT .................................................................................................................................. 7

LIST OF FIGURES......................................................................................................................... 9

LIST OF TABLES......................................................................................................................... 10

LIST OF ABBREVIATIONS .......................................................................................................... 11

1. INTRODUCTION ................................................................................................................ 12

2. DNA EXTRACTION METHODS ........................................................................................... 14

2.1 Organic extraction..................................................................................................... 14

2.2 Chelex ........................................................................................................................ 15

2.3 Solid-phase DNA extraction ...................................................................................... 16

3. IMPROVING RECOVERY OF TRACE DNA ........................................................................... 17

3.1 Introduction to trace forensic samples ..................................................................... 17

3.2 Optimising the efficiency of extraction .................................................................... 18

4. METHODS OF CONCENTRATION AND PURIFICATION ...................................................... 22

4.1 Ethanol precipitation ................................................................................................ 22

4.2 Spin columns ............................................................................................................. 23

4.3 Ultrafiltration ............................................................................................................ 24

5. CENTRIFUGAL FILTERS IN FORENSICS .............................................................................. 25

5.1 Improving the profiling success ................................................................................ 26

5.2 Degraded samples..................................................................................................... 27

5.3 Trace samples ........................................................................................................... 27

6. OPTIMISING DNA RECOVERY WITH CENTRIFUGAL FILTERS............................................. 28

6.1 Membrane design ..................................................................................................... 29

6.2 DNA loss .................................................................................................................... 30

7. EXPERIMENTAL AIMS AND OBJECTIVES ........................................................................... 32

8. CONCLUSION .................................................................................................................... 33

9. REFERENCES ..................................................................................................................... 35

Page 9 of 63

LIST OF FIGURES

Figure 1 – Illustration comparing the three main methods of genomic DNA extraction….20

Figure 2 – Typical rates of DNA recovery quoted by Millipore—manufacturers of Amicon®

Ultra-0.5 filter devices..…………………………………………………………………………………………………31

Page 10 of 63

LIST OF TABLES

Table 1 – A comparison of commercially available centrifugal filters commonly used for

concentration of DNA samples in forensic casework………………………………………………………29

Page 11 of 63

LIST OF ABBREVIATIONS

DNA – Deoxyribonucleic acid

STR – Short tandem repeat

PCR – Polymerase chain reaction

pg – Picogram

µL – Microlitre

QIAamp – QIAamp® DNA Investigator kit (Qiagen)

UF - Ultrafiltration

eDNA – Environmental DNA

MW – Molecular weight

NMWL – Nominal molecular weight limit

K – Kilodalton

Page 12 of 63

1. INTRODUCTION

Biological samples can provide valuable evidence in a forensic investigation,

potentially linking persons of interest to a crime. Genetic identification is achieved

through the analysis of deoxyribonucleic acid (DNA) in which scientists target highly

variable, short repeating sequences of the genetic code in order to create a unique profile

[1]. The success of short tandem repeat (STR) typing is heavily dependent on the size,

quality and purity of DNA obtained from samples [2]. A viable genetic profile can only be

generated if an extract is sufficiently concentrated in genomic DNA and free of

polymerase chain reaction (PCR) inhibitors [3]. This can be a challenge for trace forensic

samples which only contain minute quantities of DNA. PCR inhibition can be reduced

during the extraction phase, but the loss of DNA is rarely preventable. When working with

trace amounts, this may be extremely limiting, resulting in poor or non-existent profiles.

DNA analysis is a multi-stage process encompassing the following basic steps;

sampling, extraction, quantitation, amplification, and STR profiling [1]. Obtaining the

maximum amount of DNA possible throughout each stage is crucial as it determines the

outcome of downstream applications and ultimately, the generation of a successful

profile [4]. Forensic DNA analysis is founded upon standardised technical procedures and

commercially available kits and reagents with a database of protocols for various

biological sources and substrates. As technology advances, there is a continuous demand

for the development of higher throughput laboratory instrumentation, the automation

and standardisation of steps, and new methodologies exploiting increased sensitivity [5].

DNA extraction is a heavily researched area as it is the most critical step in the analysis

workflow, directly affecting the amount of starting material available for further

examination. There are a range of methods available for different types of samples, with

Page 13 of 63

varying efficiencies depending on the chemistry they employ. Unfortunately, there is no

one-size-fits-all protocol available, and each method has its advantages and limitations.

Debate has always ensued about which procedure is the best, particularly for challenging

samples compromised in quality or quantity. Ideally, an extraction procedure should be

capable of rapidly isolating high yields of pure DNA from a range of samples [6]. Whilst

this might be achievable from samples with a high DNA concentration, in real-life

casework, many secured DNA traces contain no or too little DNA for analysis. This

consumes time, money and resources, contributing to backlogs in forensic laboratories

and leads to increased turnaround times and reduced solvability of cases [7].

Improvements to standard extraction methods are therefore needed in order to

maximise the quality and quantity of DNA isolated and generate informative STR profiles

[4]. Validated protocols already exist for low-level evidentiary samples, including reduced

volume PCR and post-PCR purification [8]. These have had limited success with improving

the DNA profile, often leading to a variety of artefacts and false alleles, causing

misinterpretation [9]. A promising method that could provide a solution is post-

extraction membrane filtration with a centrifugal concentrator. This technique has been

successfully used in DNA purification for decades, and has since been gaining momentum

for its concentrating properties, following the benefits observed in other scientific fields,

such as microbiology and environmental barcoding [10]. To date, there is only a small

amount of research exploring the use of centrifugal filters in forensic DNA analysis. The

majority of it has focused on investigating how factors such as membrane design and

centrifugal force can attribute to further DNA loss. Thus far, no study has explored the

performance of these devices across multiple manufacturers and evaluated their absolute

efficiency to improve the recovery of low concentrations of DNA.

Page 14 of 63

2. DNA EXTRACTION METHODS

The extraction of DNA from forensic samples is one of the most fundamentally

important steps in the analysis workflow [4]. Biological samples contain a mix of cellular

proteins and other substances along with the DNA molecules. These must be removed as

they may reduce or prevent PCR amplification [1]. There are various methods that exist,

depending on the type of sample being extracted and required downstream applications.

All protocols follow the same basic steps; (1) lysis of the cell to release the DNA, (2)

isolation of the DNA from other cellular material, and (3) collection and concentration of

the DNA into an appropriate format [1]. In a perfect world, the extraction process should

recover high amounts of pure, stable, high-quality DNA, and be completely free of

inhibitors [11]. Unfortunately, the efficiency of extraction and extent of purity achieved

are inconsistent across different chemistries and protocols [4]. Despite the fact that

methods have evolved to suit a wide variety of biological sources, there is no “universal”

protocol that can be successfully applied to every sample [11]. Factors such as substrate

deposition, environmental exposure, degradation, and contamination contribute to

challenges in the isolation of DNA from forensic evidence. Furthermore, trace samples

that contain limited quantities of DNA add an extra level of complication in the wasteful

analysis approach [4].

2.1 Organic extraction

Organic extraction, also known as phenol-chloroform extraction, has been used in

forensic analysis for the longest time and has, therefore, gained the title of ‘gold

standard’ when it comes to DNA extraction [4]. It is the most tested and proven method

in the forensic science community with a database of protocols for various substrates and

biological samples [1]. The process involves adding chemicals to break open the cell and

Page 15 of 63

digest inhibitory proteins, followed by a phenol/chloroform alcohol mixture to isolate the

DNA based on its solubility in the aqueous portion of the organic mixture [1]. Organic

extraction is extremely reliable and applicable to a wide range of samples, including blood

and saliva [12]. To this day, it remains a preferred method for processing challenging

samples as it provides high yields of pure DNA [4]. It is a popular choice for degraded

samples and has successfully been used with bones, teeth, hair shafts, dandruff,

fingerprints, putrefied tissues, urine, cigarette butts and burnt remains [13-18]. In a

comparison study, Iyavoo [13] found that an organic extraction method consistently

performed better with preserved bones, whilst Castella et al. [11] observed it was

superior for the isolation of touch DNA from clothes. A significant downside to this

technique is the inability to completely remove all inhibitors which interfere with analysis

[11, 13, 14]. In trace DNA analysis, this may limit the amount of recoverable DNA [1].

Popularity for using this method has also reduced due to the harsh chemicals involved,

the tediousness of the procedure and limited potential for automation [14, 16].

2.2 Chelex

A safer alternative to organic extraction is a chelating-resin suspension that can be

added directly to a sample. The resin acts by binding polyvalent metal ions such as

magnesium, thus preventing DNA degradation [1]. Heat is applied to lyse the cells, and

the cellular debris is then removed via centrifugation [12]. The chelex®100 method is

simple, inexpensive and more rapid than the phenol/chloroform method, yielding

relatively high amounts of DNA. Moreover, it is a one-tube procedure involving fewer

steps, therefore reducing the opportunity for laboratory-induced contamination [1, 4].

Although labelled a crude method, it has been applied successfully to samples such as

blood stains, tissue, hair and bone [4]. A key disadvantage of this method, like it’s organic

Page 16 of 63

counter-part is that it has a limited capacity to remove PCR inhibitors [12]. Phillips et al.

[19] failed to obtain STR profiles from bloodstains extracted with chelex due to haem and

remaining resin impurities. In a similar study, Ip et al. [20] found that chelex yielded less

DNA than other extraction methods due to co-extracted inhibitors and degradation

caused by high temperatures.

2.3 Solid-phase DNA extraction

DNA extraction employing solid phase chemistries has become the method of

choice for modern forensic laboratories, exploiting the unique properties of the DNA

molecule. Described originally in 1990, the technique is based on the affinity of DNA to

selectively bind to a silica substrate in the presence of chaotropic salts [16]. The high

concentration of these chaotropic agents disrupts the structure of the DNA, allowing a

salt bridge to form and accelerating the binding between the negatively charged DNA

molecules and the silica particles [21]. Unwanted proteins and cellular debris are washed

away, and pure, high-quality DNA is eluted from the glass substrate. The main advantage

of this method is that the substrate binds only the DNA, permitting repeated washes and

removal of all other unwanted components. This allows DNA to be isolated with much

higher purity compared to chelex and organic extraction methods [4]. The technology is

also popular due to its high throughput and ability to be automated [1]. A limitation of

silica-based methods is that they have been known to preferentially recover high

molecular weight DNA and therefore fail to extract smaller degraded fragments [4]. This is

likely because of two inherent loss mechanisms; (1) inefficient DNA adsorption onto the

substrate, and (2) ineffective elution of purified DNA from the substrate [22]. The silica

matrices contain a small number of irreversible binding sites that permanently bind

nucleic acids and may contribute to sample loss if carrier RNA is not present [6]. Despite

Page 17 of 63

these shortcomings, solid-phase methods are still deemed the most efficient at isolating

DNA and removing inhibitors.

3. IMPROVING RECOVERY OF TRACE DNA

The main and ultimate aim of the extraction process in forensic investigation is to

obtain the maximum amount of DNA possible [13]. When working with trace samples in

which the starting quantity of DNA is already low, this is paramount to obtaining

informative genetic profiles. Scientists are therefore continually working to improve

extraction methods, increasing their efficiency at recovering DNA whilst limiting the

occurrence of PCR inhibitors.

3.1 Introduction to trace forensic samples

Technological advancements over the years have drastically improved the detection

and sensitivity capabilities of DNA typing processes [4]. Consequently, there is a greater

demand for standardised methods that will generate viable profiles from forensic samples

extremely compromised in both quality and quantity [6]. Trace DNA refers to cellular

material containing minute amounts of DNA, typically less than 200 picograms [9].

Commonly encountered examples of trace DNA include bones and teeth [21], fingernails

[23], single hairs [9], mixed sexual assault specimens [24], latent fingerprints [25, 26] and

handled objects [27, 28]. These kinds of samples present a challenge for forensic DNA

laboratories because they are heterogeneous in nature, often being degraded, containing

inhibitors and little DNA [9].

Laboratory methods used to process trace samples, frequently result in the loss of a

portion of the original DNA and increase the opportunity for exogenous sources to be

introduced [6]. This can make the interpretation of low-level DNA profiles difficult due to

Page 18 of 63

stochastic effects and allele dropout commonly occurring [29]. In addition, many

conventional amplification systems require an optimum input amount to consistently

generate full profiles [6]. Most multiplexes work at their highest efficiency when 1ng of

DNA is analysed, and not more than 28–30 cycles of amplification are carried out [29]. For

this reason, some laboratories have a minimum acceptable threshold which must be

reached for an extract to progress onto downstream applications [7]. Previous studies

have determined that a sample would initially need to contain approximately 250 cells or

1.45ng to retain a sufficient amount of DNA for amplification after extraction [6, 7].

When working with trace DNA, there are several critical areas in which

improvements can be made to the methodology that will increase the success of profiling

[1]. The most common of these occurs in the PCR stage and involves increasing the

number of amplification cycles, essentially creating more copies of template DNA [29].

However, studies have found that this process also amplifies any co-extracted inhibitors,

and in fact, increases the presence of stochastic effects such as allele drop-out [8, 26]. To

overcome this problem, modified interpretation rules have been devised to account for

uncertainties in the DNA profile [29]. Unfortunately, even with these adjustments, results

are unreliable and do not carry the weight in a court of law that conventional methods

do. This has resulted in a loss of confidence in the process, and uncertainty with any

outcome achieved [9]. For that reason, attention is now being turned to the most logical

option for improving DNA analysis of trace samples—increasing the amount of starting

material at the collection and extraction stages [1].

3.2 Optimising the efficiency of extraction

For the successful analysis of trace DNA, crude methods such as chelex and organic

extraction have fallen out of favour, with losses of up to 75% being reported [9]. Forensic

Page 19 of 63

laboratories have instead turned to solid-phase extraction methods as they seem to be

the most efficient at recovering high amounts of pure DNA. In comparison with chelex

and phenol-chloroform techniques, Castella et al. [11] found that silica-based

technologies performed the best for diluted blood and saliva samples, recovering up to 4x

more DNA. 82% of extracted samples produced conclusive profiles with balanced allelic

peaks, indicating this method was most efficient at removing PCR inhibitors and enzymes

that may degrade the DNA. Efficiency is a common benchmark in the scientific literature,

providing a means for studies to simultaneously compare the performance of different

extraction methods. Ultimately though, very little research has actually considered the

initial amount of DNA present in a sample, therefore failing to reflect the absolute

efficiency of the extraction process [30]. Further investigation is needed for an accurate

comparison of which method is in fact superior at recovering DNA.

There are several different DNA extraction methods that utilise solid-phase

technology, leading to uncertainty in the estimation of true efficiency. The silica substrate

responsible for binding comes in many forms, the most common of which are as a

membrane, beads within a spin column, or as magnetic particles seen in figure 1 [16]. The

choice of method depends on factors such as; likely amount of DNA in the samples,

substrate, type of biological material, and presence of potential PCR inhibitors [14]. The

most widely used products are arguably made by leading manufacturer Qiagen. They

have been making robust and sensitive amplification kits that have effectively been used

for DNA isolation purposes for more than 20 years [1, 24]. One such kit is QIAamp® DNA

Investigator (QIAamp), which employs silica solid-phase technology and boasts rapid

purification of high-quality, ready-to-use DNA [31]. It is a common choice by modern

forensic laboratories because it is quick, cost-effective, and a lot of the labour process can

Page 20 of 63

be reduced by automation [19]. The QIAamp kit is well documented as offering good

recovery of DNA from a wide range of samples compared to other extraction procedures

[11, 19, 20]. Ip et al. [20] determined that QIAamp was more successful than chelex in

extracting DNA from serially diluted blood and 76 simulated touch samples. A 2 to 5-fold

increase in DNA concentration was observed, and a higher number of loci were

successfully amplified.

Figure 1. [12] Illustration comparing the three main methods of genomic DNA extraction. Top: Organic extraction with phenol-chloroform alcohol mixture which separates DNA into the aqueous phase. Middle: Chelex extraction in which chelating resins bind degradative compounds and DNA is released through boiling. Bottom: Solid phase extraction in which DNA selectively binds to a solid support such as a silica column or magnetic beads in the presence of high salt solutions.

Page 21 of 63

Similarly, Foran et al. [23] investigated the recovery of trace DNA from the

fingernails of mock assault victims. Real-time PCR results indicated that the kit recovered

a significantly greater amount of the perpetrator’s DNA based on a single column elution

than did organic/phenol. Whilst there is ample literature boasting QIAamp as an efficient

extraction kit for the majority of samples, including touch/handled evidence, there is little

research into its ability to recover trace amounts of DNA from other sources. The QIAamp

kit is advertised by manufacturer Qiagen as being highly suitable for the purification of

such samples, but its manual application into this sort of analysis has not yet been

thoroughly tested.

The elution of DNA from a silica substrate is another relatively under-researched

area of the extraction process which may greatly influence the amount of DNA that can

be successfully recovered. Optimisation of the elution volume is an easily overlooked step

which can assist investigators in maximising the extraction of trace forensic samples.

When dealing with small starting quantities of DNA, eluting in a large volume can be

limiting—essentially diluting the extract and potentially contributing to poor quality

profiles [9]. The QIAamp kit promotes flexible elution volumes of 20–100µL

demonstrating its versatile application in forensic analysis [31]. Unfortunately,

standardised methods are lacking as to the appropriate elution volume for specific

samples, including trace DNA. A few authors have found success in a modified technique

which involves up to 4 successive elutions of smaller volumes [28, 32]. However, the

procedure is labour-intensive and relatively untested in the field of forensics. Foran et al.

[23] studied the QIAamp® DNA Investigator kit for extraction of fingernail scrapings and

determined if the type of eluant (buffer ATE vs low TE) or volume of elutant (20, 28, 50 or

100µL of buffer ATE) increased DNA recovery. Their results showed that DNA could be

Page 22 of 63

successfully recovered in up to 4 elutions; however, the DNA quantity was negligible in

the 4th. Optimisation of the kit’s elution step showed that the volume of buffer used in a

single elution did not have a substantial effect on DNA yields, nor did replacing the kit

elution buffer with low TE. A final protocol of 3 x 20µL elutions with buffer ATE was

determined as the most efficient at recovering DNA [23]. Similarly, Desneux et al. [32]

improved the extraction efficiency of bacterial DNA from piggery effluents by using four

successive 25µL elutions, effectively doubling the DNA yield.

4. METHODS OF CONCENTRATION AND PURIFICATION

DNA extraction is not an infallible process, despite continuous adaptations to

procedures and optimisation for a range of samples. There is an inevitable trade-off

between sufficient purification and the adequate recovery of quality DNA [9]. Chelex and

organic extraction methods are inefficient at removing PCR inhibitors, while silica-based

methods experience DNA loss [6, 22]. Adding a separate clean-up phase post-extraction

is a common technique employed to further improve the quality and quantity of DNA for

downstream processing. Methods range from chemical precipitation to clean-up columns

and centrifugal filters [13]. Ultimately, a purification method should be inexpensive, quick

and be simultaneously applicable to a large number of samples [33].

4.1 Ethanol precipitation

The most widely used technique for post-extraction concentration and purification

is precipitation with absolute ethanol [34]. The “salting out” method as it is more

commonly known, involves the addition of ethanol and salt to an aqueous solution

causing the DNA to precipitate out. The DNA is then recovered via further washing with

ethanol and centrifugation [35]. Having originated from molecular biology, it has become

Page 23 of 63

a cornerstone technique for DNA purification due to its simplicity and reliability [36].

Iyavoo [13] determined that ethanol precipitation gave high yields of DNA from rib bone

samples extracted with phenol-chloroform, but inhibition was detected and characteristic

degradation observed in the larger amplicons of the electropherogram. Other limitations

of this procedure are that it is time-consuming and requires several tedious steps, such as

adjusting for salt concentration [34, 36].

4.2 Spin columns

A quicker and cleaner alternative for nucleic acid purification is the use of spin

columns and centrifugation [37]. Employing the same solid-phase technology that has

been successful for the rapid extraction of DNA from a variety of substrates, spin columns

contain a silica membrane and follow the same underlying protocol of bind-wash-elute

[24]. Used in combination with chelex or organic extraction, spin columns provide a quick

and easy additional purification step removing impurities which may have been missed

during the initial extraction. They are highly advantageous because they can be easily

incorporated into conventional desktop centrifuge equipment, aiding the demand for

increased throughput and automation of DNA analysis methods [33].

In comparison to ethanol precipitation, Greenspoon et al. [24] found that the use of

spin columns improved the quality of DNA from dried bloodstains and sexual assault

samples. DNA was isolated more consistently on a variety of problem surfaces known to

contain PCR inhibitors, and a higher yield was at least equivalent and occasionally greater

than that generated using ethanol precipitation. Many purification spin columns are

commercially available, including QIAquick, MinElute (Qiagen) and Centri-SepTM (Thermo

Fisher Scientific). They can be purchased individually as a separate purification tool, or

some come already incorporated into extraction kits, such as the QIAamp® Investigator

Page 24 of 63

kit described above. Most use a silica membrane with a reduced diameter, allowing for

elution volumes as low as 5–7µL, making them highly suited to trace analysis in which the

DNA is often degraded [37]. Previous studies have shown the successful use of spin

columns to improve recovery of DNA from latent fingerprints [38], blood, saliva and

semen [3], and ancient bone samples [37]. The disadvantage of spin columns is their

single use, making them expensive when required for high sample throughput [38]. They

also suffer the same fate of all silica-based binding technology in that DNA loss can occur

due to inefficient adsorption and elution [22].

4.3 Ultrafiltration

A more cost-efficient approach for simultaneous concentration and purification is

ultrafiltration (UF), also known as microdialysis. UF is a pressure-driven filtration process

that utilises a semi-permeable membrane to separate molecules based on their size and

shape [39]. Separation is usually achieved with centrifugal force, and filter devices are

conveniently manufactured to fit in the rotors of standard laboratory centrifuges [40]. UF

membranes consist of specifically sized pores that retain larger particles, and ‘reject’

smaller components allowing them to pass through [41]. As in all filtration applications,

the permeability of a filter medium can be affected by the chemical, molecular or

electrostatic properties of the sample [42]. Accordingly, there has been much research

into selecting the right material for the membrane which will optimise the binding

properties of DNA [10].

Ultrafiltration technology originated from protein dialysis, in which solutes were

separated by diffusion through a selective membrane barrier [43]. Since the first

discoveries of dissolved nucleic acids in marine and freshwater samples, the technology

has rapidly evolved to encompass diverse approaches throughout many different

Page 25 of 63

scientific fields [44]. The emergence of environmental DNA (eDNA) and metabarcoding

have been enormous for gaining insight into the prevalence of aquatic species and

estimating biodiversity [45]. Aqueous eDNA is challenging to isolate because it generally

occurs at very low concentrations (<200pg/L) and can be heterogeneously distributed

throughout a water body [44]. Ultrafiltration is used to concentrate and purify dissolved

DNA from large volumes of water, replacing classical dialysis methods which are time

poor. [36]. Multiple studies have determined that more DNA can be isolated through

filtration compared to chemical precipitation methodologies of the same volume [10, 36,

44, 46]. Hinlo et al. [10] also found filtration to be a superior technique, as there is no

phase change or possible degradation of the DNA. In addition to marine and

environmental biology, ultrafiltration is commonly applied to microbial analysis; purifying

bacterial and fungal DNA from soil and sediment [47], food and beverages [48], and

ventilation system air filters [49]. A common theme to these studies is that the filter

material, pore size and extraction method all have a significant effect on DNA recovery.

More research is needed in this area, particularly in regards to human forensic

investigation.

5. CENTRIFUGAL FILTERS IN FORENSICS

Filtration and microdialysis technology has been used in forensic DNA analysis since

the advent of PCR-based methods [2]. Following DNA extraction, centrifugal filter devices

can serve as powerful tools in concentration and desalting procedures [39]. Their purpose

is to improve DNA for downstream applications, by both washing away PCR inhibitors and

concentrating the nucleic acid in the sample. The UF membrane allows low MW inhibitory

substances to pass through into the retenate, whilst recovering and retaining high MW

Page 26 of 63

DNA above the membrane [50]. A simple wash stage or reverse spin can then elute the

recovered DNA from the membrane [51].

5.1 Improving the profiling success

Filtration-based purification methods are popular in forensic analysis as they

improve both the quantity and quality of DNA needed for successful STR profiling [33]. No

single extraction procedure can exclusively isolate and recover high concentrations of

DNA simultaneously. Organic extraction and chelex methods are notorious for co-

extracting impurities that are subsequently responsible for preventing or reducing the

amplification of DNA [3, 13, 15]. Commonly encountered PCR inhibitors include heme

from blood, humic acid from soil samples, collagen from bones, phenol and

ethylenediaminetetraacetic acid (EDTA) [13]. Solid-phase extraction methods frequently

result in loss of DNA, either through multiple wash stages, inefficient binding or elution

from a silica substrate [22]. This may produce non-existent or partial profiles with a high

incidence of stochastic effects, therefore making interpretation difficult [13, 29].

Centrifugal filters have successfully been applied as an additional clean-up step in several

previous studies, following chelex [2, 11, 15, 20, 52], phenol-chloroform [13, 21, 23, 24,

26, 50, 53] and silica-based [20, 26] extraction methods. Purification improved the

profiling success of DNA from a variety of biological sources including those deposited on

substrates previously known to inhibit PCR, such as denim [26, 39, 53]. Likewise, the

concentrating potential of such filters has been demonstrated in Noren et al. [2] who

showed that filtration improved amplification of challenging samples, with increased

allele peak heights observed for saliva extracted from envelopes.

Page 27 of 63

5.2 Degraded samples

The ability to recover DNA from skeletal remains is a valuable tool for the

identification of remains and missing persons. However, extraction often yields low levels

of DNA due to the hard and complex structure of bones and teeth [4]. Abrasive organic

methods are still preferred for ancient biological material, but endogenous environmental

inhibitors frequently co-extract [21]. The potential for degraded DNA due to

environmental exposure is also high, which can further inhibit analysis and impact STR

profiles [13]. Concentration and purification with a centrifugal filter can help to address

these problems; removing small sheared DNA fragments and MW inhibitors, whilst

retaining high quality, amplifiable DNA. Iyavoo et al. [13] improved the purity of extracted

DNA from pig bones using an amicon filter (Millipore) instead of conventional ethanol

precipitation. Similar results were found by Amiel [21] and Yang et al. [37] with

quantification revealing no inhibitors were present after purification. Akane et al. [15]

used centricon filters (Millipore) to remove haemoglobin impurities from putrefied liver,

aged bloodstains and adipose tissue samples.

5.3 Trace samples

Centrifugal filters can help to improve the analysis of trace samples containing low

amounts of DNA. Concentrating the DNA post-extraction can increase the amount

available for input into an STR typing reaction [9]. Foran et al. [23] found that organic

extraction, followed by concentration with an amicon filter increased the yield of

exogenous DNA recovered from fingernail scrapings. Solomon et al. [26] evaluated the

effect of concentrating DNA from latent fingerprints with a microcon (Millipore) filter. The

results showed an increase in peak height, improved interlocus peak balance and

identified three additional STR alleles compared to untreated fingerprint samples.

Page 28 of 63

Another common source of trace evidence is in regards to firearms and the DNA

found on bullets, cartridge cases and unspent ammunition. Analysis of this type of

evidence can be extremely useful, directly linking a person of interest to a crime. As with

all trace evidence, difficulties arise with analysis due to low amounts of DNA and

degradation [28]. Fan et al. [52] explored the validity of centrifugal concentrators for

improving DNA recovery from firearms and cartridge cases in forensic casework. Applying

a microcon filter post-extraction with chelex resulted in an average of 16.2% of expected

alleles being observed compared to 1.6% with just chelex. Alternatively, Dieltjes et al. [28]

did not concentrate their samples post-extraction when recovering DNA from

ammunition. Out of 4,085 individual items, the STR success rate was very low at 6.9%

even employing highly sensitive PCR methodology such as a successive elution stage.

Current research suggests that they might have improved their results if they had applied

a centrifugal filter post-extraction.

6. OPTIMISING DNA RECOVERY WITH CENTRIFUGAL FILTERS

Centrifugal filters come in various designs based on membrane type, orientation

and area, and pore size [2]. Given their widespread integration into forensic DNA analysis,

there are now multiple manufacturers of these filter devices including Pall, Vivaproducts

and Millipore. Table 1 represents some of the most common filters used in previous

studies and compares the various membrane types and sizes.

Page 29 of 63

Table 1. A comparison of commercially available centrifugal filters commonly used for concentration of DNA samples in forensic casework.

Manufacturer Name NMWLa Membrane type Orientation Reference

Merck/

Millipore

Amicon®

Ultra-0.5

3,10,30,

50,100K

Ultracel®

(low-binding

regenerated cellulose)

Vertical [23, 50, 54,

55]

Amicon®

Ultra-2 30K Ultracel® Vertical [2]

Amicon®

Ultra-4

10,30,

50K Ultracel® Vertical [21]

Microcon® 50,100K Regenerated cellulose Vertical [25, 26, 50,

52, 54]

Pall Microsep

Advance 30K

Omega® (modified

polyethersulfone) Vertical [2]

Sartorius Vivaspin®

500 50K

Hydrosart®

(regenerated

cellulose)

Horizontal [50]

aNMWL= Nominal molecular weight limit refers to the minimum size particles must be to be

retained above the membrane (measured in kilodaltons).

6.1 Membrane design

As it stands, there is an overwhelming choice of commercially available devices and

no specific guidelines as to which product is the most suitable beyond the size and

general type of the target molecule. Multiple studies have indicated that the design of the

membrane has a significant effect on the overall performance of the filter and

subsequent ability to recover DNA. For example, Noren et al. [2] compared two well-

known purification devices with various mock crime scene stains. The Amicon Ultra 30K

(Millipore) and Microsep 30K (Pall) gave significantly different recovery rates of 62–70%

and 14–32% respectively. Noren et al. [2] concluded that the different performances of

the filter devices were likely caused by the quality of the filters and plastic wares.

Beckwith et al. [50] found that the orientation of the membrane plays a role in DNA

recovery, with vertical membranes offering improved removal of inhibitors compared to

horizontal membranes. They also determined that although polyethersulfone membranes

work well with proteins [44], modified regenerated cellulose (such as Hydrosart®) offer

Page 30 of 63

better recoveries of nucleic acids [50]. This is perhaps why the microsep filter in the study

by Noren et al. [2] was so ineffective at recovering DNA since its membrane is made of a

modified polyethersulfone material. Amicon Ultra devices incorporate an Ultracel®

membrane—a highly uniform regenerated cellulose material that has a specialised

hydrophilic structure to ensure high adsorption of DNA molecules [42]. The vertical design

of their membrane gives an additional advantage due to a functional dead space

preventing drying out of the sample if spun for too long [54].

Amiel [21] investigated the importance of membrane pore size and the effect that

it has on the recovery of ancient DNA. Ultrafiltration membranes are rated according to

their nominal molecular weight limit (NMWL) or MW cut-off, which is based on their

ability to retain >90% of a solute of a known MW, measured in kilodaltons (K) [41]. Due to

the unique structures of nucleic acids, selecting an appropriate NMWL is related more

closely to the length of the molecule rather than the weight [42]. A 30K membrane is the

recommended size for recovering DNA fragments 137–1,159 base pairs long [21].

Particularly with degraded samples such as bone, there is a trade-off between catching

small DNA fragments but also retaining small MW inhibitors. Using 50K filters, Amiel

observed DNA loss and allelic dropout, due to the smaller DNA fragments passing through

the filter. For bone, she found the optimal size for recovery and retention of DNA was

10K. The 10K filter had the highest quantification values, allele peak heights and more

reportable alleles compared to 30K and 50K NMWL.

6.2 DNA loss

Factors such as membrane design do not seem to make a massive difference when

purifying samples containing high amounts of DNA. Beckwith et al. [50] found that the

Page 31 of 63

recovery of concentrated DNA from mouth swabs and liquid blood were statistically

insignificant between the Microcon 50K and Vivacon 50K filters. However, when

processing trace quantities, maximising the amount of recoverable DNA is extremely

important for successful STR typing [50]. An evaluation of commonly used centrifugal

filter devices showed DNA loss in multiple studies [2, 21, 54, 55]. The results from real

forensic casework are vastly different from the 90% or higher rates of recovery quoted by

manufacturers. Figure 2 shows the typical recovery of nucleotides from Amicon® Ultra-0.5

devices, accessed directly from the Millipore user guide.

Figure 2. [51] Typical rates of DNA recovery quoted by Millipore — manufacturers of

Amicon® Ultra-0.5 filter devices.

Noren et al. [2] evaluated the Microsep® 30K and the Amicon®-Ultra 30K for their

ability to recover DNA from various mock crime scene samples. The amicon filter

experienced significantly less DNA loss than the microsep, with average recovery rates of

30–38% and 68–86% respectively, experienced across five replicates. Garvin et al. [54]

also observed up to 67% loss of DNA from low-level blood, semen and buccal samples

purified with an amicon filter. They theorised that the loss could be attributed to the high

force of the centrifuge and spent considerable time optimising the speed of the rotor to a

Page 32 of 63

lower setting. Similarly, Schiffner et al. [25] identified that the centrifuge speed and time

recommended by the manufacturer resulted in low DNA recovery, observing an average

of only 36% true alleles. Reducing the speed of the centrifuge increased the spin time but

also improved the percentage of correctly identified alleles to 81.53%.

Other authors have suggested that DNA loss can be attributed to the entrapment of

DNA in the device membrane [23, 25, 55]. Membrane fouling is a common problem in

filtration devices, blocking pores and reducing permeability and removal efficiency [40].

Doran et al. [55] found that using Amicon® filter devices of all MW cut-offs (3K-100K)

resulted in poor recovery, with low MW DNA being trapped on the membrane. They

investigated DNA loss by pre-treating filters with different materials such as glycogen,

bovine serum albumin and salmonid RNA. They found the most success with yeast RNA

and salmonid DNA, which gave rise to greatly improved DNA recovery and good

amplification curves. This result was confirmed in other studies by Schiffner et al. [25] and

Foran et al. [23] who used RNA treated filters to improve the yield of DNA recovered from

various samples. Pre-treating centrifugal filters appears critical to prevent loss of DNA on

the membrane, and it is expected that manufacturers will integrate this into their filter

design in time. Overall, future investigation needs to revolve around applying centrifugal

filters to samples that are expected to contain small quantities of DNA. More research

needs to be conducted on the most efficient filters for this purpose including the

membrane design, optimal MWCO and centrifuge conditions.

7. EXPERIMENTAL AIMS AND OBJECTIVES

This study aims to optimise methodologies for efficient DNA extraction of low

concentration forensic samples. Better techniques need to be established which will

Page 33 of 63

maximise the recovery of trace DNA samples during the extraction process, thereby

increasing the likelihood of generating a viable profile. The experiment will evaluate the

absolute efficiency of silica column purification followed by concentration with a

centrifugal filter when applied to samples known to contain low levels of DNA.

The objectives of this research project are as follows:

i. Investigate the efficiency of QIAamp® Investigator kit to extract DNA from low

yield forensic samples

ii. Determine the optimum elution technique which will maximise DNA

concentration required for downstream processing of forensic DNA samples

following extraction with QIAamp® Investigator kit

iii. Quantitatively compare the performance of centrifugal filters from three

different manufacturers (Pall, Millipore and Vivaproducts) and their ability to

concentrate DNA post-extraction

8. CONCLUSION

Trace samples are particularly challenging for forensic analysis as they contain low

quantities of DNA and often produce partial or non-existent STR profiles. There is no

standard method that can effectively and consistently isolate the DNA in sufficient quality

and quantity adequate for successful STR analysis. Silica-based extraction methods

appear the most promising, yielding pure extracts but suffering from sample loss.

Research has suggested that this problem may be overcome with the use of an

appropriately sized filtration device, applied to concentrate and purify the DNA sample

post-extraction. Unfortunately, there is an overwhelming choice of filter devices available

commercially and no specific guidelines available as to which is the most suitable for trace

Page 34 of 63

DNA analysis. It has also been proven that factors such as membrane design, pore size

and pre-treatment with a carrier molecule can all affect the recovery and retention of

DNA. Furthermore, multiple studies have suggested that ultrafiltration can further

contribute to DNA loss, therefore negating their use for this purpose. Additional

investigation is needed into optimising silica-based extraction methods for trace DNA

analysis, including modification of the elution procedure, and exploration into the ability

of different centrifugal filters to improve the recovery of DNA.

Page 35 of 63

9. REFERENCES

1. Butler JM. Advanced Topics in Forensic DNA Typing : Methodology. San Diego,

UNITED STATES: Elsevier Science & Technology; 2011.

2. Norén L, Hedell R, Ansell R, Hedman J. Purification of crime scene DNA extracts

using centrifugal filter devices. Investigative Genetics. 2013;4(1):8.

3. Hudlow WR. The NucleoSpin((R)) DNA Clean‐up XS kit for the concentration and

purification of genomic DNA extracts: an alternative to microdialysis filtration.

Forensic science international: genetics. 2011;5(3):226-30.

4. Lee S, G. Shewale J. DNA Extraction Methods in Forensic Analysis. 2017. p. 54, 74-

81.

5. Pádár Z, Zenke P, Kozma Z. Forensic DNA Technological Advancements as an

Emerging Perspective on Medico-Legal Autopsy: A Mini Review. Post Mortem

Examination and Autopsy-Current Issues From Death to Laboratory Analysis: InTech;

2018.

6. Cavanaugh SE, Bathrick AS. Direct PCR amplification of forensic touch and other

challenging DNA samples: A review. Forensic Science International: Genetics.

2018;32:40-9.

7. Mapes AA, Kloosterman AD, Marion V, Poot CJ. Knowledge on DNA Success Rates to

Optimize the DNA Analysis Process: From Crime Scene to Laboratory. Journal of

Forensic Sciences. 2016;61(4):1055-61.

8. Smith PJ, Ballantyne J. Simplified Low-Copy-Number DNA Analysis by Post-PCR

Purification. Journal of Forensic Sciences. 2007;52(4):820-9.

9. van Oorschot RA. Forensic trace DNA: a review. Investigative genetics. 2010;1(1).

10. Hinlo R, Gleeson D, Lintermans M, Furlan E. Methods to maximise recovery of

environmental DNA from water samples. PLOS ONE. 2017;12(6):e0179251.

11. Castella V, Dimo-Simonin N, Brandt-Casadevall C, Mangin P. Forensic evaluation of

the QIAshredder/QIAamp DNA extraction procedure. Forensic Science International.

2006;156(1):70-3.

12. McKiernan HE, Danielson PB. Chapter 21 - Molecular Diagnostic Applications in

Forensic Science. In: Patrinos GP, editor. Molecular Diagnostics (Third Edition):

Academic Press; 2017. p. 371-94.

13. Iyavoo S. Development and application of a PCR multiplex to assess the quality and

quantity of forensic DNA extracts [PhD]. Ann Arbor: University of Central Lancashire

(United Kingdom); 2014.

14. Goodwin W. Forensic DNA typing protocols. Second ed. New York: Humana Press;

2016.

Page 36 of 63

15. Akane A, Shiono H, Matsubara K, Nakamura H, Hasegawa M, Kagawa M. Purification

of forensic specimens for the polymerase chain reaction (PCR) analysis. Journal of

forensic sciences. 1993;38(3):691.

16. Shewale JG, Liu RH. Forensic DNA analysis: current practices and emerging

technologies. Boca Raton, FL: CRC Press, Taylor & Francis Group; 2014.

17. Hochmeister MN. DNA technology in forensic applications. Molecular Aspects of

Medicine. 1995;16(4):315-437.

18. Watanabe Y, Takayama T, Hirata K, Yamada S, Nagai A, Nakamura I, et al. DNA

typing from cigarette butts. Legal Medicine. 2003;5(1):S177-S9.

19. Phillips K, McCallum N, Welch L. A comparison of methods for forensic DNA

extraction: Chelex-100® and the QIAGEN DNA Investigator Kit (manual and

automated). Forensic Science International: Genetics. 2011;6(2):282-5.

20. Ip SCY, Lin S-w, Lai K-m. An evaluation of the performance of five extraction

methods: Chelex® 100, QIAamp® DNA Blood Mini Kit, QIAamp® DNA Investigator

Kit, QIAsymphony® DNA Investigator® Kit and DNA IQ™. Science & Justice.

2015;55(3):200-8.

21. Amiel M. Effects of Amicon® Ultra-4 Centrifugal Devices on DNA Yield From Bone

Samples and Effects of Amplicon Rx™ Post-PCR: ProQuest Dissertations Publishing;

2014.

22. Katevatis C, Fan A, Klapperich CM. Low concentration DNA extraction and recovery

using a silica solid phase. PLOS ONE. 2017;12(5):e0176848.

23. Foran D, Lisa Hebda M, Doran A. Document Title: Trace DNA from Fingernails:

Increasing the Success Rate of Widely Collected Forensic Evidence. 2010.

24. Greenspoon SA, Scarpetta MA, Drayton ML, Turek S. QIAamp spin columns as a

method of DNA isolation for forensic casework. Journal of Forensic Science.

1998;43(5):1024-30.

25. Schiffner LA, Bajda EJ, Prinz M, Sebestyen J. Optimization of a simple, automatable

extraction method to recover sufficient DNA from low copy number DNA samples

for generation of short tandem repeat profiles. Croatian medical journal.46(4):578-

86.

26. Solomon AD. An Optimized DNA Analysis Workflow for the Sampling, Extraction,

and Concentration of DNA obtained from Archived Latent Fingerprints. Journal of

forensic sciences.63(1):47-57.

27. Wickenheiser RA. Trace DNA: A review, discussion of theory, and application of the

transfer of trace quantities of DNA through skin contact. Journal of Forensic

Sciences. 2002;47(3):442-50.

Page 37 of 63

28. Dieltjes P, Mieremet R, Zuniga S, Kraaijenbrink T, Pijpe J, de Knijff P. A sensitive

method to extract DNA from biological traces present on ammunition for the

purpose of genetic profiling. International Journal of Legal Medicine.

2011;125(4):597-602.

29. Gill P. Application of Low Copy Number DNA Profiling. Croatian medical journal.

2001;42(3):229-32.

30. Butts E, editor Exploring DNA extraction efficiency. Forensics@ NIST 2012 Meeting;

2012.

31. Qiagen. QIAamp DNA Investigator Handbook. 06/2012.

32. Desneux J, Pourcher AM. Comparison of DNA extraction kits and modification of

DNA elution procedure for the quantitation of subdominant bacteria from piggery

effluents with real‐time PCR. MicrobiologyOpen. 2014;3(4):437-45.

33. Marziali A. NUCLEIC ACIDS | Centrifugation. In: Wilson ID, editor. Encyclopedia of

Separation Science. Oxford: Academic Press; 2000. p. 3517-24.

34. Fregel R, González A, Cabrera VM. Improved ethanol precipitation of DNA.

ELECTROPHORESIS. 2010;31(8):1350-2.

35. Green MR, Sambrook J. Precipitation of DNA with ethanol. Cold Spring Harbor

Protocols. 2016;2016(12):1116-20.

36. Matsui K, Ishii N, Honjo M, Kawabata Z. Use of the SYBR Green I fluorescent dye and

a centrifugal filter device for rapid determination of dissolved DNA concentration in

fresh water. Aquatic Microbial Ecology. 2004;36:99-105.

37. Yang DY, Eng B, Waye JS, Dudar JC, Saunders SR. Improved DNA extraction from

ancient bones using silica‐based spin columns. American Journal of Physical

Anthropology: The Official Publication of the American Association of Physical

Anthropologists. 1998;105(4):539-43.

38. Sinelnikov A, Reich K. Materials and methods that allow fingerprint analysis and

DNA profiling from the same latent evidence. Forensic Science International:

Genetics Supplement Series. 2017;6:e40-e2.

39. McClintock JT. Forensic DNA analysis: a laboratory manual. London; Boca Raton, Fla;

CRC; 2008.

40. Sahai R. MEMBRANE SEPARATIONS | Filtration. In: Wilson ID, editor. Encyclopedia

of Separation Science. Oxford: Academic Press; 2000. p. 1717-24.

41. Cheryan M. MEMBRANE SEPARATIONS | Ultrafiltration. In: Wilson ID, editor.

Encyclopedia of Separation Science. Oxford: Academic Press; 2000. p. 1797-802.

42. Merck/Millipore. New Amicon Ultra centrifugal filter devices combine high speed

and high recovery for protein concentration and desalting. 2003:20.

Page 38 of 63

43. Torto N, Laurell T, Gorton L, Marko-Varga G. Recent trends in the application of

microdialysis in bioprocesses. Analytica Chimica Acta. 1998;374(2):111-35.

44. Goldberg CS. Critical considerations for the application of environmental DNA

methods to detect aquatic species. Methods in ecology and evolution.7(11):1299-

307.

45. Deiner K, Lopez J, Bourne S, Holman L, Seymour M, Grey EK, et al. Optimising the

detection of marine taxonomic richness using environmental DNA metabarcoding:

the effects of filter material, pore size and extraction method. Metabarcoding and

Metagenomics. 2018;2:1.

46. Borrell YJ, Miralles L, Do Huu H, Mohammed-Geba K, Garcia-Vazquez E. DNA in a

bottle—Rapid metabarcoding survey for early alerts of invasive species in ports.

PLOS ONE. 2017;12(9):e0183347.

47. Roose-Amsaleg CL, Garnier-Sillam E, Harry M. Extraction and purification of

microbial DNA from soil and sediment samples. Applied Soil Ecology. 2001;18(1):47-

60.

48. Daufin G, Escudier JP, Carrere H, Bérot S, Fillaudeau L, Decloux M. Recent and

emerging applications of membrane processes in the food and dairy industry. 2001;

p89-102.

49. Luhung I, Wu Y, Xu S, Yamamoto N, Chang VW-C, Nazaroff WW. DNA accumulation

on ventilation system filters in university buildings in Singapore. PloS one.

2017;12(10):e0186295.

50. Beckwith M BA, Robertson A, Phillips W, Jimenez M, Baldwin B, Vagell M. The Role

of Ultrafiltration Membranes in the Recovery of DNA with Centrifugal

Concentrators. San Antonio, TX: International Symposium on Human Identification;

2010.

51. Merck/Millipore Ltd. User Guide: Amicon Ultra-0.5 Centrifugal Filter Devices for

volumes up to 500uL: Millipore Corporation; 2011.

52. Fan G-Y, Li W, Li S-T, Zhang Q-Y. An evaluation of the performance of DNA recovery

from fired firearms and cartridge cases using microdialysis filtration. Forensic

Science International: Genetics Supplement Series. 2017;6:e246-e8.

53. Comey CT, Koons BW, Presley KW, Smerick JB, Sobieralski CA, Stanley DM, et al.

DNA extraction strategies for amplified fragment length polymorphism analysis.

Journal of Forensic Sciences. 1994;39(5):1254-69.

54. Garvin AM, Fritsch A. Purifying and Concentrating Genomic DNA from Mock

Forensic Samples Using Millipore Amicon Filters. Journal of Forensic Sciences.

2013;58(1): S173-S5.

Page 39 of 63

55. Doran AE, Foran DR. Assessment and mitigation of DNA loss utilising centrifugal

filtration devices. Forensic Science International: Genetics. 2014;13:187-90.

Page 40 of 63

Blank Page – not numbered

Page 41 of 63

Part Two

Manuscript

OPTIMISING SILICA-BASED SOLID PHASE DNA EXTRACTION METHODS FOR LOW CONCENTRATION FORENSIC SAMPLES

Page 42 of 63

ABSTRACT

In forensic laboratories, increased extraction efficiency of trace evidence is paramount

because analytical success is intrinsically dependent on the quantity of DNA recovered.

Moreover, highly concentrated nucleic acids are vital for effective downstream analysis

and higher-quality results. This study investigated the efficiency of extraction with

QIAamp DNA Investigator kit, and explored improvements to the methodology that

would maximise the recovery of low concentration forensic samples. Addition of an RNA

carrier and performing two successive elutions of 50µL improved the recovery of DNA to

95%. Concentration with a centrifugal filter post-extraction is not recommended for trace

evidence as substantial DNA loss was observed. Additional research is required into the

causes of DNA loss with these filter devices and investigation of preventative measures

before they can be recommended for forensic casework.

KEYWORDS

Forensic DNA analysis; absolute extraction efficiency; elution volume; centrifugal filters;

QIAamp DNA Investigator Kit

Page 43 of 63

TABLE OF CONTENTS

ABSTRACT ............................................................................................................................ 42

KEYWORDS........................................................................................................................... 42

LIST OF FIGURES................................................................................................................... 44

LIST OF TABLES..................................................................................................................... 45

LIST OF ABBREVIATIONS ...................................................................................................... 46

1. INTRODUCTION ............................................................................................................ 47

2. MATERIALS & METHODS .............................................................................................. 49

2.1 Sample preparation .............................................................................................. 49

2.2 Estimation of DNA input ....................................................................................... 49

2.3 DNA extraction ...................................................................................................... 50

2.4 Post-extraction concentration .............................................................................. 50

2.5 DNA quantification ................................................................................................ 51

3. RESULTS & DISCUSSION ............................................................................................... 51

3.1 Efficiency of Extraction ......................................................................................... 51

3.2 Optimising the Elution Procedure......................................................................... 54

3.3 Post-extraction Concentration with Centrifugal Filters ........................................ 55

4. CONCLUSION ................................................................................................................ 58

5. REFERENCES ................................................................................................................. 60

6. APPENDIX ..................................................................................................................... 63

Page 44 of 63

LIST OF FIGURES

Figure 1 – Recovery of DNA from low concentration forensic samples with QIAamp DNA

investigator kit & effect of various procedural modifications……………………………………….…53

Appendix 1 – Linear plot of DNA quantity versus cyclic threshold for calibration in

Quantifiler Trio qPCR………………………………………………………………………………………………………63

Page 45 of 63

LIST OF TABLES

Table 1 – Details of the six extraction methods compared in this study including elution

volume and use of a concentrating centrifugal filter device……………………………………………50

Table 2 – Spin conditions for filter devices used to concentrate samples post-

extraction……………………………………………………………………………………………………………………….51

Table 3 – Effect of different elution procedures on DNA recovery………………………………….55

Table 4 – Performance of centrifugal filter devices from different manufacturers and their

ability to recover DNA.……………………………………………………………………………………………………56

Page 46 of 63

LIST OF ABBREVIATIONS

STR – Short tandem repeat

DNA – Deoxyribonucleic acid

PCR – Polymerase chain reaction

QIAamp – QIAamp® DNA Investigator kit

MW – Molecular weight

Page 47 of 63

1. INTRODUCTION

The success of short tandem repeat (STR) typing in forensic investigation is heavily

dependent on the size, quality and purity of deoxyribonucleic acid (DNA) that can be

isolated from samples [1]. A viable genetic profile can only be generated if an extract is

sufficiently concentrated in genomic DNA and free of polymerase chain reaction (PCR)

inhibitors [2]. Trace forensic samples can be particularly challenging to investigators

because they contain only minute quantities of DNA. Standard laboratory methods

frequently result in the loss of a portion of the original DNA, and increase the opportunity

for exogenous sources to be introduced [3]. This may produce non-existent or partial

profiles with a high incidence of stochastic effects, therefore making interpretation

difficult [4, 5]. Maximising the recovery of DNA is therefore vital to improving the analysis

of trace samples [6].

DNA extraction is perhaps the most critical step in the analysis workflow, directly

affecting the amount of starting material available for further examination.

Unfortunately, there is no one-size-fits-all protocol available, and each method has its

advantages and limitations. For the successful analysis of trace DNA, forensic laboratories

employ solid-phase extraction methods as they are the most efficient at recovering high

amounts of pure DNA [6, 7]. One example of solid-phase methodology is a commercial kit

called QIAamp® DNA Investigator (QIAamp). Manufactured by Qiagen, this kit is suitable

for purification from trace samples, generating high-quality, ready-to-use genomic DNA

[8]. Isolation is achieved through the selective binding properties of a silica column,

allowing unwanted impurities to be washed away, and pure, concentrated DNA to be

eluted from the membrane. QIAamp is reported in the literature as an efficient extraction

kit for the majority of forensic samples [9-12], but there is little research into its ability to

Page 48 of 63

recover trace amounts of DNA. Furthermore, previous studies have failed to reflect the

absolute efficiency of the extraction process by considering the initial amount of DNA

present in a sample [13].

DNA extraction is not an infallible process, despite continuous adaptations to

procedures and optimisation for a range of samples. [7]. Solid-phase methods frequently

result in the loss of DNA, either through multiple wash stages, inefficient binding or

elution from a silica substrate [14]. Maximising the recovery of DNA during the extraction

of trace forensic samples is paramount as any loss of DNA may limit the amount available

for STR profiling. Conventional amplification systems often require a minimum input

amount of DNA to work at optimal levels and consistently generate full profiles [3, 5].

Standard extraction procedures do not currently recover DNA from trace samples in

sufficient concentrations required for downstream analysis. Modifications are therefore

needed to improve the quality and quantity of DNA that can be efficiently isolated.

The elution volume is an easily overlooked step which may limit the concentration

of DNA, potentially contributing to poor quality profiles [15]. Unfortunately, standardised

methods are lacking as to the appropriate elution volume for specific samples, including

trace DNA. Several authors have found success in a modified technique involving multiple

elutions of smaller volumes [10, 16, 17], but the procedure is labour-intensive and

relatively untested for the recovery of trace DNA. An alternative method to combat this is

the application of specialised filters designed to purify and concentrate DNA post-

extraction. Ultrafiltration membranes contain specific size pores that allow low molecular

weight (MW) inhibitory substances to pass through, whilst recovering and retaining high

MW DNA [18]. A simple wash stage or reverse spin can then elute the recovered DNA

from the membrane [19]. Centrifugal filters have successfully been applied in many

Page 49 of 63

previous studies, including the analysis of trace samples [1, 10, 16, 17, 20-23].

Unfortunately, DNA losses of up to 70% have been found by some authors [21, 24]

negating their use in forensic casework. There is also an overwhelming choice of

commercially available devices and no specific guidelines as to which product is the most

suitable beyond the size and general type of the target molecule. This study aims to

explore the performance of different centrifugal filters and evaluate their ability to

recover DNA from low concentration forensic samples.

2. MATERIALS & METHODS

Human ethics approval for this study (Project No: 2019/025) was successfully obtained

from the Human Research Ethics and Integrity Committee of Murdoch University on the

4th of April 2019.

2.1 Sample preparation

A male volunteer, who had not eaten for 30 minutes, thoroughly rinsed his mouth for 15

seconds with 0.9% saline to remove any food particles and contaminants. A saliva sample

was then immediately collected by the participant vigorously rinsing his mouth for 1

minute, with approximately 30mL of 0.9% saline.

2.2 Estimation of DNA input

A cell count was performed using a standard 9-grid haemocytometer and compound

microscope at 40X magnification. Duplicates of 10µL saliva suspension were analysed

across all 9 grids by two independent reviewers and an average concentration of 70

cells/1µL obtained across all four replicates. The assumption of each cell containing

roughly 6 picograms of DNA, was used to determine that 15µL of cell suspension would

represent approximately 6.3ng of DNA.

Page 50 of 63

2.3 DNA extraction

All samples were manually extracted using QIAamp® DNA Investigator kit (Cat. No. 56504)

following the manufacturer’s (Qiagen) protocol for isolation of total DNA from small

volumes of blood or saliva [8]. For each method (see table 1 below), 5 replicates were

tested and a blank extraction. Samples were eluted in either 50µL or 100µL of buffer ATE

depending on the method. For procedures B and G, two successive eluates of 50µL were

performed as per Desneux and Pourcher [16], with the volume obtained after each

centrifugation combined into one tube (100µL).

Table 1. Details of the six extraction methods compared in this study, including elution

volume and use of a concentrating centrifugal filter device.

Method Extraction Elution volume Concentrating filter device

A QIAamp Investigator 100µL N/A

B QIAamp Investigator 50µL N/A

C QIAamp Investigator 50µL+ 50µL N/A

D QIAamp Investigator 100µL Amicon® Ultra-0.5 30K

(Millipore)

E QIAamp Investigator 100µL MicrosepTM

Advance 30K (Pall)

F QIAamp Investigator 100µL Vivaspin®500 30K (Sartorius)

G QIAamp Investigator + carrier

RNA added to Buffer AL 50µL+ 50µL N/A

*QIAamp Investigator refers to manual extraction with QIAamp DNA Investigator kit (Qiagen).

2.4 Post-extraction concentration

Following extraction, samples from methods D-F were further concentrated with a

centrifugal filter according to table 1. 100µL extracts were placed directly onto the

membrane of either an Amicon® Ultra-0.5 30K (Merck Millipore), MicrosepTM Advance

30K (Pall) or Vivaspin®500 30K (Sartorius) filter device. Centrifugation was carried out

according to the manufacturer’s instructions [19, 25, 26] with detailed spin conditions

described in table 2 below. One wash stage with 300µL low TE buffer (10mM Tris-pH 7.5;

0.1mM EDTA) was performed with the Amicon filter. After spinning, the DNA concentrate

Page 51 of 63

was collected from the Microsep filter by pipetting directly from the sample reservoir. For

the Amicon and Vivaspin filters, the membrane was inverted and immediately reverse

spun for 2 minutes at 1,000 x g and 2,500 x g respectively.

Table 2. Spin conditions for filter devices used to concentrate samples post-extraction.

Method Filter device Device

MWCOa

Centrifuge

rotor

Spin time

(min)

Speed

(x g)

D Amicon® Ultra-0.5 30K 40° Fixed angle 10 14,000

E Microsep

TM

Advance 30K Swing bucket 10 4,000

F Vivaspin®500 30K 40° Fixed angle 10 5,000 aMWCO=Molecular weight cut-off rating (Kilodaltons)

All DNA extraction and purification procedures were performed in a dedicated DNA

laboratory which was sterilised with 16% sodium hypochlorite before use. Personal

protective equipment (hooded coveralls, gloves, hairnet, face mask, shoe coverings) was

worn at all times and every effort made to reduce contamination throughout the analysis.

2.5 DNA quantification

DNA extracts were prepared as 20µL reactions using QuantifilerTM Trio DNA

Quantification Kit (Cat. No. 4482910) according to the manufacturer’s (Applied

Biosystems) user guide [27]. 5 standards were also prepared by 10-fold dilution of

Quantifiler THP DNA standard (50ng/µL) for production of a standard curve. Quantitation

was performed in a 384 well plate and run for 40 cycles in a QuantStudioTM 6 Flex Real-

Time PCR System according to the manufacturer’s (Applied Biosystems) instructions.

Results were analysed in Microsoft Excel.

3. RESULTS & DISCUSSION

3.1 Efficiency of Extraction

Page 52 of 63

For the successful analysis of trace DNA, forensic laboratories preferentially employ

solid-phase extraction methods as they are the most efficient at recovering high amounts

of pure DNA [6, 7]. The QIAamp DNA Investigator kit which utilises a silica column

substrate, is well documented as offering good recovery of DNA from a wide range of

samples compared to other extraction procedures [9, 11, 12]. Quantification results from

this study reveal that the efficiency of extraction of low concentration forensic samples

using QIAamp DNA Investigator kit is overall average. Referring to figure 1 below, elution

volumes of 100µL and 50µL show a recovery efficiency of 49% and 48% respectively. This

equates to a loss of approximately 50% of the initial sample. The low extraction efficiency

is likely to result in reduced DNA quantity available for amplification, which may fail to

produce a complete STR profile. DNA losses of up to 85% have been reported in similar

studies, although recovery is dependent on the extraction chemistry and source of DNA

[13, 28, 29]. To date, there has been no research into the absolute extraction efficiency of

Qiagen’s Investigator kit; therefore a comparison cannot be made.

DNA loss has previously been attributed to multiple wash stages and the

transference of samples between tubes [30]. Manual pipetting is a common source of

material loss in laboratories which has been largely reduced with the onset of automated

platforms. A more likely reason for decreased extraction efficiency is due to the capacity

of DNA to adsorb to plastic consumables and extraction matrices [3, 29]. The literature

theorises that poor DNA recovery using a solid silica phase is dependent on two inherent

mechanisms; (1) inefficient adsorption onto the silica column, and (2) ineffective elution

of purified DNA from the substrate [14].

The QIAamp DNA investigator kit is promoted as suitable for DNA purification from

trace samples, offering flexible elution volumes of 20-100µL [8]. Figure 1 shows that

Page 53 of 63

lowering the elution volume from 100µL to 50µL had a negligible effect on the efficiency

of extraction with Qiagen’s investigator kit. Alternatively, performing a successive elution

of 2x50µL recovered up to 15% more DNA from the initial sample, increasing the

efficiency to 63%. This suggests that a portion of DNA remains bound to the substrate and

cannot be recovered efficiently in a single elution step as per the manufacturer’s

instructions. While the driving forces for DNA elution from silica are not well understood,

pH and temperature appear important factors in regulating the binding affinity [14, 31].

Additional research is needed into the optimal conditions required for efficient recovery

of trace DNA during the elution stage.

Figure 1. Recovery of DNA from low concentration forensic samples with QIAamp DNA

investigator kit & effect of various procedural modifications–elution protocol and addition

of carrier RNA. Absolute extraction efficiency% calculated based on initial DNA estimate

and actual DNA amount obtained from quantification results.

Further optimisation of the extraction method revealed that adding a carrier RNA to

the cell lysis buffer significantly enhanced the recovery of DNA. Figure 1 shows a 32%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Ab

solu

te e

xtra

ctio

n e

ffic

ien

cy (

%)

Optimising the Efficiency of Extraction of Low Yield Forensic Samples with QIAamp DNA Investigator Kit

100µL elution 50µL elution

2x50µL successive elutions 2x50µL successive elutions + RNA carrier

Page 54 of 63

increase in the absolute extraction efficiency when carrier RNA was applied to the

successive elution method. This improved the overall efficiency of QIAamp Investigator to

95%. A similar result has been previously reported by Kishore et al. [29] who increased

DNA recovery from low-yield forensic samples by an average of 24% after addition of

carrier RNA.

The mechanism by which carrier RNA enhances the recovery of DNA extractions is

unknown, but one possibility is that the RNA blocks sites within the silica column, which

would otherwise retain sample DNA [29]. The extraction efficiency therefore increases

due to irreversible binding of the RNA carrier, which subsequently improves the elution of

target DNA [32]. Qiagen recommends adding carrier RNA to the lysis buffer for the

purification of DNA from minimal amounts of a sample [8]. When working with trace

samples, this appears a critical step in maximising DNA recovery and should be

permanently included in the extraction method.

Efficiency is a common benchmark in the scientific literature, providing a means for

studies to simultaneously compare the performance of different extraction methods. In

order to reflect the absolute efficiency of the extraction process, the initial quantity of

DNA present in a sample must be taken into account [13]. The results of this study are

limited by the method used to estimate the exact amount of DNA available before

extraction. Cell-counting is inexpensive and straightforward, but labour-intensive, and

subject to human error [33].

3.2 Optimising the Elution Procedure

The results of this study confirmed that eluting in a lower volume gave a higher

concentration of DNA. Table 3 shows that in a 100µL elution, DNA was recovered at a

concentration of 0.031ng/µL. Reducing the elution volume to 50µL; recovered DNA at

Page 55 of 63

approximately double the concentration at 0.061ng/µL. In terms of total DNA yield,

decreasing the elution volume did not improve the amount of DNA recovered. The 100µL

and 50µL recovered 3.080ng and 3.040ng, respectively. For the purposes of downstream

analysis, it is likely that eluting in a lower volume and obtaining a higher concentration of

DNA would produce higher-quality amplification results. However, further recovery of

DNA would be negligible, which may limit successful profiling.

Table 3. Effect of different elution procedures on DNA recovery.

Method Initial DNA

a

(ng)

Volume of extract

(µL)

Concentration

(ng/µL) Actual DNA (ng)

A 6.300 100 0.031 ± 0.002 3.080 ± 0.217

B 6.300 50 0.061 ± 0.012 3.040 ± 0.600

C 6.300 100 0.040 ± 0.007 3.960 ± 0.733 aInitial DNA estimated from cell count (see method). All values represent an average of n=5

obtained using Quantifiler Trio.

Alternatively, eluting in multiple successive stages appears advantageous to

increasing both the concentration and recovery of DNA during extraction. Employing

2x50µL elutions obtained 3.960ng of DNA at a concentration of 0.040ng/µL. This result is

comparable to other studies, which also observed improved DNA yield from multiple

repetitions of the elution stage [10, 16, 17]. Foran et al. [10] found that trace DNA from

fingernail scrapings was recovered efficiently in up to four elutions of 20µL, although DNA

quantity was negligible in the fourth. For the successful analysis of trace forensic samples,

modification of the final elution step could help to improve both the quantity and quality

of DNA available for downstream analysis.

3.3 Post-extraction Concentration with Centrifugal Filters

Filtration-based purification methods are popular in forensic analysis as they

improve both the quantity and quality of DNA needed for successful STR profiling. This

study demonstrates that centrifugal devices can be used to concentrate samples

Page 56 of 63

effectively post-extraction. The average DNA concentrations after filtration were

significantly higher in comparison to the concentrations calculated from the non-purified

100µL extracts (see tables 3 and 4). The Vivaspin (Sartorius) device gave the largest

concentration at 0.376ng/µL, followed by the Amicon (Merck/Millipore) device at

0.079ng/µL and lastly the Microsep (Pall) device at 0.043ng/µL. It is important to note

that whilst the Vivaspin filter achieved a much higher concentration than the other

devices, an average volume of 5.74µL is not practical for downstream analysis in a

realistic case. It is possible that a reduced spin time or maximum filling of the device with

TE buffer would help to prevent this in future use. In contrast, the Microsep device

recovered the lowest concentration, most likely because it was not spun for long enough

as indicated by the comparatively larger volume retained following concentration.

Through observations with the other filters, it is possible that further centrifugation of the

Microsep filter may have increased the concentration but resulted in less recovery of

DNA.

Table 4. Performance of centrifugal filter devices from different manufacturers and their

ability to recover DNA.

Filter device Concentrate

volume (µL)

Concentration

(ng/µL)

Actual DNA

(ng)

DNA Recoverya

(%)

Amicon® Ultra-

0.5, 30K

(M/Millipore)

21.72±1.73 0.079±0.012 1.717±0.254 56

MicrosepTM

Advance, 30K

(Pall)

54.92±3.00 0.043±0.009 2.385±0.615 77

Vivaspin®500,

30K (Sartorius) 5.74±1.57 0.376±0.072 2.097±0.386 68

aDNA recovery% is calculated based on comparison to the quantity of DNA in 100µL extract (see

method A). *All values represent an average of n=5, obtained using Quantifiler Trio.

Overall, there appears to be a trade-off between adequate sample concentration

and DNA loss when applying a centrifugal filter post-extraction. The amount of DNA

Page 57 of 63

recovered varied slightly, with the Microsep device recovering the most at 77%, followed

by the Vivaspin device at 68% and lastly the Amicon device at 56%. In comparison to the

49% efficiency of DNA recovery found solely from a 100µL extraction with QIAamp

Investigator, filtration is losing up to 23% additional DNA. For trace samples containing

only minute quantities of DNA, a loss of this proportion is likely to be severely limiting in

terms of successful STR analysis.

Whilst manufacturers quote typical DNA recoveries of greater than 90%, others

have noted similar high losses to what has been found in this study. Noren et al. [1]

compared the Amicon Ultra 30K and Microsep 30K devices and observed a recovery of

62-70% and 14-32%, respectively. The difference in recovery rate achieved in Noren et al.

compared to this study is possibly due to changes in the spin conditions. Noren et al. [1]

also applied a wash phase to their Microsep device, which they later concluded resulted

in lower DNA quantities. They suggested that more thorough washing/centrifugation

makes the DNA bind tighter to and/or get trapped within the filter. In this study, a single

wash was only applied to the Amicon device as per the manufacturer’s protocol.

Considering that this device is not typically used for purification of trace DNA, it is

reasonable to presume that the addition of a wash cycle may have contributed to low

DNA recovery. It is recommended that the wash cycle be omitted in future studies;

however, the trade-off between purity and recovery should be investigated further.

Several authors have also suggested that DNA loss can be attributed to the

entrapment of DNA in the device membrane [10, 21, 30]. Membrane fouling is a common

problem in filtration devices, blocking pores and reducing permeability and removal

efficiency [34]. DNA loss was the highest when using the Amicon device, only recovering

1.717ng of DNA or 56% of the input DNA quantity. Previous studies have implied spinning

Page 58 of 63

the device at maximum speed (14,000 x g) as per the manufacturer’s instructions is not

ideal for the recovery of trace DNA [10, 20, 21, 24]. Additional research by Garvin et al.

[24] found that a force of 3,500 x g allowed purification of low amounts of DNA with high

efficiency, in a reasonable time of 30 minutes. In a separate study, Doran et al. [21]

determined that DNA loss could be reduced by pre-treatment of the membrane with

yeast RNA. This prevents non-specific adsorption of nucleic acids in the same way that

RNA carrier acts on the silica column as discussed above. The use of RNA treated filters to

improve the yield of DNA recovered from various forensic samples has already been

confirmed in Schiffner et al. [30] and Foran et al. [10].

The different performances of the filter devices are due to the variation in

manufacturer design, specifically the membrane construction, and quality of the filters

and plastic wares [18]. Previous literature indicates that modified regenerated cellulose

membranes such as those found in Amicon and Vivaspin devices offer better recoveries of

nucleic acids [18, 35]. The results of this study refute that claim as the Microsep filter,

which incorporates a polyethersulfone membrane had the highest recovery of DNA.

Unfortunately, due to time and budget constraints, this study performed only

quantitative analysis. For a complete assessment of the performance of each filter and

the ability to recover quality DNA, STR profiling should be conducted in future

experiments. In addition, quantifying the unwanted flow-through could assist in

determining how much DNA is being lost through centrifugation versus how much is

irreversibly adhering to the membrane itself.

4. CONCLUSION

This study has demonstrated that the standard extraction of low concentration

forensic samples with QIAamp DNA Investigator kit is moderately efficient. Optimisation

Page 59 of 63

of the extraction method to include RNA carrier and employing successive elutions of

small volumes can significantly increase the efficiency of extraction, with DNA recoveries

of approximately 95%. Application of a centrifugal filter post-extraction can further

increase the concentration of samples for downstream analysis, but any benefit is

counteracted by substantial DNA loss. Additional research is required into the causes of

DNA loss with these filter devices and investigation of preventative measures before they

can be recommended for forensic casework.

Page 60 of 63

5. REFERENCES

1. Norén L, Hedell R, Ansell R, Hedman J et al. Purification of crime scene DNA extracts

using centrifugal filter devices. Investigative Genetics. 2013;4(1):8.

2. Hudlow WR. The NucleoSpin((R)) DNA Clean‐up XS kit for the concentration and

purification of genomic DNA extracts: an alternative to microdialysis filtration.

Forensic science international: genetics. 2011;5(3):226-30.

3. Cavanaugh SE, Bathrick AS. Direct PCR amplification of forensic touch and other

challenging DNA samples: A review. Forensic Science International: Genetics.

2018;32:40-9.

4. Iyavoo S. Development and application of a PCR multiplex to assess the quality and

quantity of forensic DNA extracts [PhD]. Ann Arbor: University of Central Lancashire

(United Kingdom); 2014.

5. Gill P. Application of Low Copy Number DNA Profiling. Croatian medical journal.

2001;42(3):229-32.

6. Butler JM. Advanced Topics in Forensic DNA Typing: Methodology. San Diego,

UNITED STATES: Elsevier Science & Technology; 2011.

7. Lee S, G. Shewale J. DNA Extraction Methods in Forensic Analysis. 2017. p. 54, 74-

81.

8. Qiagen. QIAamp DNA Investigator Handbook. 06/2012.

9. Castella V, Dimo-Simonin N, Brandt-Casadevall C, Mangin P. Forensic evaluation of

the QIAshredder/QIAamp DNA extraction procedure. Forensic Science International.

2006;156(1):70-3.

10. Foran D, Lisa Hebda M, Doran A. Document Title: Trace DNA from Fingernails:

Increasing the Success Rate of Widely Collected Forensic Evidence. 2010.

11. Ip SCY, Lin S-w, Lai K-m. An evaluation of the performance of five extraction

methods: Chelex® 100, QIAamp® DNA Blood Mini Kit, QIAamp® DNA Investigator

Kit, QIAsymphony® DNA Investigator® Kit and DNA IQ™. Science & Justice.

2015;55(3):200-8.

12. Phillips K, McCallum N, Welch L. A comparison of methods for forensic DNA

extraction: Chelex-100® and the QIAGEN DNA Investigator Kit (manual and

automated). Forensic Science International: Genetics. 2011;6(2):282-5.

13. Butts E, editor Exploring DNA extraction efficiency. Forensics@ NIST 2012 Meeting;

2012.

14. Katevatis C, Fan A, Klapperich CM. Low concentration DNA extraction and recovery

using a silica solid phase. PLOS ONE. 2017;12(5):e0176848.

Page 61 of 63

15. van Oorschot, RA. Forensic trace DNA: a review. Investigative genetics. 2010;1(1).

16. Desneux J, Pourcher AM. Comparison of DNA extraction kits and modification of

DNA elution procedure for the quantitation of subdominant bacteria from piggery

effluents with real‐time PCR. MicrobiologyOpen. 2014;3(4):437-45.

17. Dieltjes P, Mieremet R, Zuniga S, Kraaijenbrink T, Pijpe J, de Knijff P. A sensitive

method to extract DNA from biological traces present on ammunition for the

purpose of genetic profiling. International Journal of Legal Medicine.

2011;125(4):597-602.

18. Beckwith M BA, Robertson A, Phillips W, Jimenez M, Baldwin B, Vagell M. The Role

of Ultrafiltration Membranes in the Recovery of DNA with Centrifugal

Concentrators. San Antonio, TX: International Symposium on Human Identification;

2010.

19. Merck/Millipore Ltd. User Guide: Amicon Ultra-0.5 Centrifugal Filter Devices for

volumes up to 500uL: Millipore Corporation; 2011.

20. Amiel M. Effects of Amicon® Ultra-4 Centrifugal Devices on DNA Yield From Bone

Samples and Effects of Amplicon Rx™ Post-PCR: ProQuest Dissertations Publishing;

2014.

21. Doran AE, Foran DR. Assessment and mitigation of DNA loss utilising centrifugal

filtration devices. Forensic Science International: Genetics. 2014;13:187-90.

22. Fan G-Y, Li W, Li S-T, Zhang Q-Y. An evaluation of the performance of DNA recovery

from fired firearms and cartridge cases using microdialysis filtration. Forensic

Science International: Genetics Supplement Series. 2017;6:e246-e8.

23. Solomon AD. An Optimized DNA Analysis Workflow for the Sampling, Extraction,

and Concentration of DNA obtained from Archived Latent Fingerprints. Journal of

forensic sciences.63(1):47-57.

24. Garvin AM, Fritsch A. Purifying and Concentrating Genomic DNA from Mock

Forensic Samples Using Millipore Amicon Filters. Journal of Forensic Sciences.

2013;58(1):S173-S5.

25. Sartorius Stedim Lab Ltd. Vivacon500 Technical data and operating instructions.

2016(6).

26. Pall. Microsep Advance User Guide. 2014.

27. Quantifiler HP and Trio DNA Quantification Kits: User Guide: Thermo Fisher

Scientific; 2017.

28. Wood I, Park S, Tooke J, Smith O, Morgan RM, Meakin GE. Efficiencies of recovery

and extraction of trace DNA from non-porous surfaces. Forensic Science

International: Genetics Supplement Series. 2017;6:e153-e5.

Page 62 of 63

29. Kishore R, Reef Hardy W, Anderson VJ, Sanchez NA, Buoncristiani MR. Optimisation

of DNA extraction from low-yield and degraded samples using the BioRobot® EZ1

and BioRobot® M48. Journal of Forensic Sciences. 2006;51(5):1055-61.

30. Schiffner LA, Bajda EJ, Prinz M, Sebestyen J. Optimization of a simple, automatable

extraction method to recover sufficient DNA from low copy number DNA samples

for generation of short tandem repeat profiles. Croatian medical journal.46(4):578-

86.

31. Vandeventer PE, Mejia J, Nadim A, Johal MS, Niemz A. DNA Adsorption to and

Elution from Silica Surfaces: Influence of Amino Acid Buffers. The Journal of Physical

Chemistry B. 2013;117(37):10742-9.

32. Kashkary L, Kemp C, Shaw KJ, Greenway GM, Haswell SJ. Improved DNA extraction

efficiency from low-level cell numbers using a silica monolith based microfluidic

device. Analytica Chimica Acta. 2012;750:127-31.

33. Samie L, Champod C, Glutz V, Garcia M, Castella V, Taroni F. The efficiency of DNA

extraction kit and the efficiency of recovery techniques to release DNA using flow

cytometry. Science & Justice. 2019.

34. Sahai R. MEMBRANE SEPARATIONS | Filtration. In: Wilson ID, editor. Encyclopedia

of Separation Science. Oxford: Academic Press; 2000. p. 1717-24.

35. Merck/Millipore. New Amicon Ultra centrifugal filter devices combine high speed

and high recovery for protein concentration and desalting. 2003:20.

Page 63 of 63

6. APPENDIX

Appendix 1. Linear plot of DNA quantity versus cyclic threshold for calibration in Quantifiler Trio qPCR. Standards were prepared by 10-fold dilution of 50ng/µL.

y = -1.523ln(x) + 21.06 R² = 0.9978

y = -1.472ln(x) + 24.231 R² = 0.9982

y = -1.538ln(x) + 25.596 R² = 0.9984

12

17

22

27

32

0.001 0.01 0.1 1 10 100

Thre

sho

ld c

ycle

(C

T)

Quantity (copies)

Standard Curve

Large amplicon

Male Y

Small autosomal