Directions for developing a presumptive
immunogenic DNA quantification method
By Mareen Boel (10544291)
Supervised by A. van Dam Co-assessed by A.D. Kloosterman
MSc Forensic Science, University of Amsterdam Submitted on 24-01-2018
Abstract
On-site testing at the crime scene to determine the quantity of DNA present on an item of evidence would help investigators to anticipate the result of further analysis and thus assist in selecting the items for submission. This article seeks the appropriate strategy to develop such a tool based on the principle of lateral flow assay, because these devices offer a number of benefits. Various targets are examined in order to identify the molecules that are potentially indicative for DNA quantity. The shelterin complex, consisting of proteins that bind to the telomere, is considered a suitable option. Furthermore, the advantages and disadvantages of diverse detection methods are considered and a semi-quantitative approach with a sandwich format is recommended to realize an effective and convenient method.
2 Introduction
DNA evidence from a crime scene has generally a high probative value due to its individualizing character. Therefore, it is valuable to find and collect items containing DNA evidence from a crime scene. DNA can be present in all kinds of traces. Traces consisting of body fluids, such as blood, saliva, semen and vaginal secretion contain high concentrations of DNA. These kind of traces are relatively easy to detect, because they tend to be deposited in stains and thus be visible to the naked eye. Other traces left by other forms of contact may not be visible to the naked eye but can still contain DNA. For example, contact traces can contain released skin cells contain DNA which can be collected and used for comparative DNA analysis (Kloosterman & Meulenbroek, 2008). However, those invisible traces require more effort to be detected. The investigator has to deduce the location of possible traces based on a scenario and is therefore often dependent on information provided by other items of evidence and clues. The submission of evidence for DNA traces is thus based on a fairly subjective process, as it is a result of a reasoning process by the investigators. Moreover, the uncertainty about the presence of a sufficient amount of DNA required for a DNA profile leads to the submission of multiple evidential items, occupying the laboratory its capacity without any assurance of obtaining any result (Meulenbroek, 2006).
For this reason, assessment of the quantity of DNA present would provide a helpful tool during the investigation of contact traces, especially when these results are produced at the crime scene. Although DNA quantification methods are available, they are limited by being laboratory bound and time consuming (Jackson & Jackson, 2011). The development of an on-site DNA quantification method would be a valuable contribution to crime scene investigation, for it would provide guidance in the selection process preceding submission for DNA analysis (Mapes et al., 2015). Some progress has been made in developing on-site DNA profiling devices and Lab-on-Chip technology which are intertwined with on-site quantification methods, for some of these devices are able to quantify DNA
BOX 1 - Requirements for a presumptive DNA quantification method
The general requirements for the development of a presumptive on-site method are discussed, in order to maintain a structured overview and to provide transparency regarding which standards the method is measured by.
Ideally, a presumptive DNA quantification method would be non-destructive, allowing the same sample to be used again for further analysis. However, immunogenic technology is inevitably destructive to the material (Virkler & Lednev, 2009). Nevertheless this study focuses on finding targets for immunogenic techniques. Immunogenic techniques are based on the principle of antibodies adhering to antigens. An antigen is a substance that can evoke a immune response in an alien organism. Therefore an antigen can be various substances such as peptides, proteins and hormones. Consequently, antibodies can be produced by the immune system of the host organism when evoked by an antigen. This principle is applied to a lateral flow assay by using two sorts of antibodies that can bind to the analyte, the desired substance of detection. First the analyte binds to antibodies with a conjugated tag. Subsequently, the solution flows to the second type of antibodies which are immobilized in the form of a test line. The analytes bound to the conjugate bind to the antibodies on this test line and are thus detectable with the help of the tag (Posthuma-Trumpie et al., 2009). The sample that has been administered to the device cannot be reused, the remaining part however, containing solution that has not been in contact with antibodies, may still be usable for further analysis. (Box continues on page 3)
3 as well (Dawnay et al., 2014). However, thus far these devices cannot be applied at a crime scene, because currently none of the devices can perform all of the required steps for DNA analysis at the crime scene, because they are partly still dependent on non-portable equipment. Furthermore, most of them have only been tested with purified samples, instead of “dirty” samples from a crime scene. And finally, these devices cannot handle low quantity samples yet (Bruijns et al., 2016). Therefore, although on-site devices might become of great importance in the future, it will require some time to overcome all remaining obstacles. Hence, the exploration for a presumptive DNA quantification tool should be continued and extended, so that a straightforward tool might serve us in the meantime.
One of the techniques suggested to apply for presumptive DNA quantification is immunogenic technology (Van de Kerkhof, 2017), which is explained in Box 1. Immunogenic assays, such as the lateral flow immunoassay (LFI), have been applied at crime scenes before (Virkler & Lednev, 2009). Apart from the technique being rapid, easy to use and having low production costs (Van de Kerkhof, 2017), immunogenic technology is preferable because it can recognize distinctive analytes in a sample and the technique can be adjusted to recognize the analyte of your choice. This selection however, has consequences for the specificity and sensitivity of the test and is therefore
BOX 1 - Continued
Hence, destructivity does not necessarily pose a problem. Other characteristics underpinning the implementation of this specific method are the low costs and the previous use in presumptive identification tests (Turrina et al., 2008; Virkler & Lednev, 2009). Moreover, the alternatives include spectroscopic methods and even though these methods are non-destructive, they would not qualify on many other aspects, such as heaviness of equipment, time consumption and the need for DNA extraction (Van den Kerkhof, 2017), for portability, low time consumption and low number of steps are requirements as well. It cannot be dependent on heavy equipment, because it should be easily transportable for being able to enter any kind of crime scene. Additionally, limited time consumption is vital in practice, because securing evidence is done as swift as possible. Another obvious requirement is the low difficulty of the method, whereas non-specialists should be able to adequately use it at the crime scene.
A logical requirement as well, but more complicated to assess, is a high quality test performance. Immunogenic technology is a very sensitive method and can detect low amounts of antibodies, but the obstacles that could be encountered by the method should be anticipated. The main component of the method that establishes the accuracy of the test, will be the strength of the correlation between the target and the amount of DNA. Therefore, this is an important measure to assess early in the development of the method. Subsequently, the lower threshold of the test should be determined and if possible conformed to the threshold of low copy number (LCN) DNA profiling, because this technology is able to yield a profile from a very small amount of cell material (Gill, 2001). Contact traces can contain 0 to 1 ng DNA and the minimum amount of DNA required for LCN analysis is 6 pg (Kloosterman & Meulenbroek, 2008). By taking the same threshold for the presumptive testing, it is ensured that the presumptive result will confirm whether or not sufficient DNA is present for further analysis. Simultaneously, it should be taken into account during the development of the method that only human DNA is relevant for profiling. In the process of comparing targets and detection methods, human specificity is strongly preferred if possible. Finally, the test performance of the method should be appraised and circumstances that could lead to false positive results should be anticipated. However, this will be entrusted to further research, since this requires empirical experimentation.
4 crucial to the process of development. To find the most favorable targets and detection methods, the different possibilities should be scrutinized and assessed for their potential regarding an on-site DNA quantification tool. To provide an overview of the possibilities and future directions the following questions will be answered. What targets are potentially indicative of DNA quantity and what aspects of detection should be considered?
Exploration of various targets
Diverse approaches for detecting and quantifying DNA with an immunoassay technique can be suggested. To assess which approach would be best for quantification, a closer look should be taken at the molecule of interest; that is DNA. DNA, deoxyribonucleic acid, is built from nucleotides, containing a phosphate molecule and a sugar molecule with a base attached to it. Human DNA contains four different bases; adenine, guanine, cytosine and thymine. Adenine and guanine are purines, while cytosine and thymine are pyrimidines (Alberts et al., 2014). Meanwhile, DNA is not the only occupant of the nucleus, it has been accompanied by many proteins which are often interacting with the DNA. The proteins that bind with the DNA are, naturally, called DNA-binding proteins (DBPs) and can be divided into two groups, the specific and the non-specific DBPs. The specific DBPs only bind to DNA at a location with a specific sequence, for example transcription factors that bind to gene regulatory sequences in the DNA (Alberts et al., 2014). The binding of non-specific DBPs is not dependent on a particular sequence, but can theoretically bind anywhere. In practice, these proteins still have some preferences for location and binding is not completely left to random processes. They are distributed over the DNA molecule and generally provide structure to the DNA (Alberts et al., 2014). An example of non-specific DBPs are histones, which assist in the organized packaging of the DNA. Also non-histone chromosomal proteins are present in the nucleus with approximately the same function but other ways of functioning.
Table 1. An overview of the various possibilities that can be explored as a potential target for a DNA quantification tool.
Category Sub-category Group Target Examples
D N A -b in d in g pro te in s N o n -specifi
c Histones Canonical histones H1, H2A, H2B, H3, H4
Non-canonical histones H2A.X, H2A.Z, H3.3 Non-histone chromosomal proteins HMG proteins HMGA, HMGB, HMGN Sp ecific Transcription factors (TFs) General TFs Specific TFs TBP p53 Telomere binding proteins
Shelterin complex TRF1, TRF2, POT1
N u cleic acids D N A (do u b le strand ed) Bases Purines A, G Pyrimidines T, C
Sequence specific Human specific sequence -
RN A (sin gle strand ed) Not specified - -
5 It is expected that proteins that bind DNA both specifically and non-specifically can predict the amount of DNA present and therefore these proteins are object of investigation within this study. This search for a suitable target commences with a subdivision of the main groups of molecules that could provide a target for a quantification tool. These molecules and the group they appertain are displayed in table 1 and are scrutinized in the consecutive paragraphs.
Non-specific DNA-binding proteins Histones
The most abundant non-specific BDP within the cell’s nucleus are histones. Histones facilitate the formation of chromatin. Chromatin is the term for compact packaging of DNA and it is constructed with nucleosomes. Nucleosomes consist of DNA that is wrapped around an octamer of histone proteins. Multiple nucleosomes, connected by a small stretch of DNA and a linker histone form the organized structure of chromatin (Alberts et al., 2014). This octamer of histones consists of several kinds of histones, 2A, 2B, 3 and 4 and every variant occurs two times. Due to the binding of the histones, the DNA helix can wind around this octamer. The many interactions that occur between the histone and the DNA molecule allow the histones to bind to a variety of sequences on the DNA without much specificity. The nucleosome is completed by adding a linker histone H1 to the continuing stretch of DNA which connects the nucleosomes together (Alberts et al., 2014).
The main function of these canonical histones comprises providing an organized structure to the DNA whilst maintaining access to the DNA for gene transcription and replication. Fine tuning of this process is performed by modification of the acidic tails of the histones that protrude the nucleosome, as well as by incorporation of histone variants (Jin et al., 2005). The canonical histones that form the classic nucleosome can then be replaced by paralogues with distinctive acidic tails and perform additional functions, such as chromosome segregation, meiotic recombination and sex chromosome condensation (Talbert & Henikoff, 2010). Due to the their distinctive functions, many of these variations on the canonical histones appear at different locations and different moments throughout the cell cycle.
Histones being so closely associated with DNA suggests many opportunities for being a target of a DNA quantification tool. However, other characteristics of histones might oppose this potential. Histones are known to have a high evolutionary conservation in eukaryotes. As a consequence the detection of histones in a sample may be an indication for the presence of DNA, but this DNA can originate from any eukaryotic donor. This obstacle does not exclusively apply to histones, but will also concern other proteins and therefore may need to be solved by other approaches. However, more critical to the development of a quantification tool is that the quantity of histones may not be representative for the quantity of DNA as a consequence of the instability of concentration for different variants throughout the cell cycle. As it is not possible to determine the current stage cells in a sample are in, this will prevent an accurate estimation of DNA quantity.
Histones summarized
(+) Present in abundant concentrations in cell nuclei. (-) Presence only gives indication of eukaryotic donor.
6 HMG proteins
The high mobility group (HMG) proteins are the second most abundant protein assisting in chromatin formation after histones (Bianchi & Agresti, 2005). Multiple variants of this protein family are present within the nucleus, called HMGA, HMGB and HGMN. The added letters are in reference to the structural domains that are distinctive between the variants. A refers to the AT-hook present in the structure, while B refers to the boxes and N is assigned due to the nucleosome binding domain. Even though these variants have such different structures, their function is very much alike. Nevertheless, due to the dissimilar structures, the binding to the chromatin is established in different ways. Also, these non-specific DBPs might not be dependent on very specific DNA sequences, they do have preferences for the binding location. HMGA can bind to AT-rich sequences in the minor groove of a DNA helix where it can increase the binding of other proteins and thus influence gene regulation. Simultaneously, HMGA is believed to be involved in chromatin structure formation by binding to the nuclear matrix (Bianchi & Agresti, 2005). The HMGB proteins also bind to the minor groove of the DNA helix, independent of sequences. Due to the boxes within the HMGB protein, HMGB is able to bend the DNA helix quite sharply. The conformational change this entails, also affects the binding of other proteins. Because of these features, HMGB can eventually facilitate gene transcription. Additionally, HMGB also fulfills a role in maintenance of DNA and the repair of it. In contrast to HMGA and HMGB, HMGN proteins bind within the nucleosome instead of to the DNA helix and is able to unfold the chromatin, thus facilitating processes such as transcription, replication and DNA repair (Bianchi & Agresti, 2005). Not only the location of binding varies between these proteins, also the concentration of HMG proteins is differing. For example, the HMGA proteins are only present in high concentrations in embryonic cells according to Sgarra et al. (as cited in Bianchi &Agresti, 2005) and in adult cells HMGA only occurs in low concentrations, approximately one molecule per 1000 nucleosomes. HMGN and HMGB proteins appear respectively in concentrations of one molecule per 100 nucleosomes and one molecule per 10-15 nucleosomes (Catez & Hock, 2010). Furthermore, the distribution over the nucleus is not the same, whereas HMGB is uniformly dispersed over the nucleus, but the HMGN is preferably located at more active regions (Catez & Hock, 2010).
Although there seems to be sufficient ground for these proteins providing a target for a DNA quantification tool, some impediments may be present as well. Certain HMG proteins, for instance HMGB1, does not exclusively occur in the nucleus of the cell, but also fulfills extranuclear functions (Müller et al., 2004). This certainly gives rise to some concern, because the extranuclear concentrations of these proteins are therefore no longer necessarily associated with the intracellular concentrations which are expected to be in proportion to the DNA. Not all HMG proteins are yet known to have extranuclear functions, but caution is advised since extranuclear HMG levels might introduce aberrancies in the correlation with the amount of DNA. Moreover, HMG proteins have been found to be absent in cells that are no longer able to proliferate (Hock et al., 2007). This poses a problem when attempting to detect HMG proteins in microtraces containing epithelial cells, which are assumed to be terminally differentiated cells.
HMG proteins summarized
(+) Present in cell nuclei, HMGB in highest concentration.
(-) Extranuclear presence may prevent accurate representation of DNA quantity. (-) Absent in terminally differentiated cells, such as epithelial cells in microtraces.
7 Specific DNA-binding proteins
Transcription factors
Proteins that have been known for a long time to bind to DNA, are proteins with a key function in the transcription of genes; the transcription factors. To allow transcription of a gene, multiple proteins have to bind the DNA, among which the transcription factors. Transcription factors can be divided into two main groups. The first are the general transcription factors, which are required for the transcription of all eukaryotic genes. These proteins fulfill their function by binding to a specific DNA sequence, called the TATA box. The start of transcription is signaled by the TATA box and therefore it is present upstream of a gene. Around this TATA box, an initiation complex is formed, which ultimately allows RNA-polymerase to bind to the DNA and transcribe the gene (Alberts et al., 2014). The second group of transcription factors consists of sequence-specific DNA binding factors, which are only engaged in the transcription of specific genes, because every individual gene requires its own specific regulation. This regulation can be established through numerous ways of interacting with the general transcription factors or the promotor site of the gene (Kadonaga, 2004).
Difficulties that may arise when the concentration of specific transcription factors are measured, are determined by three main factors. The first comprises of the activity of the transcription factors. As gene expression is a highly regulated process, due to its responsibility for the maintenance of biological processes and thus for proper functioning of the cell, the regulation of the specific transcription factors is critical as well. Two mechanisms are at work to balance the activity of transcription factors in a cell. Both the synthesis of the transcription factors and the molecular activity of the transcription factors are controlled. The combination of these two mechanisms provides the system some flexibility to respond to the demand of the cell (Latchman, 1993), because the changes in activity can thus succeed each other rapidly. However, these fluctuationscan impede the use of transcription factors, because these fluctuations cannot be incorporated into the correlation between the transcription factor and the amount of DNA, but will rather distort it. Consequently, the amount of transcription factors cannot provide a reliable prediction regarding the amount of DNA. Furthermore, the regulation of transcription factors is highly dependent on cell type and the phase in which the cell is encountered. Both are hard to determine when encountered at a crime scene and thus it becomes difficult to construct a prediction for the amount of DNA.
Nevertheless, the large amount of genes in a human cell requires a variety of forms and shapes of specific transcription factors. This variety could allow some transcription factors to be more convenient by being stable over time and cell type. Furthermore, it might be possible to select a transcription factor that is specific for mammals, which narrows the scope of possible donors. However, despite elaborate research that has been performed, much has yet to be explored. The research into different transcription factors is disproportionate as a consequence of the medical focus to which especially the cancer research has contributed. As a result, the amount of knowledge about certain transcription factors varies strongly (Vaquerizas et al., 2009). Hence, it is necessary to focus on transcription factors with more information available.
A specific transcription factor that has received a lot of attention over the past few decades is p53. It received this attention on account of its importance in gene regulation and conducting the cell cycle (Millau et al., 2009). This protein stood out because it was fundamental to many pathways, including one for DNA repair, and was encountered in a multitude of cancer cells. However, despite the importance of this protein, it is only detected in low concentrations, as low as 100 nM according to Ma et al. (as cited at Bionumbers.org, 2018) and in addition the concentration is highly unstable
8 (Alberts et al., 2014). Therefore, there are no indications that this transcription factor would be an advisable target for a DNA quantification tool. An example of a general transcription factor is the TATA-binding protein (TBP). The main task of TBP is to separate the two DNA strands which allows RNA polymerase to bind (Alberts et al., 2014). Due to its universal function in human cells, it can be expected that it is widely prevalent and also perhaps more stable during the cell cycle, compared to specific transcription factors. These two examples illustrate that most of the difficulties apply to the specific transcription factors and may affect the general transcription factors to less extent. Therefore, general transcription factors as target for a DNA quantification tool are recommended before specific transcription factors.
Still, a considerable amount of transcription factors could be further examined. However, this research focusses on the distinct groups of targets that hold potential for a quantification tool and therefore this article will now proceed with inspecting another group of targets.
Telomere binding proteins
At every end of a chromosome a characteristic structure is found, called a telomere. They have a chromosome protecting function by preventing degeneration and fusing of chromosomes. A protein complex called the shelterin complex helps to successfully maintain this function and binds to the telomere ensuring the maintenance of the telomere (Xin et al., 2008). The end of telomeres are often left with a 3’ overhang, due to the mechanism of DNA replication. This asymmetry of DNA strands can be recognized as a double stranded break of the DNA, which will attract unwanted DNA repair proteins (De Lange, 2005). The binding of the shelterin complex remodels the DNA molecule and forms a loop (Wei & Price, 2003), thus preventing the attraction of undesired repair proteins.
The shelterin complex consists of six proteins of which three bind directly to the DNA. These proteins attach specifically to the sequence of TTAGGG which is abundant in human telomeres and are therefore called TTAGGG repeating factors (TRF). TRF1 and 2, together with the protecting of telomere protein 1 (POT1) bind to the DNA and are consequently bound by three other proteins that complete the shelterin complex (De Lange, 2005).
The specific and exclusive binding of the shelterin complex to the telomere, might suggest a strong association with the quantity of DNA. However, telomeres are known to shorten with every replication, which causes cells that have passed through numerous rounds of replication to have decreased telomere length. Nonetheless, the concentration of shelterin components has been found to be independent of telomere length (Takai et al., 2010). Consequently, the association between the shelterin complex and the amount of DNA is still expected to yield a high predictive value. The presence of shelterin proteins is appreciable, since TRF1 and 2 occur in high concentrations in the cell, varying between 30 to 200 thousand molecules per cell. The presence of POT1 as well, with 20 thousand copies per cell, is plentiful (Takai et al., 2010). Moreover, these proteins are abundant throughout the cell cycle (De Lange, 2005), which certainly offers an advantage over other proteins that have been discussed before. Furthermore, these proteins are promising, because they bind to
Transcription factors summarized
(+) Present in cell nuclei.
(+) Due to variety of sorts, it may be possible to select a mammal specific transcription factor. (-) Low predictive value due to fluctuations in concentration caused by activity control, cell cycle and cell type.
9 human specific telomere sequences, which might indicate a profit for developing a human specific test.
Nucleic acids DNA Staining
The possibility of targeting the DNA directly should also be considered. Ever since the existence of DNA was known, researchers have been seeking methods to visualize DNA. They have succeeded several times and thus there are currently many techniques available to use for this purpose. Aside from spectroscopic methods that are able to detect DNA, science uses multiple dyes to color DNA including intercalating dyes, such as ethidium bromide, minor groove binders, such as DAPI and a group of other nucleic acid stains of which acridine orange is one (ThermoFisher Scientific, 2018). However, these methods have never been used outside of a laboratory setting because this was never necessary. For this reason, most of the techniques available can stain DNA when present in a gel or during a PCR-reaction. Some of these stains subsequently require excitation under UV-light to optimally detect the staining and none of the above are generally available at a crime scene. Moreover, quite a few of these techniques include toxic substances, which preferably should not be handled by non-specialists (Huang & Fu, 2005).
A staining method that can stain DNA and does not require a gel, but merely a cellulose membrane is the method described by Merril and Pratt (1985). They used a silver stain to color the DNA and succeeded in visualizing small amounts of DNA up to 10 ng. However, for forensic purposes the sensitivity required would be a thousand times smaller still. This sensitivity of a few picograms was reached by Lomholt and Frederiksen (1986). To establish this sensitivity, however, the membrane had to be abandoned and again a gel was used. The exchange of the membrane for a gel was due to the regulation of the pH, because the gel had now to be washed in ethanolamine. Moreover, the time consumed to complete this procedure goes up to 3,5 hours. Thus, realization of a tool which incorporates this test would meet the challenges of controlling the pH in a cellulose membrane or developing a portable gel construction and reducing time consumption. More importantly, finding a solution to these problems would require a renewed search.
The problem of sufficient sensitivity is not limited to only a few of these staining methods, but in general a problem of directly targeting the DNA molecule. The signal of low quantities of DNA using a stain, will be weak nonetheless. This in contrast with the potential detection of an indicating molecule, because these molecules may be present in larger amounts. Furthermore, an immunogenic method allows an amplification of the signal and would therefore provide an advantage over merely staining the DNA molecules. Additionally, the lack of safety issues in an immunogenic method pleads in favor of this technique.
Telomere binding proteins summarized
(+) Present in abundant concentrations in cell nuclei. (+) Presence stable throughout cell cycle.
(+) Potential for human specific test. (-) None encountered.
10 DNA specific antibodies
Subsequently, the feasibility of targeting DNA with antibodies should be explored, because this provides an opportunity as well (Jacob & Srikumar, 1985). However, little is known about this technique since more targeted and thus efficient methods to detect DNA quickly replaced this immunogenic approach. Meanwhile, many antibodies only adhered to denatured DNA which disqualifies them to apply in forensic practice. Furthermore, these antibodies are presumably not human specific, because they interact with the oligodeoxynucleotide structure of the molecule, which is a common feature of DNA is various organisms (Alberts et al., 2014). And since the antibodies were apparently barely applied, the degree of sensitivity is unknown. Another option would be using antibodies against specific DNA sequences. Theoretically, this could also be a possible method to explore, but in practice it has a couple of downsides. The sequence that is being targeted should be specifically human, for obvious reasons. Preferably, it should also be present at multiple locations in the genome, because it would sent a stronger signal when it was emitted from multiple locations.
However, due to medical interest, tests have been developed to detect nucleic acids with the use of antibodies. Although these techniques are accessible, it differs whether they target single stranded (Corstjens et al., 2003) or double stranded DNA, whereas the latter is required for the current tool. Even LFA devices for the detection of (double stranded) nucleic acids have been designed and produced, due to a similar demand in, again, the medical field and other fields such as food industry and environmental monitoring (Posthuma-Trumpie et al., 2009). However, the applicability of these devices is not a guaranteed success for the forensic field. The “nucleic acid lateral flow immunoassay” (NALFIA) allows detection of double stranded DNA, but requires tagged amplicons for recognition by an tag-specific antibody (Posthuma-Trumpie et al., 2009). This implies pretreatment of the sample and the necessity of a PCR reaction. The “disposable nucleic acid biosensor” (DNAB) is another DNA detecting device, which is very sensitive (Mao et al., 2009) and able to perform without PCR amplification. However, the sequence that is recognized by the test still has to reach a minimum amount to be detected. Therefore, by omitting the PCR, the test demands an increased amount of DNA of a thousand copies (Zhang et al., 2004). This number exceeds the minimum amount of DNA required for LCN analysis to a large extent, because this consists of only 6 pg (See Box 1). Thus, it has to be concluded that none of the existent methods are ready to be applied in forensic practice.
DNA specific antibodies summarized
(+) Directly targeting DNA with antibodies. (+) Potential for human specific test.
(-) Requires PCR or requires a too high detection threshold.
DNA staining summarized
(+) Directly measuring the DNA concentration. (-) Regulating the right sensitivity is a challenge. (-) Weak signal for low quantities.
11 RNA targeting
A molecule strongly associated with DNA is RNA. RNA is the messenger molecule that provides the main communication between the nucleus and the extranuclear space, although this arises through many variations of the molecule. Due to the process of transcription, the information preserved within the DNA can be used in the sustenance of the cell (Alberts et al., 2014). Because of this close cooperation between the two, RNA molecules should be mentioned as contender. Many methods are available to detect RNA, but these methods will not be listed here, because it is debatable whether the use of these methods would offer any advantage over the alternative techniques with other targets that have been discussed in this article.
The first objection is the extranuclear function of RNA, which indicates that it would be possible to detect RNA without finding other material originating from the cell nucleus. Hence, the predictive value is considered fairly low. Another obstacle includes the instability of the molecule. As an important link in controlling the production of proteins, the ease of degradation of the molecule forms a substantial role in the fine tuning of this process (Belasco & Brawerman, 2012). This property of RNA bears two consequences. Firstly, it implies large fluctuations in concentrations within a living cell, which is similar to one of the difficulties that was suggested for transcription factors, because this is expected to strongly reduce the predictive value for the quantity of DNA. Secondly, the rapid degradation of RNA molecules may pose a problem when they have to be recovered from a crime scene (Bauer, 2007), for the recovery of traces rarely takes place immediately after deposition of the trace. Moreover, the time that elapses after deposition will vary strongly between cases and more importantly, the time that has elapsed will be unknown to the investigators. This knowledge should deter researchers for now from investing in this direction of research.
Detection
(Semi-)quantitative testing
When it is decided to use antibodies for quantification, a lateral flow device can be used. One of the major tasks of lateral flow devices in many different areas, is point of care (POC) testing, which aims at providing swift answers about the presence of a specific analyte (Sajid et al., 2015). For this reason, however, the design of the LFA mostly allows qualitative testing of a sample, indicating the presence or absence of the analyte. Information about the quantity of DNA would be desirable for crime scene investigators, because this information can assist in anticipating the quality of the definitive test results. Therefore this information plays a major role in the prioritization of sample submission to the laboratory (Service Level Agreement, 2017). Nevertheless, the test result of a standard LFA presents as a visible test line and the density of this test line expresses the concentration of the analyte in the sample. Therefore, visual inspection can contribute some information about the quantity of the analyte, but only to the extent that is visible to the eye. Also, the interpretation of this visual inspection may become difficult for non-specialists. Although, accordingly, this indication would be informative at a crime scene, additional quantitative
RNA targeting summarized
(-) Low predictive value for DNA quantity due to extranuclear function and instability of the molecule.
12 information would be preferable. Consequently, the standard device is not considered satisfactory and extended alternatives have to be explored.
Currently, two alternatives are available that satisfy the need for quantitative information; a semi-quantitative approach and a quantitative method. The semi-quantitative approach is based on the addition of multiple test lines with increasing concentrations, thus allowing to read out the approximate concentration. The quantifying techniques use portable spectroscopic devices that can measure the concentration of the signal on the test line of the LFA. However, techniques that provide a higher level of detail are apt to need more time for analysis and cost more money as well (Posthuma-Trumpie et al., 2009). For this reason, it can be discussed that a quantifying technique should show a clear advantage over a semi-quantitative approach regarding the test performance, for example the sensitivity that it offers. However, not much literature gives insight in this particular aspect of quantitative testing with an LFA.
Another extended LFA device that has been developed is a multianalyte LFA, which is able to detect multiple analytes in one test (Fenton et al., 2008). In the medical field this offers the advantage of testing for multiple diseases at once, but for DNA quantity testing this advantage could instead be utilized to detect multiple targets that are able to predict the DNA quantity and together provide a more precise estimation. However, quantitative information remains essential and thus the incorporation of multianalyte testing should be subordinate to finding the right method for quantitative testing. When this method has been selected, the exact improvement of precision that can be offered by multianalyte testing can be determined and taken into consideration as added value to the tool.
Ideally, the target and the detection method should be adaptable to both a semi-quantitative approach and a quantifying technique. However, selection may be inevitable and since the main requisite of this presumptive test includes determining the minimum amount of DNA present, the semi-quantitative approach can be regarded as adequate. The additional information that can be provided by a quantifying technique may not be worth the additional amount of money, especially when the technique also absorbs additional costly time at the crime scene.
Test performance
During the development of a presumptive DNA quantification tool, the first concern should be the performance of the test, since the tool should provide added value during a crime scene investigation. However, to achieve the optimal design for this tool, many parameters have to be considered and fine-tuned. Only a rough outline of the main features of an LFA and their options will be discussed.
Many factors will affect the sensitivity of the test, a precise calibration of a suitable tag, a convenient target and the right detection method is crucial to obtaining sufficient sensitivity (Posthuma-Trumpie et al.,2009). Meanwhile, a suitable tag can influence the sensitivity of the test, but is also responsible for the visibility of the result, which should also be a priority in the designing process. The selection of a detection method does not only affect sensitivity of the test, but alters the specificity of the test as well. Therefore, many decisions regarding the design of the LFA have yet
(Semi-)quantitative testing summarized
The semi-quantitative approach and the sandwich format are advised, since they provide sufficient information and are most convenient for non-specialists.
13 to be made and are largely reliant on the choice of target. The only recommendation that will be made regarding the design of the LFA concerns the test format. Two options are available for an antibody based LFA, a sandwich format and a competitive format. However, the competitive format has the adverse property of producing a seemingly ambiguous result. The absence of the analyte will color the test line, whereas the presence of the analyte will leave the test line blanc. This might lead to confusion among non-specialists (Posthuma-Trumpie et al., 2009) and therefore the sandwich format is proposed for the design of this tool.
Table 2. Summarization of the findings concerning the possible targets and detection methods.
Targets
Pros and cons
Histones (+) Present in abundant concentrations in cell nuclei. (-) Presence only gives indication of eukaryotic donor.
(-) Presence of histone variants not stable throughout cell cycle. HMG proteins (+) Present in cell nuclei, HMGB in highest concentration.
(-) Extranuclear presence may prevent accurate representation of DNA quantity.
(-) Absent in terminally differentiated cells, such as epithelial cells in microtraces.
Transcription factors
(+) Present in cell nuclei.
(+) Due to variety of sorts, it may be possible to select a mammal specific transcription factor.
(-) Low predictive value due to fluctuations in concentration caused by activity control, cell cycle and cell type.
Telomere binding proteins
(+) Present in abundant concentrations in cell nuclei. (+) Presence stable throughout cell cycle.
(+) Potential for human specific test. (-) None encountered.
DNA Staining
(+) Directly measuring the DNA concentration. (-) Regulating the right sensitivity is a challenge. (-) Weak signal for low quantities.
(-) Safety issues present, less suitable for non-specialist personnel. Antibodies
(+) Directly targeting DNA with antibodies. (+) Potential for human specific test.
(-) Requires PCR or requires a too high detection threshold.
RNA (-) Low predictive value for DNA quantity due to extranuclear function and instability of the molecule.
(-) Rapid degradation indicates unreliable source of information.
Detection
Recommendations
Qualitative vs.quantitative
The semi-quantitative approach and the sandwich format are advised, since they provide sufficient information and are most convenient for non-specialists.
Test-performance
14 Conclusions and recommendations
In this article, several possibilities have been identified for the development of an presumptive DNA quantification tool. The recommendation include the design of an LFA which targets one of the proteins of the shelterin complex. Although the search for a target was not exhaustive and more suitable targets can still be found, the proteins from the shelterin complex look promising. Nevertheless, extended research should be conducted to gather additional knowledge about the shelterin complex and methods of detection, such as available antibodies. Further recommendations regarding the detection method are still somewhat general, except for the advice to adopt the semi-quantitative approach and use a sandwich format, since the information it provides is sufficient and it is most convenient for non-specialist personnel. The findings are summarized and displayed in table 2. It can thus be concluded that there are multiple opportunities to develop a presumptive DNA quantification tool and realization of such a tool is within reach.
Acknowledgements
I would like to thank Annemieke van Dam for her support and guidance throughout the writing process and the feedback she provided me with.
References
Alberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K. & Walter, P. (2014) Molecular Biology of the Cell, sixth ed. Garland Science/Taylor & Francis Group, LLC; New York, 2014.
Ball, G., Dawnay, N., Stafford-Allen, B., Panasiuk, M., Rendell, P., Blackman, S., et al. (2015). Concordance study between the ParaDNA® Intelligence Test, a rapid DNA profiling assay, and a conventional STR typing kit (AmpFlSTR® SGM Plus®). Forensic Science International: Genetics, 16, 48-51.
Bauer, M. (2007). RNA in forensic science. Forensic Science International: Genetics, 1(1), 69-74. Belasco, J. G., & Brawerman, G. (Eds.). (2012). Control of messenger RNA stability. Elsevier.
Bianchi, M. ., and Agresti, A. (2005). HMG proteins: dynamic players in gene regulation and differentiation. Current opinion in genetics & development 15.5: 496-506.
Bionumbers.org, http://book.bionumbers.org/what-are-the-copy-numbers-of-transcription-factors/ (accessed on 25-01-2018).
Bruijns, B., van Asten, A., Tiggelaar, R., & Gardeniers, H. (2016). Microfluidic devices for forensic DNA analysis: A review. Biosensors, 6(3), 41.
Catez, F., & Hock, R. (2010). Binding and interplay of HMG proteins on chromatin: lessons from live cell imaging. Biochimica et Biophysica Acta (BBA)-Gene Regulatory Mechanisms, 1799(1), 15-27.
Corstjens, P. L., Zuiderwijk, M., Nilsson, M., Feindt, H., Niedbala, R. S., & Tanke, H. J. (2003). Lateral-flow and up-converting phosphor reporters to detect single-stranded nucleic acids in a sandwich-hybridization assay. Analytical biochemistry, 312(2), 191-200.
Dawnay, N., Stafford-Allen, B., Moore, D., Blackman, S., Rendell, P., Hanson, E. K., et al. (2014). Developmental Validation of the ParaDNA® Screening System-A presumptive test for the
15 detection of DNA on forensic evidence items. Forensic Science International: Genetics, 11, 73-79.
De Lange, T. (2005). Shelterin: the protein complex that shapes and safeguards human telomeres. Genes & development, 19(18), 2100-2110.
Fenton, E. M., Mascarenas, M. R., López, G. P., & Sibbett, S. S. (2008). Multiplex lateral-flow test strips fabricated by two-dimensional shaping. ACS applied materials & interfaces, 1(1), 124-129.
Gill, P. (2001). Application of low copy number DNA profiling. Croatian medical journal, 42(3), 229-232.
Hock, R., Furusawa, T., Ueda, T., & Bustin, M. (2007). HMG chromosomal proteins in development and disease. Trends in cell biology, 17(2), 72-79.
Huang, Q., & Fu, W. L. (2005). Comparative analysis of the DNA staining efficiencies of different fluorescent dyes in preparative agarose gel electrophoresis. Clinical Chemical Laboratory Medicine, 43(8), 841-842.
Jackson, A. R., & Jackson, J. M. (2011). Forensic science. Pearson Education.
Jacob, T. M., & Srikumar, C. (1985). Nucleic acid reactive antibodies—Specificities and applications. Journal of Biosciences, 7(1), 61-73.
Jin, J., Cai, Y., Li, B., Conaway, R. C., Workman, J. L., Conaway, J. W., & Kusch, T. (2005). In and out: histone variant exchange in chromatin. Trends in biochemical sciences, 30(12), 680-687.
Kadonaga, J. T. (2004). Regulation of RNA polymerase II transcription by sequence-specific DNA binding factors. Cell, 116(2), 247-257.
Kloosterman, A. D., & Meulenbroek, A. J. (2008). DNA-onderzoek van minimale biologische sporen; gevoelige problematiek. Expertise en Recht, 4, 108-120.
Latchman, D. S. (1993). Transcription factors: an overview. International journal of experimental pathology, 74(5), 417.
Lomholt, B., & Frederiksen, S. (1987). Detection of a few picograms of DNA on polyacrylamide gels by silver staining. Analytical biochemistry, 164(1), 146-149.
Mao, X., Ma, Y., Zhang, A., Zhang, L., Zeng, L., & Liu, G. (2009). Disposable nucleic acid biosensors based on gold nanoparticle probes and lateral flow strip. Analytical chemistry, 81(4), 1660-1668.
Mapes, A. A., Kloosterman, A. D., & Poot, C. J. (2015). DNA in the criminal justice system: the DNA success story in perspective. Journal of forensic sciences, 60(4), 851-856.
Merril, C. R., & Pratt, M. E. (1986). A silver stain for the rapid quantitative detection of proteins or nucleic acids on membranes or thin layer plates. Analytical biochemistry, 156(1), 96-110. Meulenbroek, A. J. (2006). De essenties van forensisch DNA-onderzoek. Nederlands Forensisch
Instituut, information map.
Millau, J. F., Bastien, N., & Drouin, R. (2009). P53 transcriptional activities: a general overview and some thoughts. Mutation Research/Reviews in Mutation Research, 681(2), 118-133.
Müller, S., Ronfani, L., & Bianchi, M. E. (2004). Regulated expression and subcellular localization of HMGB1, a chromatin protein with a cytokine function. Journal of internal medicine, 255(3), 332-343.
Posthuma-Trumpie, G. A., Korf, J., & van Amerongen, A. (2009). Lateral flow (immuno) assay: its strengths, weaknesses, opportunities and threats. A literature survey. Analytical and bioanalytical chemistry, 393(2), 569-582.
16 Sajid, M., Kawde, A. N., & Daud, M. (2015). Designs, formats and applications of lateral flow assay: A
literature review. Journal of Saudi Chemical Society, 19(6), 689-705.
Service Level Agreement (2017). A new service level agreement between the government and the NFI has been drafted, but is currently not retrievable.
Takai, K. K., Hooper, S., Blackwood, S., Gandhi, R., & de Lange, T. (2010). In vivo stoichiometry of shelterin components. Journal of Biological Chemistry, 285(2), 1457-1467.
Talbert, P. B., & Henikoff, S. (2010). Histone variants—ancient wrap artists of the epigenome. Nature reviews Molecular cell biology, 11(4), 264-275.
ThermoFisher Scientific, https://www.thermofisher.com/nl/en/home/references/molecular-probes-the-handbook/nucleic-acid-detection-and-genomics-technology/nucleic-acid-stains.html (accessed on 22-01-2018).
Turrina, S., Filippini, G., Atzei, R., Zaglia, E., & De Leo, D. (2008). Validation studies of rapid stain identification-blood (RSID-blood) kit in forensic caseworks. Forensic Science International: Genetics Supplement Series, 1(1), 74-75.
Vaquerizas, J. M., Kummerfeld, S. K., Teichmann, S. A., & Luscombe, N. M. (2009). A census of human transcription factors: function, expression and evolution. Nature Reviews Genetics, 10(4), 252-263.
Virkler, K., & Lednev, I. K. (2009). Analysis of body fluids for forensic purposes: from laboratory testing to non-destructive rapid confirmatory identification at a crime scene. Forensic Science International, 188(1), 1-17.
Wei, C., & Price, M. (2003). Protecting the terminus: t-loops and telomere end-binding proteins. Cellular and molecular life sciences, 60(11), 2283-2294.
Xin, H., Liu, D., & Songyang, Z. (2008). The telosome/shelterin complex and its functions. Genome biology, 9(9), 232.
Zhang, Y., Pothukuchy, A., Shin, W., Kim, Y., & Heller, A. (2004). Detection of 103 copies of DNA by an
electrochemical enzyme-amplified sandwich assay with ambient O2 as the
17
Appendix I – Search strategy
Various search strategies were applied during the process of literature thesis writing. The first references were provided by the article of Nina van de Kerkhof, who performed similar research last year, but chose to focus on one particular protein. Therefore, especially the references regarding the forensic relevance were useful to me.
To obtain knowledge about various targets that could be used, I searched terms like “presumptive DNA quantification method” or “DNA immunoassay antibodies”. However, this strategy did not produce useful results and after again consulting the article of Nina van de Kerkhof I read the articles she used for the description of the methods and started looking for articles about “Lateral flow (immune)assay” and “Lateral flow devices”, sometimes combined with “forensic” or “crime scene” which generated some relevant results. Furthermore, I decided to approach the targets from a different direction and just start at the beginning with describing them and gradually dig deeper, which was also an advice from my supervisor. One of the tips she gave me was to use the ‘cited by’ option in Google Scholar, which brought me to articles that cited the current article and this was occasionally very helpful.
At the times I got stuck, I tried to go back to the basics by just using regular google, searching the same term in a non-scientific environment. This regularly provided me with a new angle of incidence or sometimes just referred me to a Wikipedia page where I used the scientific references as a fresh start. Another approach that brought me back to the basics was by using ‘The Cell’, an educational book by Alberts et al., which provided me with the fundamental information about the cell and its occupants.
To find the numbers and concentrations of multiple molecules I wanted to describe was a bit of a challenge, because searching for “concentration HMG proteins”, for example, did not provide many useful results. The bionumbers website1 could sometimes refer me to an useful article, but much more often it couldn’t. I attempted to search for “healthy versus unhealthy” with the addition of a target on advice of my supervisor which generated some results, nevertheless not all the targets could be provided with numbers and concentrations.
Even so, these strategies, both known and previously unknown to me, and alternating them helped me to collect nearly all the information I needed and I certainly learned a lot from it.
