Protein oxidation biomarkers to reveal
the age of body fluids
Jolien M. Nienkemper 10458131
Date: 31-01-2018
Institute: University of Amsterdam Master Programme: Forensic Science Supervisor: Dr. Annemieke van Dam Co-Assessor: Prof. Dr. Ate Kloosterman
List of contents:
Introduction 2
1. What makes a good biomarker? 4
1.1 Biomarkers 4
2. The process of protein oxidation 5
2.1 Proteomics 5 2.2 Protein Oxidation 6 3. Oxidation products 7 3.1 Aliphatic residues 8 3.2 Aromatic residues 9 3.3 Sulfur-containing residues 10 4. Mass Spectrometry 12
4.1 Mass Spectrometry and biomarkers 12
4.2 From MS to immunoassays 14
Discussion 15
Conclusion 18
References 19
Appendix I: Table 1 23
Introduction
With current techniques DNA can often be extracted from body fluids found at a crime scene, making identification possible. The next step is to use the information embedded in the body fluids to determine the time passed since deposition. Such information can be used to provide intelligence on the order of events of a crime, which can help to link a piece of evidence to the crime. For example, a fingermark found at a crime scene can be linked to a suspect with the help of a database. It is possible however, that the suspect claims that he had been at the location of the crime scene, but one week prior to the date of the crime. If no other evidence can be found against the suspect, he cannot be detained. In such a situation, it would have been helpful if the forensic examiners would have been able to determine at what time the fingermark was deposited. Currently, different study groups are looking into new ways to apply varying analytical techniques on certain types of body fluids to estimate the age of that particular body fluid. Most research in this area is still in its early stages however, and needs further investigation.
Almost all body fluids contain proteins specific to that fluid, which can be used for identification of the biological material (Juusola & Ballantyne, 2003). Moreover, such proteins can function as biomarkers for age estimation. Over time proteins degrade into different products (Stadtman & Levine, 2000) which can be measured and used to determine the time since deposition. For example, it is known that the major component of blood, hemoglobin, converts into metHemoglobin once oxygen saturated. Inside of the body metHemoglobin is reduced back into hemoglobin, but outside of the body all hemoglobin will be oxygen saturated and change into hemichrome (Bremmer et al., 2011). Bremmer et al. (2011) studied the ratio of these three hemoglobin derivatives in dried blood stains with the use of diffuse reflectance spectroscopy. They found that each measurement, performed between 0 to 60 days after deposition, showed a unique combination of the three derivatives. With the help of this ratio, the age of a bloodstain could be estimated with, at most, a margin of uncertainty of 14 days. Bauer and colleagues (2003) took a different approach by looking into the degradation levels of RNA in blood, using RT-PCR to quantify the amounts of mRNA. Their results showed a significant correlation between the rate of RNA degradation and storage time. They do emphasize however, that storage conditions need to be known for
Abstract
Knowing when a body fluid stain was left at a crime scene, can provide meaningful information on the order of events of a crime. A relatively recent development in the forensic field is to study the degradation pattern of proteins present in a body fluid, to determine the time of deposition. For this method to be successful, it is necessary to establish biomarkers which are capable of giving an indication of the age of a crime scene stain. The purpose of this article is to give an overview of the most common protein oxidation products on amino acid level, and to indicate which of these products have the potential of being used as a biomarker. It is discussed whether the biomarkers can be measured using Mass Spectrometry, and subsequently whether they can be detected through an immuno-based assay. For future research, it is advised to focus on the carbonyl groups, methionine sulfoxide, methionine sulfone, cysteines oxyacids, dityrosine and the different kynurenines for a more thorough investigation into their potential as a biomarker.
reliable age estimation, since, for example, heat and light could influence RNA fragmentation. p
A similar approach was attempted for saliva samples, however, higher variability in overall protein concentrations within and between individuals and a high protein degradation rate made it hard to find a protein marker which was stable over time (Crosley et al., 2009; Ackerman et al., 2010). Saliva in general is found to be a very dynamic proteome, continuously being supplied with new proteins which are subsequently being removed by swallowing. Not only does is contain a great variety of proteins, it is also a very unstable fluid, susceptible to microorganisms and proteases (Helmerhorst & Oppenheim, 2007; Esser et al., 2008). In saliva, protein degradation starts from the moment a sample is deposited and is usually very rapid, making these proteins inadequate for long term age determination. However, a selection of degradation products has been found to increase in abundance over time and are therefore thought to be stable breakdown products of larger proteins, making them suitable biomarkers (Esser et al., 2008). Yet precautions should be taken towards the proteases in saliva and how they interact with these proteins.
Seminal plasm proteins also show rapid degradation, but when stored under optimum conditions, relatively stable biochemical parameters can be found (Jimenez-Verdejo et al., 1994). An older research of Jimenez-(Jimenez-Verdejo and colleagues (1994) studied the behaviour of a selection of semen specific enzymes and detected that, when all parameters were combined, the age of a semen stain could be calculated with a certain degree of precision. Nevertheless, none of the parameters was capable of giving a reliable age prediction on their own. A more recent study of Szykula (2016) found ‘new’ biomarkers in semen by analysing the length of proteins in aging stains with the help of mass spectrometry. Results showed dermcidin and semenogelin-2 to have the most significant change in length, making them interesting targets for validation studies.
A promising study of van Dam et al. (2014) showed that with the help of fluorescence spectroscopy the age of the 55% of the male fingermarks could be determined. The method used is based on the principle of the oxidation of the proteins and lipids present in a fingermark, with tryptophan-containing proteins (Tryp) being the main contributor. Unsaturated lipids oxidize when exposed to air, which results in reactive oxidation products (LipOx). These products react with proteins and form fluorescent oxidation products (FOX) (Eq. (1)).
[Eq. (1)]: LipOx + Tryp → FOX
During the oxidation reaction tryptophan is degraded, causing the fluorescence intensity generated by tryptophan to decrease. FOX is being produced as a result of the reaction, which leads to an increase of FOX-induced fluorescence. From this fluorescence spectrum, a Tryp/FOX-ratio can be extracted, which can be used to determine the time of deposition. A downside to this method is the fact that fluorescence spectroscopy is not sensitive enough to measure the fluorescent reaction products of, for example, female donors, because of the lower excretion of skin components by women. Mass spectrometry is a technique which is far more sensitive and is capable of quantifying very minute amounts of proteins. This is particularly useful since there is still no method available to amplify proteins, as they do with PCR for DNA (Kussman et al., 2009). Since crime scene stains can have a low volume, it is important that the technique used to analyse these samples is sensitive enough to measure such small quantities.
This review will focus on the different forms of protein oxidation modifications and will try to answer the question ‘which biomarkers seem most promising in revealing the age of body fluids?’.
Sub questions in need of an answer are:
- What makes a good biomarker?
- Which protein oxidation products are there?
- Can MS be used to demonstrate the presence of oxidative modifications? - How do these biomarkers translate to practice?
Proteins are chosen as the subject of this study since they are known to be the major component of almost every biological system, regardless of whether one considers tissue level, cellular level or body fluid level (Davies, 2005). Also, protein oxidation products are generally thought to be more stable compared to lipid oxidation products m. Moreover, proteins are found to be the main contributor to oxidation reactions with free radicals. Such knowledge, together with the finding that proteins have low turnover rates, and are therefore likely to accumulate higher amounts of oxidation products, makes that proteins are thought to be a suitable candidate for oxidation biomarker research (Davies et al., 1999).
1. What makes a good biomarker?
1.1 Biomarkers
Before answering the main research question of ‘which biomarkers seem most promising in revealing the age of body fluids?’ it is discussed here what exactly a biomarker is, and what makes a good biomarker.
Any biological molecule found in an organism that can be measured and quantified is a potential biomarker. In the medical field a biomarker is often a molecule that signals the presence of abnormal processes or diseases, but in the forensic context a biomarker can be used to indicate fairly common processes. One such example is using hemoglobin derivatives as biomarkers to measure the level of degradation of blood proteins. The body fluid from which biomarkers are aimed to be extracted should meet certain criteria. First of all, the fluid should be easily collectable. In forensic casework, this should be less of a problem since the body fluid stains that are of interest, are already excreted from the body when found at a crime scene, in contrast to the medical field, where the fluid still needs to be extracted from the patient. The difficulty lies in the correct sampling method, which should ensure the collection of all relevant biological material present in a body fluid. The fluid should also remain stable when being processed; during sample collection, transportation, storage and preparation for analysis (Good et al., 2007). Preferably, at the time of analysis the composition of the body fluid should be equal to that at the moment of collection. Therefore, storing the sample under the right conditions is of utmost importance, since temperature, humidity, light and pH are found to have an influence on the oxidation rate of proteins (Wellington, 2017). Furthermore, the sampling procedure needs to be reproducible. For example, blood needs to go through a few preparation steps before it can be analysed. These modifications cause the samples collected from one person to be highly variable, which results in unreproducible sampling (Good et al., 2007). And lastly, the analysis of a sample needs to render stable results; each time a certain fluid is collected, it should contain the biomarker of interest. When at a crime scene however, investigators do
not have the luxury of selecting a particular type of body fluid; they have to work with what is found at the scene. Therefore, this study will focus on reviewing biomarkers which may occur in every type of body fluid, and do not necessarily need to be specific for any particular cell type. Two of the most important criteria a marker needs to meet are that the biomarker shows a stable degradation pattern over a longer period of time, and that the products of this degradation can be measured.
Once a preliminary study results in a particular chemical that is determined to be a potential biomarker, it should be ensured that this biomarker is well studied in the population, and preferably have low variability within and between people. A publicly accessible proteomics database could be of help in such a process. After a specific biomarker has been found, the correlation between the concentration of the biomarker and the research objective, in this case time of deposition, should be investigated. To determine whether the chosen biomarker is suited for usage in the field, its specificity, reproducibility, sensitivity, selectivity and accuracy should be assessed with the use of assays (Ngamchuea, 2017). Once a biomarker is found to meet all the set criteria, it is argued by Good and colleagues (2007) it can then only be applied for its intended use, unless proven otherwise. For example, a biomarker developed to screen for a particular disease, might not be able to monitor the therapeutic effects.
The following chapters will elaborate on the different forms of oxidation and its products, how these modifications can be measured with mass spectrometry and how these biomarkers can eventually be put into practice. In the concluding remarks, recommendations shall be made regarding the most promising biomarkers in revealing the age of body fluids.
2. The process of protein oxidation
2.1 Proteomics
To be able to find a proper biomarker, some knowledge of proteins as a whole is required. The study of proteomics is a helpful tool for a better understanding of the functioning of proteins. Proteomics is the study of the proteome, which is the extensive collection of expressed proteins in a biological system. The Human Genome Project, which initiated the sequencing of the entire human genome, gave a great boost to the study of proteomics (Domon & Broder, 2003). Gained knowledge on the nucleotide sequence of DNA and how this sequence interacts with amino acids to produce proteins, gave insight into the proteins structural diversity. The thousands of genes embedded in the genome were found to encode for many more proteins, through, amongst others, alternative splicing. All of these proteins can subsequently undergo post-translational modifications, degradation, or form large complexes, resulting in an even greater variety (Kussman et al., 2006; Altelaar et al., 2013). Large databases are being built with the use of proteomics, which contain the amino acid sequences of proteins. The structure, function in an organism and how proteins interact with other chemicals is also being studied with proteomics. Understanding the amino acid sequence and structure of a protein, makes it possible to predict which type of modifications might occur and the corresponding change in mass. This information can be very useful when interpreting the mass-to-charge spectra obtained by Mass Spectrometry (MS) and selecting suitable biomarkers. Proteomics can also be used to find antibodies for the
biomarkers identified with MS. Therefore, proteomics is thought to be a key element in the search for biomarkers suited for age determination of a biological fluid.
2.2 Protein Oxidation
The breakdown or denaturing of the structure of a protein leads to inactivation of that protein, since the biological activity of a macromolecule is highly dependent on its peptide structure (Jimenez-Verdejo, 1994). The longer a protein is exposed to degrading substances, the more profound the degradation will be. Therefore, time is a very important determinant in the forensic analysis of biological samples. One degradation mechanism all macromolecules are faced with is oxidation (Chang et al., 2000). The process of oxidation can occur in many different ways, one of which is through a reaction with an oxidant. An oxidant is a molecule which is often very electronegative and eager to take electrons from other molecules. Without the presence of oxidizable compounds, oxidants tend to be rather stable substances. In contrast, free radicals are profoundly unstable and therefore highly reactive. A free radical is an atom, ion or molecule which possesses one unpaired electron, which can be generated by a wide variety of different processes. Free radicals can interact with other molecules by donating their free electron, which can, in turn, result in new radicals. Whenever a free radical comes into contact with an oxygen molecule, reactive oxygen species (ROS) are formed. One such ROS is the hydroxyl radical (OH·), which is the most reactive species of ‘activated oxygen’ both in vivo and in vitro and one of the causes of protein oxidation (Finaud et al., 2006). Hydroxyl radicals are a product of metal-catalysed oxidation; hydrogen peroxide (H2O2), which is derived from dioxygen, is converted into hydroxyl radicals in the presence of a metal, for instance iron (Fe) or copper (Cu) (Eq. 2).
[Eq. (2)]: H2O2 + Fe2+/Cu+ → OH⋅ + OH− + Fe3+/Cu+
Practically all amino acid side chains in a protein can be oxidized by hydroxyl radicals. However, when iron or copper concentrations are low, such metal-catalysed oxidation (MCO) reactions will be limited to the amino acid residues near the metal binding sites of the protein (Stadtman & Levine, 2000). The oxidation of sulfur-containing amino acids can be reversed with the use of enzymes, but the oxidation of other amino acids is found to be irreversible. Proteins which contain oxidized amino acids which are non-reducible, are targeted for degradation by proteasomes, to clear the cell from the inactive proteins.
Yet, there are also oxidized amino acids which have newly formed carbonyl groups (see Appendix I: Table 1), which cannot go through this process, which results in an accumulation of proteins containing these oxidized amino acids. Such accumulation is called oxidative stress. This leads to a reduced protein turnover, a decline in genetic transcription and can eventually result in cell death (Finaud et al., 2006) (Image 1).
From the aforementioned it may be deduced oxidation only has a negative influence on the biological system. In certain processes however, such as the assembly of adenosine triphosphate (ATP), oxidation is an essential step in the production cycle. Oxidation becomes harmful when there are too many free radicals, relative to the number of antioxidants. Antioxidants neutralize free radicals and make them less reactive or even unreactive (Stadtman & Levine, 2008).
Outside of the body, cells lose part of their antioxidant defence system. Moreover, the enzymes responsible for reduction of a protein are depleted as a result of excretion of the body fluid, since no knew enzymes are being produced. This results in an accumulation of reactive oxygens and an increase of the protein oxidation rate (Lund et al., 2011; Zhang et al., 2013). Furthermore, extracorpuscular a body fluid is exposed to higher levels of dioxygen compared to intracorpuscular. Dioxygen is an essential component for the metal-catalysed oxidation reactions, therefore an increase in dioxygen levels could be expected to result in an increase in the MCO reactions. A large part of the research done on the process of (metal-catalysed) oxidation of proteins outside of the body, is performed on muscle foods. The cells which are contained in meat are, like the molecules in a body fluid, separated from a living organism, and will thus not be provided with any new material. It is stated however that enough metals are present in meat to initiate and maintain the MCO reactions (Decker et al., 2000; Lund et al., 2011). Especially heme, a component of hemoglobin in blood, is a rich provider of iron molecules. As a result of degradation, the heme molecules break down and release their iron. Aside from blood, also sweat, urine, semen and saliva have been found to contain the copper and iron needed for MCO reactions ex vivo (Mitchell et al., 1949; Rodriguez & Diaz, 1995; Slivkova et al., 2009; George et al., 2017). Apart from the aforementioned, protein oxidation seems to follow a similar path outside of the body as to inside of the body.
In conclusion, the most common oxidation reactions one can expect to occur in excreted body fluids are those catalysed by a metal or directly through free radicals or other reactive oxidants.
3. Oxidation products
There are broadly two possible ways of oxidation of a protein: 1. the attack of the backbone of the protein and 2. modifications of the side-chains of the amino acids. A reaction between radicals and the backbone of a protein occurs mainly at the α-carbon site of an amino acid; the location where the side chains are attached (Davies, 2005). Such an attack will result in stabilised carbon-centred radicals, which can, through multiple reactions, lead to fragmentation of the backbone of the protein. The occurrence of backbone fragmentation is low however. Modifications to the side-chains can arise on the residues of all 20 amino acids, which generates many different products (see Appendix I: Table 1). Most side chains have multiple sites which can be attacked by an oxidant. It is dependent on the attacking oxidant which kind of damage will occur (Davies, 2012). The least reactive amino acids, are those with an aliphatic side chain. Aliphatic chains are those consisting of only carbon and
hydrogen. The most reactive amino acids residues tend to be those containing either aromatic rings (tryptophan, phenylalanine and tyrosine), from which hydrogen molecules are readily abstracted off of the benzene ring, or sulfur (methionine and cysteine), because of their high amount of easily removed electrons (Elias, 2008; Zhang et al., 2013). However, whether or not an amino acid is accessibleto a free radical, depends on its location in the tertiary structure of the protein it is in. Also, the residues that are in close proximity are determinant for the level of reactivity.
For ease of discussion of the distinct oxidation products, the different types of amino acid residues are divided into three different categories; aliphatic, aromatic and sulfur-containing. Some of the most common oxidation products can be seen in Appendix I: Table 1.
3.1 Aliphatic residues
In proteins, aliphatic side-arms are of the least reactive residues. If, however, oxidation is to occur, this will only happen with the most reactive of radicals (Davies, 2005). The potent oxidants hypochlorous acid (HOCl) and hypobromous acid (HOBr) can react with the residues of lysine and arginine, which results in the unstable chloramines and bromamines (Image 2). Decomposition of these products can give rise to the previously mentioned carbonyl groups. Asparagine, valine and glutamine are also found to react with HOBr and HOCl, but to a lesser extent (Davies, 2005). HOCl
and HOBr are both produced by white blood cells (FU et al., 2000), and because of the relative abundance of white blood cells, it can be assumed that HOCl and HOBr are common elements ex vivo. However, most products created through a reaction with HOCl or HOBr are often rapidly decomposed and therefore unstable markers, with the carbonyl groups being an exception.
HOCl does not only react with the aliphatic groups, but also exchanges electrons with the cyclic residue of histidine. The imidazole group in histidine side chains makes it an oxidatively labile amino acid (Elias et al., 2008). The hydroxyl radical (OH·) transfers electrons to the imidazole group, which in turn can react with O2, forming a peroxyl radical. After a chain of reactions, amongst others, asparagine, aspartic acid and unstable hydroperoxides are formed. Histidine is most sensitive to MCO reactions however, which result in 2-oxo-histidine, which in turn can oxidize into different products. The histidine oxidation products are poorly characterized and only 2-oxo-histidine has been used as a biomarker for histidine oxidation (Davies et al., 1999; Hawkins & Davies, 2001). However, it is yet unknown whether 2-oxo-histidine is a reaction product specific to the oxyradical chemistry, therefore, the potential of this marker is low.
Like histidine, valine, leucine, isoleucine, glutamic acid, arginine and proline are also known to produce unstable hydroperoxides (Hawkins & Davies, 2001). This is a result of peroxyl formation through a reaction of the carbon-centred radicals with O2, on either the side chains or the α-carbon sites of the amino acids. Hydroperoxides are thought to be unstable because of their rapid degradation when exposed to light, heat, reducing agents or metal ions (Davies, 1999). Moreover, peroxyls as well as hydroperoxides can, through a series of steps, generate carbonyl groups and alcohols (Hawkins & Davies, 2001; Davies, 2005) (Image 3). These alcohols are, in general, stable products which are hardly
susceptible to subsequent oxidation (Davies, 1999). It should be noted that the carbonyl product is prone to react with the α-amino group of lysine (Stadtman & Levine, 2000). This could occur within the same protein, or with a protein in its vicinity, which results in intra- or intermolecular protein cross-linked derivatives. Such structures cannot be degraded by the proteasome, and therefore contribute to the accumulation of oxidized products. However, since carbonyls are such a common product of oxidation, they are often used as biomarkers to measure oxidative damage. Yet, oxidation is not the only process through which carbonyl groups are generated. Glycation of proteins, for example, is also known to add carbonyl groups to the residues of amino acids. Glycation is a step in the Maillard reaction, which is a common process occurring both inside and
outside of the body (Draelos, 2013). Considering that carbonyl groups are not only a product of oxidation, measuring carbonyl groups as a way of quantifying the level of oxidation, can only be used presumptively (Levine et al., 1994).
Some other products which occur are aminomalonic acid derived from
glycine, 5-hydroxy-2-aminovaleric acid which is formed by oxidation of either proline or arginine, and α-ketoisocaproic acid, isovaleric acid, isovaleraldehyde and isovaleraldehyde oxime which are all products of oxidation of leucine. Each of these reaction products however, can also be generated by other reactions or other amino acids, and are therefore found to be nonspecific markers (Davies et al., 1999).
3.2 Aromatic residues
The aromatic residues of phenylalanine, tryptophan and tyrosine and their susceptibility to oxidation, makes these amino acids biomarkers with a high potential. The cyclic residues are not sensitive to MCO reactions, but rather react with the Reactive Oxygen Species (Domingues et al., 2003; Zhang et al., 2013). The products resulting from these oxidations are rather well studied. Oxidation of the ring of phenylalanine results in 2-hydroxyphenylalanine, also known as ortho-tyrosine, 3-2-hydroxyphenylalanine, known as meta-tyrosine, and 4-hydroxyphenylalanine. Since the latter product is equal to tyrosine, this cannot be used as a marker, but the other two products seem to be stable products (Davies et al., 1999).
Tyrosine is oxidized into tyrosyl phenoxyl radicals, which can be converted into dityrosine through self-dimerization (Image 4). This structure is
resistant to degradation by lytic enzymes, and therefore thought to be a relatively stable marker (Zhang et al., 2013). 3,4-dihydroxyphenylalanine, known as DOPA, is another oxidation product of tyrosine. This molecule is a catechol which is sensitive to subsequent oxidation, which can lead to the formation of quinone and cyclized products (Davies, 1999). In turn, these products tend to oxidize even further into different molecules, which makes it an unstable product. Again, HOCl, or other chlorinating species, can react with tyrosine, which leads to 3-chlorotyrosine. In the presence of H2O2 and Br2 oxidized tyrosine gives rise to 3-bromotyrosine. A
reaction with nitrating species results in 3-nitrotyrosine (Davies et al., 2005). If sufficient
Image 3. Degradation of a hydroperoxide leading
to the formation of a carbonyl group.
oxidants are present, 3-chlorotyrosine, 3-bromotyrosine and 3-nitrotyrosine can undergo further reactions to form 3,5-dichlorotyrosine, 3,5-dibromotyrosine and 3,5-dinitrotyrosine, respectively. This should be kept in mind if one would want to use the 3-chloro-, 3-bromo-, or 3-nitrotyrosine as a marker.
The oxidant HOCl moreover reacts with tryptophan, to form the stable 2-hydroxyindole (Pattinson & Davies, 2001). When tryptophan reacts with a radical and subsequently with O2, this results in different products. Part of these products can be characterized by an addition to the ring, which results in 2-, 4-, 5-, 6-, or 7-hydroxytryptophan or an opening up of the ring, which leads to N-formylkynurenine and kynurenine. The latter product is prone to oxidize further into 3-hydroxykynurenine (Perdivara et al., 2010) (Image 5). The unstable hydroperoxides are also a result of tryptophan oxidation with O2, as well as alcohols and cyclized products.
3.3 Sulfur-containing residues
The residues of methionine, cysteine, and cystine, the oxidized dimer form of cysteine, are the most susceptible to oxidation reactions. Their sulfur atoms readily react with all forms of ROS and compared to other amino acids, the conditions needed for oxidation to occur are rather flexible (Zhang et al., 2013). Of all amino acids, cysteine and methionine are found to react most rapidly with HOCl (Pattinson & Davies, 2001). When comparing reaction rates of free amino acids with HOCl, methionine and cysteine are found to be approximately 100 times more reactive than the other amino acids. HOCl in reaction with cysteine results in disulfides and oxyacids, while a reaction with methionine leads to methionine sulfoxide. Different types of oxidative reactions can result in methionine sulfoxide, but only the oxidation of methionine side-arms with ROS HOCl is found to be reversible (Stadtman & Levine, 2000). The enzyme methionine sulfoxide reductase, which is highly abundant in most animal tissues, can converse methionine sulfoxide back to methionine. Considering there is no supply of this enzyme outside of the body, the reduction of methionine sulfoxide will most probably be very limited in excreted body fluids. If not reduced, the sulfoxide can oxidize into the very stable and irreversible methionine sulfone, though the occurrence of this reaction is low (Davies, 2005).
Cys-SOH, one of the oxyacids formed by oxidation of cysteine, is highly unstable and prone to react with any thiol in its vicinity, which results in the disulfide cystine. Since cystine is a naturally occurring product, this is not considered to be an oxidation marker. Cys-SOH can also react with HOCl, which leads to sulfenyl chloride. A reaction between sulfenyl chloride and another amine can result in sulfonamide (Wang et al., 2004). Sulfenyl chloride is short-lived, and therefore thought to be an unstable marker (Davies, 2005). A second option for Cys-SOH is to further oxidize into the more stable Cys-SO2H or Cys-SO3H. As with methionine, the oxidation of cysteine to Cys-SOH is found to be reversible (Barelli et al., 2008) (Image 7).
Most of the abovementioned oxidation products were discovered by studying free amino acids, but it is suggested that the formed products are very similar to the oxidation products generated by amino acids in proteins (Elias et al., 2008). The difference between free amino acids and those contained in a protein is the tertiary structure of the protein, which causes other amino acids to be in close proximity. This could lead to radical transfer reactions, where one oxidation reaction leads to the creation of a radical which can lead to the oxidation of some other amino acid residue located somewhere else in the protein. Such radical transfer is found to occur over very large distances, depending on the structure of the protein (Davies, 2005). Since the residues of tryptophan, tyrosine, histidine, cysteine, cystine and methionine are found to be the most readily oxidizable, a radical reaction is most likely to be transferred to any of these residues. It depends however, on the selectivity of the radical that is attacking. The hydroxyl radical, for example, is highly reactive and will therefore react with basically anything that is near, therefor its selectivity will be low (Elias et al., 2008). A less reactive species, such as peroxide H2O2, is far more selective and will target the more readily oxidizable amino acids more frequently (Davies, 2005).
Image 6. Oxidation products of methionine.
4. Mass Spectrometry
4.1 Mass Spectrometry and biomarkers
If one would want to use the protein oxidation products as an indicator of the age of a body fluid stain, it is necessary that the presence of these oxidative modifications can somehow be identified, and that íf they are found to be present, they can be quantified. The question is whether Mass Spectrometry (MS) can be used for this purpose. With MS, the mass-to-charge ratio (m/z) of ionized analytes is being measured. The change in chemical composition of a protein caused by oxidative modifications, leads to an alteration of the mass-to-charge ratio of the overall protein and of the amino acid side chains where the oxidation occurred (Verrastro et al., 2015). An MS measurement results in a spectrum showing all the charges present in a sample relative to the mass of the corresponding ions. Databases which are created with the help of proteomics are being used for the interpretation of the MS results, by matching the MS spectra to peptide sequences from the database. With the help of this matching-system, it can be deduced which oxidative modifications are present in a protein. However, because of the large number of different modifications, and the possibility of two products having the same mass, the identification of these modifications remains challenging. The subsequent section will elaborate on some of the most commonly studied oxidation products identified with Mass Spectrometry.
To be able to locate oxidative modifications more easily, chemical reagents can be used to react with the modifications. Such reagents can serve as a tag which label the modified proteins and can both facilitate enrichment and detection. Enrichment is needed to isolate the proteins when they are present in a lower abundance and/or to reduce the sample complexity to aid the eventual spectrum analysis. Isolation of a protein with the help of a tag results in targeted MS analysis, since now only the tagged protein will be subjected to further measurement, instead of all proteins present in the sample. The reagent 2,4-dinitrophenylhydrazine (DNPH) is a well-established tag which can react with carbonyl groups (Robinson et al., 1998; Barelli et al., 2008; Verrastro et al., 2015). It moreover serves as the matrix needed for MALDI-MS and eliminates the need for further enrichment. Together with the use of the antibody anti-DNPH the carbonylated proteins can be isolated from the sample. ESI-MS can also be used to analyse the labelled proteins, and with the use of this technique a proteome-wide study was conducted on protein carbonyl groups. This study identified 210 carbonylated proteins and detected 643 carbonyl locations in a specific proteome. One disadvantage of this DNPH-labelling technique is that relatively large quantities (up to a milligram) of protein are needed. One other way to tag carbonyl groups is through derivatization with biotin hydrazide. The biotinylated proteins can then be selected and isolated through avidin affinity chromatography, after which LC-MS/MS can be used for identification (Mirzaei & Regnier, 2007).
The oxidation products of cysteine, such as the disulfides and sulfenic acids, have also been found to be detectable with the use of biotinylation. Because of the strong binding of biotin to avidin, avidin affinity chromatography can again be used to isolate the products, and MS for identification (Verrastro et al., 2015). A downside to biotinylation is that it compromises the efficiency of the ionisation needed for MS, as well as the fragmentation of the peptide. The presence of oxyacids can also be measured without the use of a tag, through non-targeted analysis of the change of mass of the parent protein with ESI-MS. Formation of Cys-SOH results in an increase of 16 Da relative to the mass of cysteine, while
sulfinic and sulfonic acid result in an increase of 32 Da and 48 Da, respectively (Baez et al., 2015). When analysing an untargeted sample with MS, it is preferred to single out a particular modification for the system to recognize. Both sulfenic (oxidation of C) and sulfonic acid (trioxidation of C) are default modifications which can be selected for with MS. It should be kept in mind however, that oxidative modifications, under normal conditions, are not fixed modifications which are applied universally, to every residue. They are variable modifications which may be present on a residue, but not necessarily. Therefore, when analysing a protein or peptide for the presence of, for example, ‘oxidation of C’ all cysteines in that protein/peptide have to be reviewed consecutively for the presence of the default mass increase of 16 Da. This leads to a longer search time and reduced specificity (“Modifications”, 2016).
Methionine, the other sulfur-containing amino acid, was found to oxidize into methionine sulfoxide. With the use of untargeted MS, it can be seen that this product has a mass increase of 16 Da compared to methionine. One other way to prove the presence of methionine sulfoxide is through the detection of its fragmentation product with Collision-Induced Dissociation (CID). This product is 64 Da lower in mass than its precursor, which can be explained by the loss of methanesulfenic acid (CH3SOH) from the residue of methionine sulfoxide, which has a mass of 64 Da (Guan et al., 2003). This particular mass loss of 64 Da is unique to peptides containing methionine sulfoxide. This information is especially useful when knowing that the residues of methionine sulfoxide and phenylalanine share the same nominal mass. Methionine sulfone can be detected with the help of ESI-MS by measuring a mass increase of 32 Da, relative to the mass of methionine. Methionine sulfoxide (oxidation of M) and methionine sulfone (dioxidation of M) are also default modifications which can be selected for with MS.
Of the numerous oxidation products of tyrosine, 3-nitrotyrosine is best studied in relation to MS. Immunoprecipitation methods have been performed where anti-3-nitrotyrosine antibodies are used as tags to isolate the oxidized proteins, after which MS is used for identification (Verrastro et al., 2015). However, because of the low abundance of tyrosine modifications and the high amount of background proteins which are unmodified, the identification of tyrosine modifications has been found to be challenging. In accordance, a study of Orhan et al. (2004) showed that concentrations of 3-nitrotyrosine, 3-chlorotyrosine and 3-bromotyrosine were too low to be measured with MS. However, they did find dityrosine to be detectable in urine samples with the highly selective and sensitive method HPLC–APCI–MS/MS. Within the same experiment the research group was able to identify the oxidation product of phenylalanine, ortho-tyrosine, but concentrations of meta-tyrosine were, again, too low. Phenylalanine nor tyrosine show default modifications in MS which can be selected for to single out the oxidation products.
Perdivara and colleagues (2010) found that the process of separating tryptophan-containing proteins by SDS-PAGE, may result in tryptophan oxidations. For this reason, it is not preferred to use gel electrophoresis as a separation method, as it could lead to an overestimation of tryptophan oxidation products. With electrospray MS however, tryptophan modifications can rather easily be identified without the use of a tag. Tryptophan itself was found to have a m/z of 205, the modification hydroxytryptophan a m/z of 221 (+16 Da) and m/z 237 (+32 Da) corresponds to either dihydroxytryptophan or isomeric N-formylkynurenine (Domingues et al., 2003). N-formylkynurenine/dihydroxytryptophan (dioxidation of W) is a default modification which can be selected for when performing MS. Kynurenine shows a weight change of +4 Da relative to tryptophan and 3-hydroxykynurenine a change of +20 Da.
In conclusion, the detection and identification of oxidative modifications with Mass Spectrometry is still a challenging task, but is has been proven to be possible. MS shows great potential for usage in the forensic field to help establish biomarkers potentially capable of determining the time of deposition of biological traces.
4.2 From MS to immunoassays
Subsequent to the identification of potential markers, is should be tested whether the level of oxidation actually correlates to the time of deposition, with the help of a validation method. The validation phase can be performed with MS, but most often the biomarkers established through MS are analysed in an immuno-based assay (Good et al., 2007). If the markers are proven to be successful in giving an indication of the time of deposition, the immunoassays could also be used to put the markers into practice. MS is a highly sensitive tool, and suitable for the discovery of markers, but because of its lengthy procedure, and the high costs of one run, it is preferred to have a different method for everyday use. The advantage of an immunoassay, such as ELISA, is that it is relatively fast, simple to read, comparatively low in costs and that it could be made into a kit which is portable to the crime scene. Therefore, it should be assessed whether the aforementioned biomarkers measured through targeted analyses can also be identified with the use of, for example, a fluorescent tag bound to the antibody, instead of MS. And for the biomarkers measured with non-targeted analyses antibodies need to be established so that they can be transformed into an immuno-based assay.
For cysteine, an antibody exists which is able to detect both sulfenic acid, sulfinic acid and sulfonic acid. Visualization of sulfenic acid with the use of this antibody has been proven to be successful (Seo & Carroll, 2009). For 3-hydroxykynurenine, a product of tryptophan, an antibody has successfully been identified and has been proven to be compatible with ELISA (Staniszewska & Nagaraj, 2005). Dityrosine too has an antibody which can be used for ELISA. However, for some other products this transfer to an immuno-based assay will pose a difficult task. Wehr & Levine (2012) investigated whether they could find an antibody specific for methionine sulfoxide. The scientists screened 15 billion clones, and could not identify a single antibody for methionine sulfoxide. Hence, they found it surprising to discover that Oien and colleagues (2009) did claim to have detected a methionine sulfoxide specific antibody. Therefore, Wehr & Levine (2013) tested the specificity of this antibody, and found that it was not capable of detecting methionine sulfoxide in a protein highly exposed to this oxidation nor could it distinguish between two particular forms of modifications. Yet, a view studies have used the antibody and were able to measure methionine sulfoxide levels. Such contradicting results indicate the need for further research into antibodies for methionine modifications and oxidation products as a whole.
Discussion
In the past few decades, protein oxidation has received more and more attention, because of their involvement in many processes such as ageing, the development of diseases, and deterioration of the quality of food products. The interest of the forensic field in protein oxidation, however, is a rather recent evolution. For this reason, the study of the relation between protein oxidation and age determination of body fluids, is still in its early stages, which makes insight into this correlation and the establishment of stable biomarkers ever so important.
The aim of this article was to give an overview of what oxidative products are most common, and which ones have the potential of being a good biomarker. It is suggested that the oxidation products formed by the aliphatic amino acids are no promising biomarkers. This is based on the finding that, first, aliphatic residues are of the least reactive, and therefore their oxidation products less abundant, and second, the products are not thought to be specific nor stable enough. An exception to the latter is the generation of carbonyl groups by some aliphatic amino acids. As can be seen in Appendix I: Table 1, a fairly large number of the amino acids form carbonyl compounds after oxidation, namely; lysine, leucine, isoleucine, arginine, valine, proline, serine, threonine, tryptophan and histidine. Carbonyls are by far the most commonly used marker for protein oxidation, and because of their large numbers, less sensitive measuring methods are needed compared to, for example, quantifying the oxidation products of tyrosine, which are present in far lower numbers (Dalle-Donne et al., 2003). Moreover, the fact that so many different oxidized amino acids are found to receive a carbonyl group, suggest that carbonyl compounds can be expected to be found in basically any protein and thus in any body fluid. Also, since crime scene stains can be of a very low volume, it is necessary that an oxidation product is abundantly present, for it to be able to be measured in such small crime scene stains. As previously mentioned, measuring carbonyl groups could result in an overestimation of the occurred oxidation, since there are also other processes through which carbonyls are produced. Secondly, measuring the carbonyls with the use of antibodies could also lead to overestimation, given that these antibodies are not entirely specific to carbonyl groups. However, this overestimation does not necessarily need to be a problem as long as it is occurring constantly. The level of carbonyls is expected to increase over time at a stable rate; a constant overestimation of abundance at each moment of measurement, would still result in a steady increase. This increase could therefore still be used to be correlated to the time of deposition of the source. Berardo and colleagues (2015) showed that, in the first ten days after deposition of a meat sample with an oxidant (H2O2), the formed carbonyl compounds displayed a steady increase, with a potential of increasing even further after this timeframe (Figure 1).
Notable is that they found that pH has an influence on the initial carbonyl levels, but after a period of ten days, carbonyl levels were equal for both pH groups. pH has an influence on oxidation in general, as do temperature, light exposure and other environmental factors (Wellington et al., 2017). Since the conditions after deposition at a crime scene cannot be controlled, only the collection and storage of the samples can be regulated. Several studies (Estévez & Cava, 2004; Soyer et al., 2010) showed that even during refrigerated storage at -18 ˚C, carbonyl levels keep rising. Therefore, Schipper et al. (2007) recommends to store samples at -80 ˚C, to minimize protein degradation. Two other aspects that should be taken into consideration are the age and health of the donor of the stain and their relation to protein levels in general, but specifically protein oxidation levels. Increased levels of carbonyls are found in patients suffering from, for example, Parkinson's disease, Alzheimer’s disease, muscular dystrophy, diabetes, induction of renal tumours and rheumatoid arthritis (Stadtman & Levine, 2000). Despite the mentioned difficulties, carbonyl groups remain a biomarker with a high potential.
Where carbonyl groups display a stable increase over time, methionine oxidation exhibits a different degradation pattern. Kotiaho and colleagues (2000), studied the reaction between gaseous oxidant ozone, O3, and an aqueous solution containing methionine; a reaction similar to this is not unlikely to occur in excreted body fluids. The rate at which methionine is oxidized into methionine sulfoxide (MetO), followed by an increase of methionine sulfone (MetO2) is depicted in Figure 2. This pattern occurs within a relatively short timeframe of 20 minutes. It is thought however, that a sample left at a crime scene will show a similar degradation pattern but at a much lower rate. This is based on the finding that the results of this experiment were obtained under constant ozonation of the sample, which is thought not to be representative of normal circumstances. Crime scene stains will be exposed to less severe oxidizing conditions, and therefore it will most likely take longer for the oxidation products to reach these intensity levels.
Figure 1. (A) and (B) show the change in abundance of carbonyls formed in oxidized meat over
Possibly, the ratio of methionine sulfoxide and methionine sulfone could be used as a measure of deposition time. Using a ratio as a biomarker could eliminate part of the inter-individual variance; a person with elevated methionine-containing protein levels could still display a MetO/MetO2-ratio equal to that of a person with normal protein levels. Taken together with the fact the methionine and its oxidation products are rather easily measured with the use of MS, it is advised to study methionine and its potential as biomarker more thoroughly.
It is proposed that the oxidation products of cysteine, more specifically the oxyacids, might show a similar degradation pattern. Cys-SOH quickly oxidizes into the more stable Cys-SO2H, which in turn can oxidize further into Cys-SO3H, if sufficient oxidants are present. Potentially, the ratio of Cys-SO2H/Cys-SO3H could be correlated to oxidation time, however this has not been studied yet. A simple experiment would be to expose cysteine-containing proteins to an oxidant for a specific amount of time, and measure the change in abundance of the oxidation products.
A comparable experiment should be performed on tryptophan and its oxidation products. The oxidation products of tryptophan could potentially be used as biomarkers, but little is known about how these products change in abundance over a longer period of time. Because of the fluorescent property of tryptophan and its oxidation products, fluorescence spectroscopy could also be used to measure the level of oxidation. Dalsgaard and colleagues (2007), studied the level of protein oxidation in milk by separately measuring the levels of carbonyls, tryptophan, N-formylkynurenine, kynurenine and dityrosine over a period of 50 hours. To confirm the believe that a decrease in tryptophan correlated to an increase of N-formylkynurenine and kynurenine, they performed a linear regression and found an R2 of ≥ 0.88 and ≥ 0.84, respectively. As expected, the carbonyl compounds too showed a stable increase over time as well as the dityrosine concentrations. It would be interesting to study these changes in abundance over an extended period of time to see whether these biomarkers are suitable for longer time intervals. Moreover, it is suggested to study the oxidation products of one amino acid in one assay, to give a better overview of the interaction between different products.
Figure 2.Depicted is the intensity change of methionine (Met), methionine sulfoxide (MetO) and methionine sulfone (MetO2), as a function of ozonation time (edited from: Kotiaho et al., 2000).
The previous experiment proved dityrosine to be a successful biomarker, yet, other products of oxidized tyrosine and phenylalanine do not seem to be promising markers because of their overall low abundance, considering the often small volume of crime scene stains. Despite the generally small size of a body fluid stain, the information embedded in these stains can be of great value to a criminal case. Having a stable biomarker which can be correlated to the age of a body fluid stain, can be of great use in the forensic field to establish when a crime scene stain was deposited. Such intelligence could help the crime scene investigators to determine which marks are relevant to the crime, or to verify a person’s statement and possibly to solve a crime.
Conclusion
In conclusion, it is advised to study the oxidative products of methionine, cysteine, tyrosine and tryptophan and their potential as biomarker more thoroughly, as well as the carbonyl compounds. It is suggested to combine multiple biomarkers into one panel, to ensure the detection of at least one of them in ~100% of the samples collected. Therefore, it should be investigated whether it is possible to produce an immuno-based assay which is capable of visualizing a combination of (some of) these oxidation products. It is advised to focus on the carbonyl groups, methionine sulfoxide, methionine sulfone, cysteines oxyacids, dityrosine and the different kynurenines. Future research should test the biomarkers under a greater variety of conditions, so as to mimic real-life circumstances. Heat, light, humidity, pH, deposition surface and the age, gender, health and diet of the donor should be taken into consideration.
Research on protein oxidation applied in the forensic field is still in its early stages, nevertheless very promising.
References
Ackermann, K., Ballantyne, K. N., & Kayser, M. (2010). Estimating trace deposition time with circadian biomarkers: a prospective and versatile tool for crime scene reconstruction. International journal of
legal medicine, 124(5), 387-395.
Altelaar, A. M., Munoz, J., & Heck, A. J. (2013). Next-generation proteomics: towards an integrative view of proteome dynamics. Nature Reviews Genetics, 14(1), 35-48.
Baez, N. O. D., Reisz, J. A., & Furdui, C. M. (2015). Mass spectrometry in studies of protein thiol chemistry and signaling: opportunities and caveats. Free Radical Biology and Medicine, 80, 191-211. Barelli, S., Canellini, G., Thadikkaran, L., Crettaz, D., Quadroni, M., Rossier, J. S., Tissot, J., & Lion, N. (2008). Oxidation of proteins: basic principles and perspectives for blood proteomics. Proteomics-Clinical Applications, 2(2), 142-157.
Bauer, M., Polzin, S., & Patzelt, D. (2003). Quantification of RNA degradation by semi-quantitative duplex and competitive RT-PCR: a possible indicator of the age of bloodstains?. Forensic science
international, 138(1), 94-103.
Berardo, A., Claeys, E., Vossen, E., Leroy, F., & De Smet, S. (2015). Protein oxidation affects proteolysis in a meat model system. Meat science, 106, 78-84.
Bremmer, R. H., Nadort, A., Van Leeuwen, T. G., Van Gemert, M. J., & Aalders, M. C. (2011). Age estimation of blood stains by hemoglobin derivative determination using reflectance spectroscopy.
Forensic science international, 206(1), 166-171.
Crosley, L. K., Duthie, S. J., Polley, A. C., Bouwman, F. G., Heim, C., Mulholland, F., Horgan, G., Johnson, I.T., Mariman, E.C., Elliot, R.M., de Roos, B., & Daniel, H. (2009). Variation in protein levels obtained from human blood cells and biofluids for platelet, peripheral blood mononuclear cell, plasma, urine and saliva proteomics. Genes & nutrition, 4(2), 95.
Dalle-Donne, I., Giustarini, D., Colombo, R., Rossi, R., & Milzani, A. (2003). Protein carbonylation in human diseases. Trends in molecular medicine, 9(4), 169-176.
Dalsgaard, T. K., Otzen, D., Nielsen, J. H., & Larsen, L. B. (2007). Changes in structures of milk proteins upon photo-oxidation. Journal of agricultural and food chemistry, 55(26), 10968-10976. van Dam, A., Schwarz, J. C., de Vos, J., Siebes, M., Sijen, T., van Leeuwen, T. G., Aalders, M.C.G., & Lambrechts, S. A. (2014). Oxidation monitoring by fluorescence spectroscopy reveals the age of fingermarks. Angewandte Chemie International Edition, 53(24), 6272-6275.
Davies, M. J., Fu, S., Wang, H., & Dean, R. T. (1999). Stable markers of oxidant damage to proteins and their application in the study of human disease. Free Radical Biology and Medicine, 27(11), 1151-1163.
Davies, M. J. (2005). The oxidative environment and protein damage. Biochimica et Biophysica Acta
(BBA)-Proteins and Proteomics, 1703(2), 93-109.
Davies, M. J. (2012). Free radicals, oxidants and protein damage. Aust Biochem, 43, 8-12.
Decker, E., Faustman, C., & Lopez-Bote, C. J. (2000). Protein oxidation and implications for muscle food quality. Antioxidants in muscle foods, 3-23.
Domingues, M. R. M., Domingues, P., Reis, A., Fonseca, C., Amado, F. M., & Ferrer-Correia, A. J. (2003). Identification of oxidation products and free radicals of tryptophan by mass spectrometry.
Journal of the American Society for Mass Spectrometry, 14(4), 406-416.
Domon, B., & Broder, S. (2004). Implications of new proteomics strategies for biology and medicine.
Journal of proteome research, 3(2), 253-260.
Draelos, Z. D. (2013). Aging skin: The role of diet: Facts and controversies. Clinics in dermatology,
31(6), 701-706.
Elias, R. J., Kellerby, S. S., & Decker, E. A. (2008). Antioxidant activity of proteins and peptides.
Critical reviews in food science and nutrition, 48(5), 430-441.
Esser, D., Alvarez-Llamas, G., De Vries, M. P., Weening, D., Vonk, R. J., & Roelofsen, H. (2008). Sample stability and protein composition of saliva: implications for its use as a diagnostic fluid.
Biomarker insights, 3, 25.
Estévez, M., & Cava, R. (2004). Lipid and protein oxidation, release of iron from heme molecule and colour deterioration during refrigerated storage of liver pâté. Meat Science, 68(4), 551-558.
George, T., Honnamurthy, J. B., Shivashankara, A.R., Suresh, S., Manjeshwar Shrinath Baliga, M.S. (2017).
Correlation of blood and salivary levels of zinc, iron and copper in head and neck cancer patients: an investigational study. Avicenna J Med Biochem, 5(1), 35-39.
Good, D. M., Thongboonkerd, V., Novak, J., Bascands, J. L., Schanstra, J. P., Coon, J. J., Dominiczak, A. & Mischak, H. (2007). Body fluid proteomics for biomarker discovery: lessons from the past hold the key to success in the future. Journal of proteome research, 6(12), 4549-4555.
Guan, Z., Yates, N. A., & Bakhtiar, R. (2003). Detection and characterization of methionine oxidation in peptides by collision-induced dissociation and electron capture dissociation. Journal of the
American Society for Mass Spectrometry, 14(6), 605-613.
Hawkins, C. L., & Davies, M. J. (2001). Generation and propagation of radical reactions on proteins.
Biochimica et Biophysica Acta (BBA)-Bioenergetics, 1504(2), 196-219.
Helmerhorst, E. J., & Oppenheim, F. G. (2007). Saliva: a dynamic proteome. Journal of dental
research, 86(8), 680-693.
Houée-Lévin, C., Bobrowski, K., Horakova, L., Karademir, B., Schöneich, C., Davies, M. J., & Spickett, C. M. (2015). Exploring oxidative modifications of tyrosine: An update on mechanisms of formation, advances in analysis and biological consequences. Free radical research, 49(4), 347-373. Jimenez-Verdejo, A., Osuna, E., Garcia-Olivares, E., & Luna, A. (1994). Study of the enzymatic activity of GGT, LDH, PAP and PSA in semen stains: application to age calculation. Forensic science
international, 68(1), 7-15.
Kotiaho, T., Eberlin, M. N., Vainiotalo, P., & Kostiainen, R. (2000). Electrospray mass and tandem mass spectrometry identification of ozone oxidation products of amino acids and small peptides.
Journal of the American Society for Mass Spectrometry, 11(6), 526-535.
Kussmann, M., Raymond, F., & Affolter, M. (2006). OMICS-driven biomarker discovery in nutrition and health. Journal of biotechnology, 124(4), 758-787.
Levine, R. L., Williams, J. A., Stadtman, E. P., & Shacter, E. (1994). Carbonyl assays for determination of oxidatively modified proteins. Methods in enzymology, 233, 346-357.
Lund, M. N., Heinonen, M., Baron, C. P., & Estevez, M. (2011). Protein oxidation in muscle foods: A review. Molecular nutrition & food research, 55(1), 83-95.
Mirzaei, H., & Regnier, F. (2007). Identification of yeast oxidized proteins: chromatographic top-down approach for identification of carbonylated, fragmented and cross-linked proteins in yeast. Journal of
Chromatography A, 1141(1), 22-31.
Mitchell, H. H., Hamilton, T. S., & Haines, W. T. (1949). The dermal excretion under controlled environmental conditions of nitrogen and minerals in human subjects, with particular reference to calcium and iron. Journal of Biological Chemistry, 178, 345-361.
“Modifications”. (2016). Retrieved from http://www.matrixscience.com/help/pt_mods_help.html
Ngamchuea, K., Chaisiwamongkhol, K., Batchelor-McAUley, C., & Compton, R. G. (2017). Chemical Analysis in Saliva and the Search for Salivary Biomarkers–A Tutorial Review. Analyst.
Oien, D. B., Canello, T., Gabizon, R., Gasset, M., Lundquist, B. L., Burns, J. M., & Moskovitz, J. (2009). Detection of oxidized methionine in selected proteins, cellular extracts and blood serums by novel anti-methionine sulfoxide antibodies. Archives of biochemistry and biophysics, 485(1), 35-40. Orhan, H., Vermeulen, N. P., Tump, C., Zappey, H., & Meerman, J. H. (2004). Simultaneous determination of tyrosine, phenylalanine and deoxyguanosine oxidation products by liquid chromatography–tandem mass spectrometry as non-invasive biomarkers for oxidative damage.
Journal of Chromatography B, 799(2), 245-254.
Pattison, D. I., & Davies, M. J. (2001). Absolute rate constants for the reaction of hypochlorous acid with protein side chains and peptide bonds. Chemical research in toxicology, 14(10), 1453-1464. Perdivara, I., Deterding, L. J., Przybylski, M., & Tomer, K. B. (2010). Mass spectrometric identification of oxidative modifications of tryptophan residues in proteins: chemical artifact or post-translational modification?. Journal of the American Society for Mass Spectrometry, 21(7), 1114-1117.
Rodríguez R, E., & Díaz R, C. (1995). Iron, copper and zinc levels in urine: relationship to various individual factors. Journal of Trace Elements in Medicine and Biology, 9(4), 200-209.
de Roos, B., & McArdle, H. J. (2008). Proteomics as a tool for the modelling of biological processes and biomarker development in nutrition research. British Journal of Nutrition, 99(S3), S66-S71. Schipper, R., Loof, A., de Groot, J., Harthoorn, L., Dransfield, E., & van Heerde, W. (2007). SELDI-TOF-MS of saliva: methodology and pre-treatment effects. Journal of Chromatography B, 847(1), 45-53.
Seo, Y. H., & Carroll, K. S. (2009). Profiling protein thiol oxidation in tumor cells using sulfenic acid-specific antibodies. Proceedings of the National Academy of Sciences, 106(38), 16163-16168.
Slivkova, J., Popelkova, M., Massanyi, P., Toporcerova, S., Stawarz, R., Formicki, G., Lukac, N., Putala, A. & Guzik, M. (2009). Concentration of trace elements in human semen and relation to spermatozoa quality. Journal of Environmental Science and Health Part A, 44(4), 370-375.
Soladoye, O. P., Juárez, M. L., Aalhus, J. L., Shand, P., & Estévez, M. (2015). Protein oxidation in processed meat: mechanisms and potential implications on human health. Comprehensive Reviews
in Food Science and Food Safety, 14(2), 106-122.
Soyer, A., Özalp, B., Dalmış, Ü., & Bilgin, V. (2010). Effects of freezing temperature and duration of frozen storage on lipid and protein oxidation in chicken meat. Food chemistry, 120(4), 1025-1030. Stadtman, E. R., & Levine, R. L. (2000). Protein oxidation. Annals of the New York Academy of
Sciences, 899(1), 191-208.
Staniszewska, M. M., & Nagaraj, R. H. (2005). 3-Hydroxykynurenine-mediated modification of human lens proteins structure determination of a major modification using a monoclonal antibody. Journal of
Biological Chemistry, 280(23), 22154-22164.
Verrastro, I., Pasha, S., Jensen, K. T., Pitt, A. R., & Spickett, C. M. (2015). Mass spectrometry-based methods for identifying oxidized proteins in disease: advances and challenges. Biomolecules, 5(2), 378-411.
Wang, Y., Vivekananda, S., Men, L., & Zhang, Q. (2004). Fragmentation of protonated ions of peptides containing cysteine, cysteine sulfinic acid, and cysteine sulfonic acid. Journal of the
American Society for Mass Spectrometry, 15(5), 697-702.
Wehr, N. B., & Levine, R. L. (2012). Wanted and wanting: antibody against methionine sulfoxide. Free
Radical Biology and Medicine, 53(6), 1222-1225.
Wellington, E. (2017). Effects of Different Haematocrit Values on Estimation of Time since Deposition
of Human Blood Stains Using Diffuse Reflectance Spectroscopy (Doctoral dissertation, Murdoch
University).
Zhang, W., Xiao, S., & Ahn, D. U. (2013). Protein oxidation: basic principles and implications for meat quality. Critical reviews in food science and nutrition, 53(11), 1191-1201.
Appendix I: Table 1
Table 1. Stable and unstable products formed on oxidation of amino acids’ aliphatic, sulfur-containing,
and aromatic side chains (edited from; Davies, 2005). Amino Acid Products
Glutamic acid hydroperoxides (unstable)
Leucine hydroperoxides (unstable)
alcohols α-ketoisocaproic acid isovaleric acid isovaleraldehyde oxime isovaleraldehyde carbonyls
Glycine aminomalonic acid
Valine hydroperoxides (unstable)
alcohols
chloramines (unstable - from HOCl)
bromines (unstable - from HOBr)
carbonyl compounds
Proline hydroxyperoxides (unstable)
alcohols
5-hydroxy-2-aminovaleric acid
carbonyl compounds
Arginine hydroperoxides (unstable)
chloramines (unstable - from HOCl)
bromamines (unstable - from HOBr)
carbonyl compounds
Isoleucine hydroperoxides (unstable)
alcohols
carbonyl compounds
Lysine chloramines (unstable - from HOCl)
bromamines (unstable - from HOBr)
carbonyl compounds
Histidine 2-oxohistidine
chlorinated materials (unstable - from HOCl)
asparagine
aspartic acid
hydroperoxides (unstable)
alcohol and carbonyl products
Methionine methionine sulfoxide
methionine sulfone
Cysteine cystine (disulfide)
oxyacids (Cys-SOH, Cys-SO2H, Cys-SO3H)
sulfonamides
sulfenyl chloride (unstable - from HOCl)
meta-tyrosine (3-hydroxyphenylalanine)
tyrosine
Tyrosine 3,4-dihydroxyphenylalanine (DOPA) (unstable) dityrosine 3-chlorotyrosine 3,5-dichlorotyrosine 3-bromotyrosine 3,5-dibromotyrosine 3-nitrotyrosine 3,5-dinitrotyrosine hydroperoxides (unstable)
alcohols and cyclized products
Tryptophan N-formylkynurenine kynurenine 3-hydroxykynurenine 2-, 4-, 5-, 6-, 7-hydroxytryptophan 2-hydroxyindole hydroperoxides (unstable)
Appendix II: Search Strategy
The following will describe the used strategy for finding the literature which is included in the current article.
The first ±10 articles were provided by the supervisor, which were mainly used for writing the introduction. This selection of articles contained references of interest, which were subsequently read as well.
To find articles on a new topic, some of the following search terms were entered in google scholar/Web of Science:
- Body fluid identification protein forensic - Age prediction semen proteomics - Biomarkers saliva
- Circadian rhythm body fluids age estimation - Pathways of oxidative damage
- Oxidation food industry
- Extracorpuscular/In vitro protein oxidation - Body fluids metal concentrations
- Metal-catalysed oxidation meat - Histidine residue oxidation
- Protein oxidation methionine/tryptophan/tyrosine/cysteine - Methionine oxidation immuno-based assay
- Methionine sulfoxide antibody - Hydroxykynurenine antibody
- Percentage/frequency amino acids in proteins - Oxidation milk
Often ‘Since 2010/2014’ was selected, with the aim of finding more actual articles. From the articles obtained through these search terms, again, references of interest were abstracted and looked into. For some of the articles which contained a high amount of significant information, it was checked by whom there were cited. Also, some authors seemed to be reoccurring in many different articles, therefore their names were searched in relation to the corresponding topic.
