Hair Decay and its Effect on the Detection of Metabolites for the Purpose of Donor Profiling

Hair Decay and its Effect on the Detection of

Metabolites for the Purpose of Donor Profiling

Yoram Goedhart

8553 words

10999361

16 December 2019

Supervisor: Dr. B.M. (Ben) de Rooij

Examiner: Prof. Dr. A.C. (Arian) van Asten

Master Forensic Science

FNWI, University of Amsterdam

2

Abstract

Human hair fibres are continuously growing and being shed. Thus, it is not uncommon for hair fibres to be left behind during a criminal act. Hair shafts are a valuable source of bioinformation, providing information on the donor’s lifestyle, e.g. drug abuse or habits, as compounds and their metabolites get incorporated into the hair shaft. When encountered at a crime scene, knowing more about the hair donor can provide police forces with an investigative lead to narrow down the search for the perpetrator. However, hairs deposited during the criminal act are often left behind in unfavourable conditions, possibly leading to degradation. Therefore, to provide more knowledge on how the hair is actually degraded, the current study discusses how various factors contribute to damaging the hair shaft, with a focus on UV radiation, pH values, the microbiome (especially fungi), and insect activity. Furthermore, the study investigates to what degree the passage of time itself has any effect to the detection of metabolites by reviewing findings based on mummified remains. Finally, the detection of metabolites in forensic casework and actively damaged hair fibres is discussed to focus more on the effect of decay on metabolite detection. It appears that environments promoting desiccation and inhibiting microbial activity are most favourable for hair preservation. Furthermore, the environment, rather than time itself, was found to be more important. Nevertheless, metabolites in hair samples from both archaeological findings as those in forensic casework proved to be stable enough for detection. This highlights the relevance of donor profiling in hair as it appears to remain feasible even when samples are in a degraded state. Future suggestions include the gathering of data with a clear forensic focus to determine which metabolites can reflect specific searchable traits.

Keywords:

3

Table of Contents

1. Introduction ...4

2. Decay of the Hair Root...6

3. Decay of the Hair Shaft ...7

3.1 Ultraviolet Radiation...7

3.2 Water pH ...8

3.3 Microbial Alteration ...8

3.4 Insect Activity ...9

4. Metabolites in Mummified Remains ... 10

4.1 Drugs of Abuse ... 11

4.2 Habits ... 12

4.3 Hormones ... 13

5. Findings from Forensic Casework and Experimental Data... 15

5.1 Hair metabolites after exposure to ultraviolet radiation ... 15

5.2 Hair metabolites after exposure to soil and water ... 16

5.3 Comparison with data obtained from mummified remains ... 17

6. Conclusions ... 18

7. Discussion ... 19

References ... 21

4

1. Introduction

Hair is a keratinised biological material with a variety of functions, such as providing warmth and protecting the scalp from UV radiation. The anatomy of the hair mainly consists of two parts; the hair root and the hair shaft. The hair root (also known as the follicle) is rooted in the dermis and is the living part of the hair fibre. It allows the hair to grow and regrow upon damage. The visible part of the hair, the shaft, does not contain living cells and consists of three parts (Figure 1). First and outermost, the cuticle, which is a protective layer of overlapping thin dead cells on the outside, often compared with tiles on a roof. Second, the underlying cortex, which is the major component of the hair shaft and mainly contains protein filaments of α-keratin and pigment granules. While the first gives the hair its structure, the latter gives the hair its colour. Third and final, the inner medulla, which consists of vacuolated cells and may or may not be present at all [1].

Hair follows a continuous cycle of growth and shedding [1, 2]. The hair first follows an active growth phase (anagen), in which cell proliferation and keratinization take place. The latter describes a process in which older cells that are pushed up from the follicle dehydrate. As a result, the cell membrane and nucleus are lost [3]. At the same time, keratin is highly produced. After the anagen growth phase, the hair enters a transition phase (catagen) in which growth stops. In this phase the hair detaches from its blood supply. Finally, the hair enters the resting phase (telogen), in which the hair follicle goes dormant. At that moment, the old hair may shed on its own, or is pushed out by the new growing hair fibre underneath. Obviously, the hair growth cycles of each individual human hair are not synchronised, as humans do not moult. Rather, it is estimated that at any given moment, the majority of the hairs are in anagen phase (80-95%), followed by telogen phase (4-20%) and finally catagen phase (1-2%) [3]. Thus, growing and shedding of hair occurs naturally and continuously and it is estimated that people shed up to one hundred scalp hairs each day [4].

Given this fact, it is not surprising that hair fibres can often be encountered on crime scenes as they can be shed during the criminal act. In that case, the question that often arises is ‘Who does the hair found on the crime scene originate from?’. For forensic identification purposes, hair analysis began mainly as an observer-based comparison of features, during a time when DNA analysis was not available yet [5]. This method relies on class characteristics

Figure 1. Schematic structure of the hair shaft.

The hair shaft consists of three components. First, the outer cuticle, which functions as a protective barrier. Second, the cortex, which makes up the bulk of the hair shaft. The cortex cells contain mainly keratin filaments and pigment granules. Third, the medulla, which consists of vacuolated cells filled with either air or water. The medulla may be continuous, segmented or not be present at all. Picture adapted from [14].

5 observed in the cuticle, cortex and medulla. However, this type of visual comparison has been

criticized in the past as it lacks scientific basis [5]. Nevertheless, even after the onset of DNA analysis, the method continued to be in use. This is due to the fact that nuclear DNA can only be extracted from the hair root, which is often only found if the hair was forcible removed from the scalp [5]. Besides that, the root has to be in either the anagen or catagen phase, whereas hair roots found on-scene are often in their telogen phase [6]. The hair shaft, on the other hand, only contains mitochondrial DNA (mtDNA), which only infers the maternal lineage and has thus a limited statistical power. Still, these kind of analyses are of interest when a suspect is already apprehended, while there are many investigations in which this is not the case. For these scenarios it would be more valuable if the hair could be used to find an investigative lead.

It is known that lifestyle characteristics of the hair donor can be traced back in the hair. Through the donor’s diet, daily habits and medicinal use, certain compounds are incorporated into the hair shaft. Since hair is of course also a growing material, the hair shaft can even provide a timeline, in which the hair shaft closer to the scalp provides more recent information on the donor’s activities. The exact mechanisms underlying the incorporation of metabolites into the hair shaft are still a topic of discussion and have resulted in multiple models describing these [1]. Nevertheless, there appears to be a general consensus that compounds and their metabolites are indeed getting incorporated into the hair shaft.

The detection of these analytes is mostly challenged by the sample preparation. Some studies do not treat the hair samples beforehand, whereas others do; and the washing solutions vary greatly. Of course, caution should be in place not to lose the analytes of interest or contaminate the hair with others. The actual analysis, on the other hand, does appear to be comparable throughout literature. Most studies rely on gas chromatography coupled with mass spectrometry (GC-MS). In this way, many substances have already been identified, such as metabolites of common drugs of abuse (e.g. cocaine, ketamine, or amphetamines) [7–9]. Furthermore, efforts have been made to determine whether a hair donor is a smoker or non-smoker based on the presence of nicotine in the hair shaft [10]. Studies like these have sparked the interest in the forensic community as a possible novel approach for non-invasive drug testing. Yet, this same idea might provide a valuable approach to obtain an investigative lead upon encountering hair on the crime scene to narrow down the list of possible suspects.

Hair found on a crime scene, however, might be exposed to many environmental factors, leading to its decay. Yet, it is unknown how this process might affect the metabolites and whether metabolites can still be extracted after a long period of time. Therefore, the current review will focus on how the human hair degrades when exposed to different variables. Moreover, the study aims to shed a light on whether donor profiling is still feasible in samples obtained from extreme environments and those more often encountered in forensic casework. The effect of the passage of time itself to the detection of metabolites will be investigated by reviewing findings based on the relatively well-preserved hair of mummified remains. Furthermore, to focus more on the effect of decay on metabolite detection, studies regarding hair fibres obtained from forensic casework and experimental data will be discussed.

6

2. Decay of the Hair Root

As stated earlier, the hair consists of a root and shaft in which only the root consists of living cells. It is important to differentiate between the two parts as it is known that soft tissues decompose more quickly and in a different manner than hard, keratinized tissue. This means that in the case of a full hair, the root is expected to decompose first and differently than the shaft. However, at the moment there appears to be a lack of experimental research investigating the decay of the hair root. This is interesting since the hair root has plenty of forensic relevance; it can be used to determine the type of removal (forcibly or naturally shed), for nuclear DNA extraction and even suggests whether the hair has been removed post- or ante-mortem. One major feature to measure hair root decomposition is the so-called

dead-man’s ring, which describes a darkening or coloured band close to the root (Figure 2) [3].

According to Koch et al. [3] this phenomenon can only occur post-mortem and thus can only be observed when the hair is still attached to the scalp. Yet, similar observations have been made in hairs that were removed ante-mortem [11]. Koch and colleagues found that hair roots obtained from cadavers that were placed in a climate-controlled room or that were immersed in water showed slower degradation than those obtained from cadavers placed in a vehicle trunk, which already showed root banding as early as four days [3]. Post-mortem root banding of hair roots from cadavers placed outside on the surface depended largely on the

outside temperatures, with temperatures below 20° Celsius delaying decomposition. Finally, human cadavers were also buried in either shallow graves or deeper graves. It was found that hair roots appeared to be well-preserved in a shallow grave over the time period of maximally 25 days [3]. It should be noted, however, that apart from one sample, the temperatures during this experiment ranged from -1° to maximally 18° Celsius, as the samples were buried during the winter months. As with the surface samples, these lower temperatures would certainly have slowed down the decomposition rate. After two years of burial in a deeper grave, mainly early signs of decomposition were observed, whereas full root banding was observed in almost the entire set of hairs tested after an additional year.

Thus, it appears that the hair root can be quite stable over time given the environment. While the decomposition of the hair shaft has not been studied here, the decomposition rate of the hair root could possibly provide an indication for the survival of the hair shaft. Roots in anagen phase decompose more quickly than those in telogen phase, in which the root is almost fully keratinized [11]. This might further suggest that the hair shaft (also mainly keratin) would

Figure 2. Post-mortem root banding.

Shown in the middle is a root end in anagen phase with extensive black root banding. The dark segmented line starting at the same location is the medulla of the hair shaft. Source: [3].

7 be more stable over time than the hair root. Then again, it should be noted that the hair root

remains enveloped in the skin during the study by Koch et al. [3], whereas the hair shaft is fully exposed. The effects of full exposure to the environments on the hair root thus remains questioned here.

On the other hand, Tridico et al. [12] state that post-mortem root banding can occur regardless of the environment, as they observed it in hairs (both archaeological and contemporary) obtained from both a tropical environment as permafrost.

3. Decay of the Hair Shaft

As stated earlier, for forensic purposes the hair shaft may be of more interest. This is due to the fact that the hair root is often no longer attached to the hairs found on crime scenes and those that are found are mostly in telogen phase [6]. Depending on where the hair is deposited, be it either inside or outside, or wet or dry conditions, the hair shaft can be exposed to a variety of environmental conditions. Therefore, this section aims to elucidate how some of these variables might affect the hair shaft over time.

3.1 Ultraviolet Radiation

That UV radiation can be harmful is common knowledge, given the wide-spread awareness regarding skin cancer. Therefore, it might not be surprising that sun exposure can also be damaging to hair fibres. In general, UV radiation degrades the hair through keratin damage and melanin oxidation [13]. A common finding is that sun exposure causes photobleaching, the fading of hair colour [14]. This process can be observed on the morphological level as the degradation of pigment granules in the cortex when hair has been deposited on the ground surface during a period of 12 months [15]. Darker hair is suggested to be better protected from ultraviolet damage than blonde hair, due to the amount of melanin sub-type eumelanin [13]. Simulated sun exposure using artificial UV sources has shown that irradiation can also lead to cavities and fractures in the cuticle [16]. Furthermore, the cells of the cuticle will be lifted, which disrupts the hydration balance and results in the hair shaft becoming more brittle. This has been investigated using an amount of artificial radiation that simulated a two month period of daily sun exposure in Brazil [16]. Obviously, it is expected that photodegradation is highly depended on the amount and intensity of UV irradiation per geographic location.

According to Lee [17], visible light also causes degradation of the pigment granules, regardless of which melanin subtype is more abundant. However, the effects of visible light on the cuticle are much less severe than those of UV light.

8

3.2 Water pH

Mills et al. [18] investigated the effect of different pH levels by submerging multiple hair fibres in water containing a pH level ranging from 3 to 11 over the course of 12 weeks. Using a light microscope, the physical characteristics of the hair were analysed and compared to those of hair fibres that had not been submerged. It was found that hairs exposed to a more acidic or neutral pH level showed minimal degradation, whereas hairs exposed to a basic pH presented a discoloration at pH 9 and a degraded cuticle at pH 11 [18].

These results are not surprising, since the pH of skin is also slightly more acidic (pH ± 5-6) and thus less degradation is expected in a more acidic environment [19]. Similarly, many studies have reported the damaging effects of using sodium hydroxide in cosmetic hair treatment, which of course also results in a basic pH value [15]. The study is however lacking in the amount of detail provided, since the amount of hair fibres analysed is not explicitly specified and the manner of analysis is rather subjective and prone to bias.

3.3 Microbial Alteration

Keratinolytic microorganisms produce the keratin-destroying enzymes called keratinases. These enzymes are proteases that specifically target keratin by catalysing the hydrolysis of the disulphide bonds which are abundantly present in keratin [20]. By doing so, microorganisms obtain their nutrients. Therefore, because of these specific enzymes, it is not surprising that keratin-containing structures of the hair are more susceptible to microbial digestion than other structures, whereas pigment granules prove to be relatively robust [15].

Through the degradation of keratin, ammonia will be released. As a result, the pH value of the environment will increase and thus become more basic [21]. As stated earlier, basic environments also cause damage and thus an additional level of degradation is created. Furthermore, a higher pH value has also been suggested to increase keratin degradation as it alters cysteine residues of keratin [20]. As a result, the accessibility of the protein for keratinases is increased.

On the morphological level, bacterial activity results in small pits on the hair shaft surface (Figure 3A) [12]. Fungi on the other hand, are described to have preferred manners of digestion, depending on their species. While some species digest from the outside in, others invade the hair shaft and digest from the inside out, starting with either the cortex or medulla [12]. Before fungal infestation takes place, fungal mycelia grow and spread on the hair shaft surface, regardless of the whether the fungi are capable of digesting keratin [12]. As a result, the overlapping cells of the cuticle are lifted, essentially allowing the fungi to perforate the cuticle and enter the cortex. In case of keratinolytic fungi, the so-called tunnelling into the cortex can now occur [15]. This describes the process in which the fungal hyphae bore transversally into the cortex due to their release of keratinases (Figure 3B-C). In a more advanced stage, hyphae will start to tunnel parallel to the hair axis. This allows the fungus to spread through the entire length of the cortex without having to degrade the outer cuticle first. Due to the hollowing out from inside, the tunnelling process eventually leads to the collapse

9 of the hair fibre. DeGaetano et al. [22] analysed the hairs of a murder victim who had been

buried for three weeks. While the hair seemed clean and intact with the naked eye, microscopy revealed transversal fungal tunnelling in the cortex. This means that the fungal attack was still in its early stages, suggesting that hair fibres are likely to survive in soil during three weeks. Of course, it should be noted that the activity of fungi and their enzymes are also under the influence of environmental factors such as temperatures or pH. In this case, however, those factors were unknown. Then again, many species of fungi exist, which all have their own preferred environments. It is thus hard to tell what kind of environments will cause keratinolytic fungi to proliferate.

Another way to analyse the effects of microbial alteration has been proposed by Domzalski [11]. Here, hair fibres were placed into jars of either soil or water. These environments were then sterilized by autoclaving and stored for three weeks. It was hypothesized that eliminating microbial alteration by sterilizing the environment would delay hair degradation. Microscopic analysis revealed that those hairs that were stored in the sterile soil, did not show any signs of degradation. On the other hand, the hairs that were stored in the non-sterile soil showed fungal tunnelling in the shaft, as well as erosion at the root end. In the case of water immersion, moderate to advanced changes were observed regardless of whether the water was sterilized. However, it did appear that roots in telogen phase were more sensitive to degradation in non-sterile water compared to roots in anagen phase.

3.4 Insect Activity

Bacteria and fungi are not the only living organisms to feed on the hair. Specialized insects are also described to play a role in hair degradation, mainly clothes moths larvae and carpet beetles. The common names of these insects already indicates that they are able to use keratin from hair and wool.

The larvae of clothes moths (Tineidae species) are known to be able to digest keratin also through keratinases and combined with a low redox potential in the gut [23]. With the use of their mandibles, specific bitemarks can be seen on the hair shaft (Figure 3D) [12]. Similar bitemarks are observed in case of carpet beetles (Dermestidae species). Thus, whereas the hair shaft is mostly digested from the inside out by fungi, the outside is attacked by bacteria, beetles and larvae.

10

Figure 3. Damage to the hair shaft due to microbial and insect activity.

A) Scanning Electron Micrograph (SEM) of microscopic pits on the cuticle due to bacterial activity. B) SEM of fungal tunnelling as seen from outside. Note the intact cuticle and the slight grooving at the top, indicating fibre collapse due to degradation of the cortex. C) Transmitted light micrograph of extensive transversal fungal tunnelling. D) SEM of bitemarks corresponding to insect activity such as from carpet beetles or clothes moths as seen from the front and the side. Pictures A, B, C and D are adapted from [12], [15], [22] and [14], respectively.

4. Metabolites in Mummified Remains

Not only the previously described environmental factors are of interest when it comes to the detection of hair metabolites, but also the passage of time itself. It is important to know whether hair metabolites remain stable enough to be detected in aged samples, when it is lucky enough to survive in the first place. On the morphological level, hair appears to be quite stable when protected from the environment. Hair samples from remains entombed in a continental climate were taken up to 25 years after death [24]. The study claimed that even after five years of burial, no morphological changes are observed. After 15 years, the cuticle is

11 observed to be partially lost and after 20 years the cortex is entirely exposed. At the same time,

the medulla is degraded and finally, at 25 years, major holes in the hair shaft are present. However, hair fibres can be preserved for much longer periods of time. This is especially the case when bodies become mummified in dehydrating environments due to extreme heat or cold. The hair of mummified (human and non-human) remains have sparked a lot of interest in the fields of archaeology, cultural anthropology and palaeontology. These fields have focussed mainly on detecting substances such as drugs of abuse that could elucidate the habits of ancient populations. This section will discuss a selection of these studies to provide an overview on the substances found.

4.1 Drugs of Abuse

Common practice in the ancient Andean populations was chewing on Coca leaves, from which cocaine can be obtained. This was done for ritualistic and medicinal purposes. Therefore, the hair of mummified remains found in the South Americas has been analysed extensively by many research groups focussing on the detection of cocaine and its metabolites benzoylecgonine (BZE) and, to a much lesser extent, ecgonine methyl ester (EME) (for an overview, see [25]). Estimated time periods range from 3000 BC to 1520 CE. It appeared that cocaine and its metabolites could still be detected in these hair samples. Overall, these studies only focussed on either cocaine or BZE. However, some studies included both parent drug (in this case cocaine) and its metabolites. Wilson et al. [26] investigated metabolites in the hair of mummified Inca children dated from the 15th _{century, thought to have been sacrificed. Levels} of cocaine detected were significantly higher than the levels of BZE. Interestingly, this study also detected cocaethylene, which is formed in the presence of alcohol, in one of the Inca children, showing that alcohol intake significantly increased shortly before the child’s death [26].

Another study, performed by Springfield et al. twenty years earlier [27], analysed hair samples from Peruvian mummies, dated 1000 CE. This study found that the metabolite levels in the hair of mummified remains were lower than those often reported in studies investigating present-day samples [27]. This could either be due to breakdown of the metabolites or perhaps due to the fact that the route of administration for cocaine is differently nowadays in modern western culture. The contrasting result with the previous study described, however, is that in this case the metabolite levels were higher than the level of the parent drug. Not only does this contrast the study by Wilson et al. [26], but with overall findings generally found within toxicological hair analysis [28]. The authors argue that this may be due to the metabolites having longer half-lives and thus they accumulate more easily. This is however not a very convincing argument as this should then also be the case in other studies, which should then have obtained similar results. Nevertheless, the study did raise an important issue regarding sample preparation, by pointing out that the washing procedure could negatively affect the extraction efficiency due to the aged hairs being more porous.

Regarding other drugs of abuse, Nowotnik et al. [29] investigated opiate usage in the hair of a known drug user whose body was found in his apartment two years after death.

12 Although not an archaeological finding such

as those from other studies discussed here, the body showed extreme mummification and was completely desiccated. Interestingly, only a single hair was tested, although pictures show more material available (Figure 4). It was found that the concentration of morphine (0.41 ng/mg; morphine is both a parent drug as well as a metabolite of heroin and codeine) was similar to those found in hair samples obtained from non-mummified or living individuals (e.g. 0.3-1.3 ng/mg, [30]). However, concentrations of heroin-specific metabolite 6-Monoacetylmorphine (6-MAM; 0.15 ng/mg) and codeine (0.05 ng/mg) were lower than those reported in literature (e.g. 0.3-7.4 ng/mg [30] and 0.15-1.87 ng/mg [31],

respectively). Given the sample size of a single hair, the question remains if this is due to variability or due to degradation resulting from the mummification process.

4.2 Habits

On behalf of habits, nicotine (11.4-57.5 ng/mg, [32]) has been found in the hair of pre-Columbian mummies in concentrations similar to those found in present-day samples from active smokers (mean ± SEM; 24.7 ± 2.97, [33]) . However, these findings should be interpreted with caution, due to the fact that the metabolite of nicotine, namely cotinine, could not be detected. Normally, the presence of cotinine in hair is an indicator of either active tobacco use or passive smoke inhalation. However, even in modern forensic toxicology, cotinine levels are lower than those of nicotine [34]. Thus, a negative result for cotinine could mean that the amount of cotinine was simply too low to be detected, but also gives rise to the possibility that the nicotine levels found are the result of external contamination, rather than smoke inhalation. Given that the hair is damaged and thus more porous, the hair is simply more prone to environmental contamination as compounds from outside smoke can be more easily incorporated [35]. The findings here, however, are supported by the fact that traces of nicotine have not been found in the washing solution, thus lowering the chance of contamination [32]. The study indicates a main issue of working with mummified remains, which is that it is simply impossible to know how the hair has been treated over time. In this case, nicotine could have been incorporated due to active tobacco usage, medicinal rituals using only tobacco smoke, or even smoking museum visitors.

The findings are supported by the results of another study by the same author [36]. Here, the hair of Italian mummified remains from the late 19th _{century has been used and} whereas nicotine could be detected in all hair samples, cotinine was only detected in half of the samples. Similar to the previous study, this means that nicotine incorporation due to

Figure 4. Remaining scalp hair of modern mummified remains.

Mummified remains of a known drug user found two years post-mortem in his home. As can be seen, multiple strands of hair could be recovered, yet Nowotnik et al. [29] only analysed a single hair for the detection of opiates. Source: [29].

13 contamination cannot be ruled out. It was also found, however, that caffeine could also be

detected in the hair shafts. This compound might be more reliable as caffeine cannot contaminate the hair through the air surrounding it. Yet, the relevance of caffeine in a modern day forensic context may be more questionable, as caffeine consumption is a general habit among the majority of adults and thus not a very specific searchable trait.

Finally, Musshoff et al. [37], determined the presence of Ethyl Glucuronide (EtG) in the hair of mummies from the catacombs of Palermo (19th _{century AD). EtG is a metabolite of} ethanol and thus an indicator of alcohol consumption. It was found that among the 32 adult individuals tested, alcohol consumption 2,5 months prior to their death could be indicated in 15 of them.

4.3 Hormones

Hormones have been investigated using remains from an Egyptian cemetery (50-450 CE) [38]. These hairs have also been preserved in a desert environment and samples were taken from five male and five female subjects. At the same time, samples from a modern population were taken. It appeared that in general, cortisol (272.5-467 ng/g), testosterone (3.37-23.34 ng/g) and estradiol (0.04-0.22 ng/g) could still be detected in concentrations not significantly different to those found in the present-day samples (cortisol: 190.57-348.2 ng/g, testosterone: 3.35-20.08 ng/g, estradiol: 0.02-0.07 ng/g) [38]. Regarding the gonadal steroids, estradiol concentrations were found to be higher in female subjects than in male subjects, as expected. However, the differences were not statistically significant. Oddly enough, the authors draw a wrongful conclusion in the discussion, stating that the difference was actually significant and thus an explanation for the finding is not provided. Interestingly, estradiol concentrations in the archaeological remains were found to be lower than those in present-day samples. Given the estimated age of the subjects taken the hair from, the authors speculate that this may not be due to degradation, but due to the presence of two populations: pre- and postmenopausal.

For testosterone, on the other hand, male subjects’ hair contained significantly higher levels of testosterone than the hair of female subjects. Out of the five female subjects, two of them showed higher testosterone concentrations in their hair than the others. These two females were different from the others as they were suggested to be pregnant prior to their death as pointed out by isotopic analysis. One of them was even found to be buried with a 38-week-old fetus.

As a side note, it should be stated that whereas most studies analysing hair samples are based on GC-MS, this study used a commercial ELISA designed for the detection of these hormones in saliva. It could thus be questioned how reliable or accurate this manner of detection is compared to the commonly used GC-MS approach. Although it does seem to have been used by other research groups investigating cortisol and other steroids, the poor accuracy of immunoassays to detect low testosterone levels in serum samples is apparently still an issue, as it has been pointed out by a recent study published this January [39].

14 Nevertheless, the findings appear promising. In a whole different environment than the

Egyptian desert, testosterone has also been detected in even older hair samples that had been preserved in permafrost, namely in that of the extinct Siberian woolly mammoth (dated 10.000 to 60.000 years old) [40]. This indicates that hormones in hair can be well preserved in both hot and cold conditions as long as liquid water is unavailable.

This section has discussed various papers regarding metabolites in the hair of mummified remains. These remains were mummified in both hot and cold environments and have been discovered up to many decennia ago to relatively recent. An overview of the studies discussed and the success rate of detecting the metabolites involved can be seen in table 1. As can be seen, in the majority of the cases, the detection of analytes of interest is well-above 50% and samples confirm the hypotheses (such as the usage of drugs of abuse) of the studies discussed.

Table 1. Overview of metabolites found in the hair of mummified remains.

Compounds/metabolites from the studies discussed are presented. The number of mummified individuals is compared with the percentage of those whose hair resulted in a positive outcome for the compound/metabolite tested.

Compound/Metabolite # Individuals

tested

% Positive Estimated time period Study

Stimulants Cocaine 10 100% 1100-1250 CE [27] 3 100% 1430-1520 CE [26] 6 33% 1910 CE [36] Cocaethylene 3 100% 1430-1520 CE [26] Benzoylecgonine 10 100% 1100-1250 CE [27] 3 100% 1430-1520 CE [26] 6 33% 1910 CE [36]

Ecgonine Methyl Ester 10 20% 1100-1250 CE [27]

Opioids Codeine Morphine 1 100% 2010 CE [29] 6-Monoacetylmorphine Habitual Caffeine 6 83% 1910 CE [36] Ethyl glucuronide 32 48% 1800-1900 CE [37] Nicotine 3 100% 1045-1135 CE [32] 6 100% 1910 CE [36] Cotinine 3 _{6 50%} 0% 1045-1135 CE _{1910 CE} [32] _[36] Hormones Cortisol Estradiol 10 100% 50-450 CE [38] Testosterone

15

5. Findings from Forensic Casework and Experimental Data

While the discussed findings from mummified remains are a relevant indicator of the stability of hair metabolites over time, forensic casework regarding donor profiling is certainly not expected to have to take such a large time-frame and extreme conditions into account. Therefore, to focus more on decay itself, rather than time, this section will discuss experimental studies in which hair fibres are actively degraded or studies investigating the hair from bodies retrieved from water or soil within a smaller time-frame.

5.1 Hair metabolites after exposure to ultraviolet radiation

As stated above in section 3.1, UV light can have a damaging effect to the hair shaft and it could be argued to be one of the most relevant factors, as hairs on the crime scene are probably left on the ground surface. Furthermore, both UV (including both of its subtypes UVA and UVB) as visible light played a role in hair decay [17], highlighting the relevance of this factor regardless of whether the hair has been shed inside or outside.

To start with, cannabis-related metabolites have been found to decrease after sun exposure [41]. Cannabis-positive hairs were either placed in the dark (control) or exposed to natural sunlight for 10 weeks. It was found that the concentrations of tetrahydrocannabinol (THC), cannabidiol (CBD) and cannabinol (CBN) were greatly reduced in the sun-exposed hair strands compared to the control strands. Even much so, that after sun exposure, almost 80% of the concentrations were below the detection limit of 0.05 ng/mg. Skopp et al. [41] suggested that light catalyses a oxidation reaction that causes THC to degrade to CBD. However, CBD is in turn further degraded due to the generation of reactive oxygen species by melanin.

Other studies regarding drugs of abuse have indicated that also cocaine and its metabolite BZE are affected by UV irradiation, as well as opiates dihydrocodeine, morphine and heroin-metabolite 6-MAM [42, 43]. It appeared that cocaine was less sensitive to photodegradation than BZE. On average, photodegradation rates did not exceed values above 21% for the tested compounds [43]. It should be stated, however, that this study only tested the effect of UVB irradiation at 300 J/cm2_{. Since this is of course only part of the full UV radiation} present in natural sunlight, it is hard to determine to what level of sun exposure this amount of UVB irradiation compares. Combined with other variables, such as rain and wind, obtained opiate concentrations were generally further lowered [42].

Long-term UV radiation has also been found to affect steroid concentrations, by placing bundles of hair outside with either 1, 3, or 6 hours of sun exposure per day over the course of two summer months (Figure 5) [44]. Grass et al. [44] found a positive correlation between the

16 percentage cortisol concentrations below the

quantification limit and the amount of sun exposure per day. This same observation was made in case of cortisone, although this steroid proved to be more robust. The other two steroids that were tested, dehydroepiandrosterone and progesterone, appeared unaffected by UV radiation.

In a separate experiment, the same research group also investigated the effects of a single 1 or 5 hour exposure period to an artificial UV source, which led to the same conclusion [44]. A Dutch research group investigated the steroid hormones cortisol and cortisone, as well [45]. They found that after exposure to natural sunlight for 40 hours, cortisol concentrations dropped with 54%, whereas cortisone remained more stable and only decreased by 32%. These findings support those found by Grass et al. [44] described above. The study also tested the effects

of extreme artificial UV exposure, which was higher than ever recorded on earth. In this case, compared to the control which was kept in the dark, cortisol levels decreased by 75% and cortisone levels by 52%. While this amount of UV irradiation cannot occur in a real-life situation, it does provide an indicator of the maximum amount of steroid loss if considering UV irradiation alone.

5.2 Hair metabolites after exposure to soil and water

On behalf of hair in exhumed remains, Cippitelli et al. [46] tested the hair of highly putrefied remains which were buried for over a year at an Italian cemetery. This was due to the public prosecutor who wanted to know if the man’s death could be attributed to opiate abuse. The researchers took samples from hair, teeth, liver and kidney. It was found that morphine and 6-MAM could be detected in all tissues with codeine also being detected in the liver. Unsurprisingly, concentrations were highest in the liver and kidney, followed by hair and finally teeth. Yet, the body in this case was buried in a metal-shell coffin, hindering the influences of insect activity and air. Thus, hair degradation could be more limited.

Conflicting results were found in another study by Pötsch et al. [47]. They investigated the effects of soil and water on the detection of (metabolites of) opiates dihydrocodeine, morphine and 6-MAM. Hair samples from known opiate abusers were placed in water or soil for a prolonged period of time. It was found that after storage in water for four weeks, only dihydrocodeine could still be detected, whereas before exposure all compounds could be found. After six months in water, only negative results were obtained. Hair samples that were

Figure 5. Natural sunlight exposure for hair bundles.

Whereas some studies make use of jars or quartz glass vials (e.g. [37]), Grass et al. [40], used a hanging device to expose hair samples in the open air. While this might be more natural than a glass container, it does expose the hair samples to more environmental factors (wind and air pollution). Yet, the hair bundles were brought inside during rain. Source: [44].

17 placed in the soil for six months also only tested positive for dihydrocodeine and only one of

the two samples tested positive for 6-MAM.

As stated above, the opiates detected by Cippitelli et al. [46] originated from hair samples in a metal-shell coffin, which can partly explain their positive findings. The negative findings of Pötsch et al. [47] might be due to their very limited sample size or due to their methodology. Given the time difference between the two studies, it is assumed that the techniques used have improved their sensitivity over time. Nevertheless, the limit of detection of morphine in Pötsch et al. [47] is equal to those reported in recent articles, such as [48]. Regarding the negative results after water exposure, hair samples placed in water were kept in jars. It is unknown whether these were isolated from light sources and thus whether light could have increased the degradation process. Yet, it could of course also simply mean that water immersion negatively affects the survival of the tested metabolites.

More regarding water exposure, in 2015, the body of a woman was retrieved from her car at the bottom of the Como lake (120 meters deep). Her body had been immersed in water for three years before she was found. In the pubic hair of the woman low levels (97 pg/mg) of EtG (metabolite of ethanol), were detected [49]. In this case, while this finding does indicate that the woman has been drinking alcohol during the period prior to her death, it is impossible to determine the amounts of alcohol. While the woman was also described to have been a marijuana user, THC could not be detected in her hair. Interestingly enough, other marijuana-related metabolites such as CBD or CBN were not tested, while other drugs of abuse (such as opiates and amphetamines) were, even though there was no indication that the woman might have been using these while she was still alive. In any case, these also led to a negative result [49]. If true negatives, these findings would be in line with Pötsch et al. [47], suggesting that water exposure does indeed hinder metabolite detection of opiates.

5.3 Comparison with data obtained from mummified remains

So far findings from both contemporary as archaeological studies have been discussed and the effect of time itself and environmental factors on the hair metabolites have been described. Some metabolites have been detected in both the hair of both ancient as modern populations, on which table 2 provides an overview. It is chosen not to include cortisol, as this hormone is often used as a biomarker for chronic stress [38]. Since the overall human lifestyle has changed drastically over hundreds of years, it is questionable how comparable the stress experience is between ancient and modern populations. Furthermore, the concentrations of cortisol before degradation found by Wester et al. [45] are already much lower than those found by Tisdale et al. [38], again disallowing a meaningful comparison. Comparing the other findings, it appears that concentrations in hair samples from mummified remains are either higher than, or comparable with, those in hair samples obtained from contemporary hair samples, which are believed to be in a more degraded state given their environment. Yet, the studies differ greatly in their methodology. Whereas some have taken multiple hair samples (originating from different individuals), others have taken only one, by either taking a bulk amount of hair or a single to a few strands. As a result, the findings presented might be the result of a (un)lucky

18 sample and the possible variability of concentrations both within hair samples as individuals

remains unclear. To be applicable in a forensic context, this uncertainty might pose a problem, as will also be discussed in section 7.

Table 2. Comparison of findings of contemporary samples to those of mummified remains.

Compounds/metabolites discussed are presented. The environment from which each hair sample is retrieved, along with the amount of hair and number of samples is shown. Concentrations (ng/mg) and, in case of more than one sample, a standard error of the mean (SEM) are provided. *, infants not taken into account. N.d., not detected.

Compound/Metabolite Retrieval Hair

amount/sample # Samples Concentration (ng/mg) ± SEM, if applicable Study Cocaine Outside weather 40-50 mg 8 1.95 ± 0.50 [42] Mummified remains 0.2 mg 1 6.2 [26] Benzoylecgonine Outside weather 40-50 mg 8 0.44 ± 0.29 [42] Mummified remains 0.2 mg 1 3.9 [26] Codeine Soil 100 mg 1 n.d. [46] Mummified remains Single hair 1 0.05 [29] Morphine Soil 100 mg 1 0.42 [46] Mummified remains Single hair 1 0.41 [29] 6-Monoacetylmorphine Soil 100 mg 1 0.06 [46] Mummified remains Single hair 1 0.15 [29] Ethyl glucuronide Water 30 mg 1 0.097 [49] Mummified remains 75 mg 32* 0.06 ± 0.03 [37]

6. Conclusions

As a biological source, the hair shaft can provide information on the donor’s possible drug use, habits and sex. This can provide law enforcement with investigative leads and narrow down the number of possible suspects. It appears that the decay of hair fibres is under influence of multiple environmental factors, such as UV radiation, pH and microbial and insect activity. It is difficult to compare the different degradation patterns of the various factors to determine which is truly most damaging to the hair fibre, since, for example, it cannot be said how much fungal tunnelling is equally as destructive as a certain amount of photobleaching. Nevertheless, out of the variables discussed, it appears that microbial alteration might be the most damaging to the hair shaft. This is supported by the fact that the hair of mummified remains has been kept intact as those environments disallow microbial proliferation. Indeed, environments that promote desiccation and inhibit putrefaction and thus activity of microorganisms are most suitable to preserve hair fibres. Those dry environments, which are often accompanied by

19 extreme heat or cold, can then preserve the hair for hundreds of years. Additionally,

metabolites of interest can still be obtained from these hairs, often in concentrations that are comparable with those from present-day samples. Thus, it seems that not time, but rather the environmental influences play a bigger role in hair degradation. Regarding less extreme environments, only a limited amount of studies have investigated hair metabolites from fibres obtained in a more realistic setting. These studies mostly investigated hair samples that were actively degraded by UV irradiation or obtained from exhumed remains. These are the cases that are more likely to be encountered during forensic casework. It appears that in these hair fibres, too, metabolites could still be detected, albeit in lower concentrations. Taken together, the relevance of donor profiling in hair is emphasized as it appears to remain feasible even when samples are in a degraded state.

7. Discussion

It appears that metabolites in hair are generally stable over time. Yet, a relevant question to be asked is whether this is indeed the case, or if negative findings simply don’t get published. Given the amount of studies on mummy hair, it appears that hair metabolites are indeed stable and thus, a publication bias seems unlikely. However, obtaining negative results when studying mummified remains, does give rise to another issue, namely that of interpretation. If negative results are found, it remains unclear whether this is due to degradation of the metabolites or simply because the metabolite was not present to begin with. In other words, the true and false negatives cannot be distinguished here. This issue is also raised by [50], who failed to detect multiple opiates, cannabis-related and cocaine-related metabolites in well-preserved prehistoric hair (100 BC). Perhaps the tested individuals never consumed any of these drugs, or maybe the method just failed to detect them possibly due to the aging of the hair. The answer to this problem is impossible to know as empirical studies simply cannot test the effect of hundreds of years of time on the hair fibres.

Of course it could be argued that a similar issue is always true for hair fibres found on the crime scene in a smaller time-frame, since the donor and their habits are unknown. The point is, however, that studies regarding the error rate in metabolite analysis are lacking. To be used successfully as an investigative lead, the forensic field needs more information regarding accuracy and reliability of the method when using realistic (i.e. sub-optimal) samples. It is important to know how much variability, both inter- and intra-individual, can be expected. As already stated in section 5.3, many studies analyse a very limited amount of hair samples and pose the closed question whether a certain metabolite can be detected, rather than how accurate their outcome is. Yet, while this may be of less importance in case of xenobiotics (i.e. compounds not naturally present in the human body), it is most important when investigating hormonal compounds which are present in both sexes. Moreover, knowing variability on compounds easily incorporated through external contamination, such as nicotine from cigarette smoke, is also beneficial as it helps to distinguish habit (smoking) from environment (smoke). Furthermore, more experimental data could in this case perhaps even be used as an indicator of whether the donor has been an active or passive smoker.

20 Nevertheless, studies investigating metabolites in hair obtained from exhumed bodies

might still be the most relevant at the moment, as the undisturbed environment may be the most realistic and the relatively warm and humid environment within the coffin is favourable for microorganisms to grow. This is an important aspect to highlight if we conclude that this very microbiome is the most relevant variable for hair decay. Yet, it should be questioned how often hair fibres are actually obtained that were immersed in water or buried deep within the soil. It seems more common sense that hairs are recovered from the surface and thus UV radiation might have more forensic relevance, even if it is less damaging than the microbiome. Furthermore, the microbiome mainly plays a role if recovered from outside. If the hair has been shed inside, it is even further expected to survive well.

Regarding which compounds are of interest, for archaeological purposes, since it is unknown what has happened to the hair over many decades, it is recommended to analyse metabolites instead of the parent drug, since it better indicates active usage. However, it should be noted that some metabolites still remain more suitable than others. For example, cocaine-metabolite EME is formed through active metabolism by enzymes, whereas BZE can be formed through spontaneous hydrolysis, thus without enzymatic activity. As such, while BZE does in fact refer from cocaine, it does not have the added value of distinguishing active usage versus contamination, like EME does. The recommendation stated above also applies to the forensic field, but to a lesser extent. This is due to the smaller time-frame and the fact that the hair fibres are not expected to have undergone certain treatments in the past like those of museum specimens.

On behalf of the future, it has already been stated that studies with a clear forensic focus are lacking. More information is desired on the error rate of the analysis and more studies that simulate realistic, common, environments are required. This also ties in with another common issue; as it is expected to find a very limited amount of hair on the average crime scene, how much more sensitive do the techniques need to be to also be applicable when the hair is limited and partially degraded? In any case, the issue of sensitivity and determining how much hair is actually needed for reliable testing is not restricted to degraded hair only, but rather applies to donor profiling of hair as a whole. Furthermore, it would be valuable for studies to investigate metabolites that result in searchable traits that can be more generally applied. Of course, it is valuable to know whether the donor is a drug user, but as of now there seems to be an abundant focus on drugs of abuse, mainly opiates and cocaine. While this is a searchable trait, it may be too specific. In other words, if a crime is not drug-related, what are the odds that an actual drug addict shed the hair during the criminal act compared to any other non-addict? Therefore, metabolites originating from over-the-counter and prescriptive drugs may be more relevant as they are expected to be more widespread and the usage of these drugs is better documented. Furthermore, more research into hormonal metabolites to determine the sex of the donor could provide a way to narrow down the list of possible suspects in half.

Nevertheless, hair and its metabolites have shown to be stable over time, when relatively protected from the environment. Given the literature discussed, it is not expected to find hair fibres in a highly degraded state within the forensic time-frame. This makes donor

21 profiling of hair a promising tool to obtain investigative leads even when hair fibres are found

in environments which would be considered to be sub-optimal.

References

[1] M. Usman, A. Naseer, Y. Baig, T. Jamshaid, M. Shahwar, and S. Khurshuid, ‘Forensic toxicological analysis of hair: a review’, Egypt. J. Forensic Sci., vol. 9, no. 1, p. 17, 2019.

[2] R. Paus, ‘Principles of hair cycle control’, J. Dermatol., vol. 25, no. 12, pp. 793–802, 1998.

[3] S. L. Koch, A. L. Michaud, and C. E. Mikell, ‘Taphonomy of hair—a study of postmortem root banding’, J. Forensic Sci., vol. 58, pp. S52–S59, 2013.

[4] R. Dhurat and P. Saraogi, ‘Hair evaluation methods: merits and demerits’, Int. J. Trichology, vol. 1, no. 2, p. 108, 2009. [5] J. M. Taupin, ‘Forensic hair morphology comparison–a dying art or junk science?’, Sci. Justice J. Forensic Sci. Soc., vol.

44, no. 2, pp. 95–100, 2004.

[6] A. Hellmann, U. Rohleder, H. Schmitter, and M. Wittig, ‘STR typing of human telogen hairs–a new approach’, Int. J. Legal Med., vol. 114, no. 4–5, pp. 269–273, 2001.

[7] W. A. Baumgartner, V. A. Hill, and W. H. Blahd, ‘Hair analysis for drugs of abuse’, J. Forensic Sci., vol. 34, no. 6, pp. 1433–1453, 1989.

[8] P. Xiang, M. Shen, and X. Zhuo, ‘Hair analysis for ketamine and its metabolites’, Forensic Sci. Int., vol. 162, no. 1–3, pp. 131–134, 2006.

[9] Y. Zhuo et al., ‘Simultaneous quantitative determination of amphetamines, opiates, ketamine, cocaine and metabolites in human hair: application to forensic cases of drug abuse’, J. Forensic Sci., 2019.

[10] M. D. Gomez-Roig et al., ‘Maternal hair testing to disclose self-misreporting in drinking and smoking behavior during pregnancy’, Alcohol, vol. 67, pp. 1–6, 2018.

[11] A. C. Domzalski, ‘The effects of environmental exposure on human scalp hair root morphology’, PhD Thesis, John Jay College of Criminal Justice New York, NY, 2004.

[12] S. R. Tridico, S. Koch, A. Michaud, G. Thomson, K. P. Kirkbride, and M. Bunce, ‘Interpreting biological degradative processes acting on mammalian hair in the living and the dead: which ones are taphonomic?’, Proc. R. Soc. B Biol. Sci., vol. 281, no. 1796, p. 20141755, 2014.

[13] G. Miolo and D. Favretto, ‘Photodegradation of Drugs of Abuse in Hair’, in Light in Forensic Science, 2018, pp. 399– 422.

[14] A. S. Wilson, ‘The decomposition of hair in the buried body environment’, in Soil Analysis in Forensic Taphonomy, CRC Press, 2008, pp. 135–164.

[15] A. S. Wilson, H. I. Dodson, R. C. Janaway, A. M. Pollard, and D. J. Tobin, ‘Selective biodegradation in hair shafts derived from archaeological, forensic and experimental contexts’, Br. J. Dermatol., vol. 157, no. 3, pp. 450–457, 2007.

[16] M. Richena and C. A. Rezende, ‘Morphological degradation of human hair cuticle due to simulated sunlight irradiation and washing’, J. Photochem. Photobiol. B, vol. 161, pp. 430–440, 2016.

[17] W.-S. Lee, ‘Hair photoaging’, in Aging hair, Springer, 2010, pp. 123–133.

[18] T. Mills, S. K. Jasra, and P. Jasra, ‘The Degradative Characteristics of Hair in The Water Samples with Altered pH’, J. Emerg. Forensic Sci. Res., vol. 2, no. 1, pp. 22–31, 2017.

[19] M. Iguchi and H. Tagami, ‘Functional analysis of the stratum corneum of scalp skin: studies in patients with alopecia areata and androgenetic alopecia’, Arch. Dermatol. Res., vol. 292, no. 12, pp. 605–611, 2000.

[20] R. Gupta and P. Ramnani, ‘Microbial keratinases and their prospective applications: an overview’, Appl. Microbiol. Biotechnol., vol. 70, no. 1, p. 21, 2006.

[21] E. Bach, F. C. Lopes, and A. Brandelli, ‘Biodegradation of α and β-keratins by Gram-negative bacteria’, Int. Biodeterior. Biodegrad., vol. 104, pp. 136–141, 2015.

[22] D. H. DeGaetano, J. B. Kempton, and W. F. Rowe, ‘Fungal tunneling of hair from a buried body’, J. Forensic Sci., vol. 37, no. 4, pp. 1048–1054, 1992.

[23] A. C. Cohen, Insect diets: science and technology. CRC press, 2015.

[24] B. S. Chang et al., ‘Ultramicroscopic observations on morphological changes in hair during 25 years of weathering’, Forensic Sci. Int., vol. 151, no. 2–3, pp. 193–200, 2005.

[25] E. L. Brown and A. S. Wilson, ‘Using evidence from hair and other soft tissues to infer the need for and receipt of health-related care provision’, Int. J. Paleopathol., vol. 25, pp. 91–98, 2019.

[26] A. S. Wilson et al., ‘Archaeological, radiological, and biological evidence offer insight into Inca child sacrifice’, Proc. Natl. Acad. Sci., vol. 110, no. 33, pp. 13322–13327, 2013.

[27] A. C. Springfield, L. W. Cartmell, A. C. Aufderheide, J. Buikstra, and J. Ho, ‘Cocaine and metabolites in the hair of ancient Peruvian coca leaf chewers’, Forensic Sci. Int., vol. 63, no. 1–3, pp. 269–275, 1993.

22

[29] J. Nowotnik, D. Rentsch, A. Bittorf, I. Lindner, and A. Büttner, ‘Mumie ohne Geheimnisse–Identifizierung, Eingrenzung der Liegezeit und Nachweis von Morphin’, Rechtsmedizin, vol. 23, no. 6, pp. 485–491, 2013.

[30] R. Kronstrand, R. Grundin, and J. Jonsson, ‘Incidence of opiates, amphetamines, and cocaine in hair and blood in fatal cases of heroin overdose’, Forensic Sci. Int., vol. 92, no. 1, pp. 29–38, 1998.

[31] P. Mangin and P. Kintz, ‘Variability of opiates concentrations in human hair according to their anatomical origin: head, axillary and pubic regions’, Forensic Sci. Int., vol. 63, no. 1–3, pp. 77–83, 1993.

[32] F. Musshoff, W. Rosendahl, and B. Madea, ‘Determination of nicotine in hair samples of pre-Columbian mummies’, Forensic Sci. Int., vol. 185, no. 1–3, pp. 84–88, 2009.

[33] M. Tsuji et al., ‘Nicotine cut-off value in human hair as a tool to distinguish active from passive smokers: A cross-sectional study in Japanese men’, Cancer Biomark., vol. 20, no. 1, pp. 41–48, 2017.

[34] W. K. Al-Delaimy, ‘Hair as a biomarker for exposure to tobacco smoke’, Tob. Control, vol. 11, no. 3, pp. 176–182, 2002.

[35] J. Thorspecken, G. Skopp, and L. Pötsch, ‘In vitro contamination of hair by marijuana smoke’, Clin. Chem., vol. 50, no. 3, pp. 596–602, 2004.

[36] F. Musshoff, H. Fels, A. Carli, and D. Piombino-Mascali, ‘The anatomical mummies of Mombello: detection of cocaine, nicotine, and caffeine in the hair of psychiatric patients of the early 20th century’, Forensic Sci. Int., vol. 270, pp. 20–24, 2017.

[37] F. Musshoff, C. Brockmann, B. Madea, W. Rosendahl, and D. Piombino-Mascali, ‘Ethyl glucuronide findings in hair samples from the mummies of the Capuchin Catacombs of Palermo’, Forensic Sci. Int., vol. 232, no. 1–3, pp. 213– 217, 2013.

[38] E. Tisdale, L. Williams, J. J. Schultz, and S. M. Wheeler, ‘Detection of cortisol, estradiol, and testosterone in archaeological human hair from the Dakhleh Oasis, Egypt’, J. Archaeol. Sci. Rep., vol. 27, p. 101968, 2019. [39] D. French, J. Drees, J. A. Stone, D. T. Holmes, and J. G. van der Gugten, ‘Comparison of four clinically validated

testosterone LC-MS/MS assays: Harmonization is an attainable goal’, Clin. Mass Spectrom., vol. 11, pp. 12–20, 2019. [40] L. Koren et al., ‘Testosterone in ancient hair from an extinct species’, Palaeontology, vol. 61, no. 6, pp. 797–802,

2018.

[41] G. Skopp, L. Pötsch, and M. Mauden, ‘Stability of cannabinoids in hair samples exposed to sunlight’, Clin. Chem., vol. 46, no. 11, pp. 1846–1848, 2000.

[42] G. Skopp, L. Pötsch, and M. R. Möller, ‘Zum suchtmittelnachweis in haaren V. Auswirkung von sonne, regen und wind auf den drogengehalt in kopfhaaren von drogenkonsumenten–ein pilotprojekt’, Rechtsmedizin, vol. 7, no. 6, pp. 176– 179, 1997.

[43] D. Favretto, M. Tucci, A. Monaldi, S. D. Ferrara, and G. Miolo, ‘A study on photodegradation of methadone, EDDP, and other drugs of abuse in hair exposed to controlled UVB radiation’, Drug Test. Anal., vol. 6, no. S1, pp. 78–84, 2014.

[44] J. Grass et al., ‘In vitro influence of light radiation on hair steroid concentrations’, Psychoneuroendocrinology, vol. 73, pp. 109–116, 2016.

[45] V. L. Wester, N. R. van der Wulp, J. W. Koper, Y. B. de Rijke, and E. F. van Rossum, ‘Hair cortisol and cortisone are decreased by natural sunlight’, Psychoneuroendocrinology, vol. 72, pp. 94–96, 2016.

[46] M. Cippitelli, D. Mirtella, G. Ottaviani, G. Tassoni, R. Froldi, and M. Cingolani, ‘Toxicological analysis of opiates from alternative matrices collected from an exhumed body’, J. Forensic Sci., vol. 63, no. 2, pp. 640–643, 2018.

[47] L. Pötsch, G. Skopp, and J. Becker, ‘Ultrastructural alterations and environmental exposure influence the opiate concentrations in hair of drug addicts’, Int. J. Legal Med., vol. 107, no. 6, pp. 301–305, 1995.

[48] H. Khajuria and B. P. Nayak, ‘Detection and accumulation of morphine in hair using GC–MS’, Egypt. J. Forensic Sci., vol. 6, no. 4, pp. 337–341, 2016.

[49] L. Morini, C. Vignali, P. Tricomi, and A. Groppi, ‘Analytical challenge in postmortem toxicology applied to a human body found into a lake after three years immersion’, J. Forensic Sci., vol. 60, no. 5, pp. 1383–1386, 2015.

[50] H. Baez et al., ‘Drugs in prehistory: chemical analysis of ancient human hair’, Forensic Sci. Int., vol. 108, no. 3, pp. 173–179, 2000.

23

Appendix A - Search strategy

General articles on hair and taphonomy of hair were the starting points. After searching by using general terms such as hair, degradation and decay, I followed with more specific searches using keywords like fungi, moisture/water and UV light to find more about possible variables. After determining which variables to discuss, specific articles focussing on hair in aged remains (such as mummies) and hair metabolites were searched for. This was proven rather difficult, as articles investigating hair samples obtained from mummified remains appeared to be quite old at first and did not go into the metabolites found. Furthermore, recent articles only seemed to analyse present-day hair samples. Another problem encountered was that many articles weren’t about the forensic field, but rather had a different focus (e.g. archaeology, palaeontology or clinical).

I also decided to focus on author Wilson who does more research on hair in a forensic context, but found mostly articles regarding mitochondrial DNA, which was outside of my scope. Therefore, I changed my search strategy once I found the term xenobiotics, which describes compounds naturally not present in an organism.

On a second round of searching, more recent articles determining the presence of specific compounds in mummified remains were found by setting a fixed time-span. The focus while searching was mainly on hormones or drugs of abuse. The problem then encountered, however, was the overwhelming amount of papers describing the use of hair in present-day drug testing. Papers which investigate compounds in hairs that have been affected by the environment seem scarce. To find more, I used synonyms such as metabolites, analytes,

compounds, etc. To focus more on degraded hair outside of mummification, I included key

words such as post mortem, putrefaction, or buried.

After these rounds, I started writing to find out where I am missing sources and to pick up sources through the articles already found.