(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Unravelling taphono-myths. First large-scale study of histotaphonomic changes and diagenesis in bone from modern surface depositions [1]

['Eline M. J. Schotsmans', 'Environmental Futures Research Centre', 'School Of Earth', 'Atmospheric', 'Life Sciences', 'University Of Wollongong', 'Wollongong', 'Pacea', 'De La Préhistoire À L Actuel', 'Culture']

Date: 2024-10

Abstract The use of diagenetic alterations in bone microstructure (‘histotaphonomy’) as indicators of funerary treatment in the past and for post-mortem interval calculations in forensic cases has received increasing attention in the last decade. Studies have used histological changes to conclude in-situ decomposition, mummification, infanticide and post-mortem interval. There has been very little attempt to experimentally validate the links between decomposition, depositional conditions, time-since-death and microscopic changes in human bone so that meaningful interpretations of archaeological and forensic observations can be made. Here, we address this problem experimentally using the largest sample of human remains from anatomical donors and the longest-term deposition framework to date. This study tests one key assumption of histotaphonomy; that putrefaction during the early stages of decay is reflected in bone microanatomy and composition. Seventeen human donors and six pigs were deposited on the surface in a known Australian environment and left to decompose between 463 and 1238 days. All remains underwent all stages of decomposition reaching skeletonisation. Rib and femur samples were analysed using conventional histological methods and scanning electron microscopy, by applying the Oxford Histological Index, and examining collagen birefringence, microcracking and re- and de mineralisation. Biomolecular changes of the femoral samples were analysed using Fourier-transform infrared (FTIR) spectroscopy. The results indicate that bioerosion in human bone does not occur due to putrefaction. There were no correlations between bone histology and the following variables: human vs pigs, season, primary vs secondary deposition, position, fresh vs frozen and time-since-deposition. Furthermore, no trends were observed between biomolecular changes and time-since-deposition. The study also shows that pigs cannot be used as substitutes for human remains for bone biodegradation research. This is the first, controlled, larger scale study of human remains providing a lack of support for a long-assumed relationship between putrefaction and bone histology bioerosion. Using bone degradation as an argument to prove putrefaction, in-situ decomposition and early taphonomic processes cannot be supported based on the experimental human data presented.

Citation: Schotsmans EMJ, Stuart BH, Stewart TJ, Thomas PS, Miszkiewicz JJ (2024) Unravelling taphono-myths. First large-scale study of histotaphonomic changes and diagenesis in bone from modern surface depositions. PLoS ONE 19(9): e0308440. https://doi.org/10.1371/journal.pone.0308440 Editor: Dong Hoon Shin, Seoul National University College of Medicine, REPUBLIC OF KOREA Received: March 21, 2024; Accepted: July 23, 2024; Published: September 26, 2024 Copyright: © 2024 Schotsmans et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data are within the manuscript and its Supporting Information files. Funding: This study was supported by the Australian Research Council with grants to EMJS [DE210101384], JJM [DE190100068], and the Australian Facility for Taphonomic Experimental Research (AFTER) [LE150100015], an Australian Government Research Training Program (RTP) grant to TJS, and a Centre for Archaeological Science (CAS-Environmental Futures) Small Strategic Grant for the analytical costs of Fourier-transform infrared (FTIR) spectroscopy and scanning electron microscopy (SEM) to EMJS. The funders had no role in study design, data collection and analysis, decision to publish and preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.

Introduction Analysing human remains and their depositional context, can provide a wealth of information about the individual themselves, their lifestyle and mortuary practices of their society. In most cases, bone is the only tissue that survives time. In addition to macroscopic anatomical methods, microscopic analyses of bone histology have been used to study age-at-death, sex, behaviour, health, diet and disease both in ancient and recent human remains [1–4]. The study of diagenetic processes in archaeology, geochemistry and sedimentology gained momentum in the mid to late 20th century, but the analysis and interpretation of diagenetic alterations in bone microstructure specifically, known as ‘histotaphonomy’, is a more recent development, increasingly studied from the 2000s onwards [5]. Bone diagenesis encompasses the physical, chemical and biological processes that alter or degrade bone [6, 7]. At a microscopic level, respective examples are cracking, de- and hyper-mineralisation and microscopical foci of destruction (MFD) created by micro-organisms [6, 8, 9]. Wedl [10] was the first to describe microscopic tunnelling in fossil bones and teeth, originally thought to be caused by fungi. The tunnels became known as Wedl tunnelling and are nowadays attributed to cyanobacteria, typical for aquatic environments [8, 11, 12]. Tunnels of bacterial origin were classified by Hackett [13] in budded, lamellar or linear longitudinal according to their microarchitecture and how they invaded the bone. While several authors [e.g. 14–16] continue to use these associate descriptive morphologies, limited research exists validating their origin in bone. Indeed, multiple alternative explanations as to their origin can be given, ranging from inherent taphonomic changes to external sources, such as technical preparation of histology sections, determining their appearance seen microscopically. Recently, Turner-Walker [8] concluded there is no convincing evidence that their different morphologies link with different bacterial aetiologies. The most established method to score microscopic diagenetic alteration in bone is the Oxford Histological Index (OHI), originally proposed by Hedges et al. [17] and further developed by Millard [18], scoring the percentage of intact bone, unaffected by bioerosion. The OHI is often used in combination with bone birefringence scoring [13, 19]. More recently, a whole range of taphonomic scoring indices have been developed and re-developed [e.g. 15, 20–23] but their reproducibility has not been tested and several face challenges due to methodological subjectivity. For example, it has yet to be understood how researchers measure or judge ‘enlarged’ lacunae or canaliculi [14, 23–25] while they appear filled instead of empty, rather than enlarged. All these unvalidated methods have one common goal; the reconstruction of post-mortem biographies of human remains. At the core of controversy in this matter is the origin of osteolytic microbiota. This has led to a dichotomy of opinions between endogenous and exogenous bacterial origins. On the one hand, it is thought that osteolytic bacteria originate from the depositional environment [16, 26–30]. In contrast, other researchers claim that bacteria originate from the gut [9, 15, 23, 31] and that ribs are more susceptible for biodegradation because of their location ‘closer’ to the intestines [9, 32–34]. Some researchers are more cautious and state that differences in bone diagenesis are difficult to explain. It is difficult to discern the many unknown factors related to the funerary conditions of the remains and the local microenvironment that change over time [8, 21, 35, 36]. The origin of the gut-soil discussion goes back to observations made by Jans et al. [9] that bones from articulated animal and human remains were more affected by bacterial degradation than disarticulated ones. Several authors have argued that bacterial degradation is linked to putrefaction and the early stages of decomposition based on arguments that butchered animals do not undergo the putrefaction stage [9, 23, 31], although this hypothesis does not take into account the treatment of butchered animals such as cooking or baking kills bacteria [8]. Observations of a lack of biodegradation in stillborn pigs [23] and a limited presence of bacterial attack in archaeological human infant bones [37] have been used as additional arguments to confirm the gut bacteria hypothesis assuming that ‘bioerosion reflects the extent to which remains are exposed to bodily putrefaction’ [23, 38]. Ever since, a myriad of researchers has gone further with interpretations by trying to deduce early post-mortem practices from histological thin sections, but without experimental validation of these histology-putrefaction assumptions, thus leading to interpretative inconsistencies reported in the literature. In 2015, Booth et al. [39] claimed that a lack of microscopic biodegradation could indicate mummification practices because “ancient mummified bones are unlikely to have been affected by putrefaction” (p1156). This statement implies a lack of understanding of mummification. Mummy studies have shown that mummies pass through the stage of heavy putrefaction before they desiccate [40, 41]. Further, Trueman and Martill [38] stated that early post-mortem processes such as the mode of death influence the potential of any bone to survive into deep time. Martinon-Torres et al. [42] reported presence of MFD in bone as an argument to identify the earliest known human burial in Africa, claiming that biodegradation in bone is evidence of putrefaction to support in-situ decomposition. Similarly, other researchers directly associate putrefaction with the presence of biodegradation in bone [e.g. 15, 43]. In contrast, some studies advise caution when using the absence/presence of bioerosion in human bone to assess ‘putrefaction’. Haddow et al. [14] could not distinguish individuals with clear archaeological and osteological indications of delayed burial on a histotaphonomic level and stressed that many other contextual variables might have played. Rather than putrefaction, Trenchat et al. [44, 45] link inter-individual differences in histological preservation to different biological parameters (primary bone) and different container use (zinc/wood coffins). Hemer et al. [46] observe different histological preservation caused by different post-mortem treatment of plague victims instead of putrefaction. And Eriksen et al. [11] and Turner-Walker [47] observe different aerobic and anaerobic bone bacterial communities between terrestrial and aquatic contexts. These last six references show the interpretative potentials of histotaphonomy and stress the importance of contextual information. In forensic science, it has been claimed that bone bioerosion can be used as a method for post-mortem interval estimation, based on animal model experiments, such as those of a rat or a pig [22–24], and suggested that stillborn and neonatal death can be distinguished histotaphonomically [23]. However, these studies show low replicability and ignore biological variation in bone histology between humans and other mammals [48]. Flawed post-mortem interval (PMI) calculations and infanticide claims have serious implications for forensic investigations and thus should be conducted with methodological rigour [8, 48]. Another problem has been the single focus on analysis of histological sections with light microscopy [e.g. 15, 25, 49]. The use of only light microscopy is insufficient for the study of diagenetic changes in bone and low-resolution identifications might lead to misinterpretations of artifacts. To understand bone diagenesis holistically, microstructural and molecular changes should also be studied. Microcracking, de- and re-mineralisation, and the presence of fungi or sediment can be examined using electron microscopy [e.g. 12, 29, 50]. Microbial attack affects both the collagen and mineral phases of bone. Collagen is also lost via chemical hydrolysis [8, 29]. Additionally, bone molecular alterations have been assessed with infrared spectroscopy in archaeology [e.g. 7, 35, 51–54], forensic experiments [55, 56] and specifically for PMI estimations [56–58]. All the above-mentioned attempts to detect in-situ decomposition, ‘arrested putrefaction’, manner of death, mummification, post-mortem interval and other archaeological and forensic interpretations are underpinned by one unvalidated assumption: that putrefaction leads to bacterial degradation in bone and is caused by gut-bacteria. The latter can only be addressed by specific bone microbial studies, while the former can be answered by experimental studies with human remains. The only study that attempted to address this experimentally was undertaken by Mavroudas et al. [25] at a taphonomy facility in Texas, USA. They examined biodegradation in the rib, metatarsal and tibia from five human donors deposited in different settings (buried in soil, buried in a coffin, semi-buried in a coffin, surface deposition and exposed in a trench) for a period between 20 and 30 months. All bones showed very little bacterial histological modification. While these results are encouraging, the small sample size (only one donor per setting), methodological restriction to ground histology, and data collection completed by one observer demonstrates the need for a larger and more consistent experiment with human remains to confirm their observations and expand the study design by further methods such as scanning electron microscopy (SEM) and Fourier-transform infrared (FTIR) spectroscopy. This experimental study sets out to test the assumption that putrefaction during the early stages of decay leads to bacterial degradation at the bone histological level. The aim is to study microscopic bone diagenetic changes in human remains, alongside pigs as non-human mammal samples, that have undergone putrefaction and decomposed while deposited on the surface in an experimental, controlled setting. Two methodological approaches are applied: (1) rib and femoral bone samples are assessed using conventional histological methods, and (2) changes in micro-structural and biomolecular properties are examined using SEM and FTIR spectroscopy; testing the null hypothesis that bone bioerosion and biomolecular properties, do not correspond to putrefaction experienced during early stages of decomposition.

Discussion This study confirms the null hypothesis that histological markers of bioerosion do not correspond to putrefaction experienced during the early stages of human decomposition. All humans in this study underwent all stages of decay, including putrefaction. They displayed well-preserved bone on a microscopic histological level with unaltered molecular composition. The results have implications for investigations into early post-mortem interval in forensic anthropology and the interpretation of early post-mortem practices in archaeology. This key finding contradicts the assumption that bioerosion reflects the extent to which remains are exposed to putrefaction [9, 15, 23, 38]. Thus, using the presence of MFD in bone as an argument to prove putrefaction, in-situ decomposition, early taphonomic processes or use it as argument for primary burial [e.g. 15, 42, 43] cannot be supported based on the experimental human data presented here. These results are in agreement with experimental data obtained by Mavroudas et al. [25] in their Texas experiment on five donors placed in different burial scenarios. There was no statistically significant correlation between the OHI scores of femoral and rib samples. Visually, many samples with an OHI 5 appeared to have the onset of degradation, but not always enough to score a lower OHI. This indicates that OHI stages should be revisited and improved for assessing early degradation. The onset of degradation was more often observed in human ribs (17/17) than femurs (14/17). For the pigs, ribs scored a lower OHI than femurs, but the pig-only sample size was too small to apply statistics. In archaeology, some studies suggest that ribs are more sensitive predictors of inter-site variation in histological preservation [16, 73], while other studies show similar preservation between ribs and femora [45]. Unfortunately, many histotaphonomy studies target the femur without clear arguments of why this bone was chosen. Booth and Madgwick [32] and Booth [33] claim that ribs are more susceptible to biodegradation because of their location closer to the gut. While this claim should be studied by microbiologists, one could argue that the 3rd rib is further away or about the same distance from the gut than the proximal end of the femur which makes this claim unstable. It is likely that time and cortical thickness have an effect on biodegradation of the bone. Hence our experimental study should be repeated over longer timeframes. A noticeable difference between ribs and femurs was the presence of microcracks. Light microscopy did not reveal cracking, while SEM revealed more microfissures in rib sections than femoral sections (Figs 2, 3 and S1 Table). This observation argues against crack-index scoring with light microscopy only [e.g. 15] and stresses the importance of using complementary methods such as SEM. There are two types of microcracks that should be distinguished: cracks that are directly related to diagenesis and cracks that are the result of bone processing. According to Hollund et al. [21] larger cracks might have a non-diagenetic cause. In this study, both rib and femur samples were processed the same. It can be said that a rib is more fragile due to the lower ratio of cortical bone compared to femurs, but this does not explain if the higher number of microcracks is caused by sample processing or diagenesis in the field. However, the human rib sections with shorter time-since-deposition seem to have less microcracks (Fig 3C and S1 Table) which could be indicative of a taphonomic cause such as drying and sun exposure. Overall, the pig samples also showed less microcracks. It is assumed that human skeletal remains which are exposed for longer timeframes, will become dryer and more exposed to the elements. Pig bones naturally have a higher fat content protecting them longer against cracking. These observations indicate that microfissures, their shape, size, exact location and correlation with time-since-deposition should be studied in more detail to increase understanding of their cause. It is timely to develop a reproducible, automated scoring method that combines several techniques and resolutions. A strong statistically significant correlation was noted between the MFD scores of the pigs and the humans. In addition, the pig sections showed the onset of demineralisation with SEM, which was not observed on the human sections (Fig 9C and S1 Table). While pigs are commonly used as analogues for humans when tackling other biological questions, this result stresses that pigs (and other mammals, such as rats–see [48]) cannot be used as substitute for human remains for studies of bone biodegradation because of variation in mammalian bone histology which has consequences for how biodegradation spreads throughout the bone microstructural system [48, 74]. The lateral decomposition and condensed body structure with short legs in pigs is an incorrect representation of human decomposition. Forensic comparison studies have shown that pigs have different decomposition patterns [61, 75–79], microbiome [80] and chemical markers [81] than humans, hence producing different volatile organic compounds [75] which leads to different insect activity [61]. Most MFD in bone is caused by bacteria. This goes back to the discussion about the origin of bacteria in bone; gut versus soil. Emmons et al. [82, 83] found that microbial communities in bone from surface-deposited and shallow buried donors were more similar to those from soils, while bones recovered from donors at the saturated, deeper areas of the grave showed increased similarity with microbial communities from human gut samples with a higher representation of anaerobic taxa. If surface depositions are more likely to be colonised by soil bacteria, this could explain the different results obtained in this study between pig and human bones and the observations that MFD was present at the side of the pig rib that was touching the soil (Fig 9). The lateral decomposition of pigs provides a direct link between soil, ribs and internal organs, while humans placed in a supine position have their spine acting as barrier between the internal organs and soil, and ribs sticking up rather than touching the ground. On the other hand, the only prone donor, also displayed an immaculate microscopic bone preservation, so this should be studied further. The reason why bone of butchered animals is better preserved microscopically in the archaeology record [9, 31] might be explained by the pre-treatment (cooking, boiling, baking, roasting) rather than their contact with soil [8]. Similarly, the well-preserved pig bones from two articulated pig carcasses deposited for three years in Turner-Walker et al.’s study [30] could be explained by the application of hydrated lime to one of the pigs [84, 85]. In addition, the depositional context of the stone-built cavity might not have favoured bacterial activity either, compared to this study where the pigs decayed in full nature. Our study demonstrates statistically significant differences in bacterial degradation of human and pigs from the same depositional context. While more research is necessary into bacterial communities in bone, it is long established that every deposition is unique due to a combination of intrinsic and extrinsic factors [40] and that the amount of oxygen in the grave is a critical factor for decomposition by aerobe bacterial communities [86–89]. The local micro-environment can change between and within each burial. Findings in the present study therefore agree with Turner-Walker [8] and Booth et al. [36] that the picture is more nuanced than bacterial degradation meaning that the bones were exposed to putrefaction. Season of deposition, primary versus secondary deposition, were not statistically significantly related with bone histology markers of bioerosion. The latter confirms archaeological results obtained by Haddow et al. [14] studying histotaphonomy of 9000-year old Neolithic Near Eastern bones. It, again, stresses caution with interpretation about in-situ decomposition, and concluding primary burial based on the presence or absence of MFD in bone only such as done by Martinon-Torres [42] to prove the earliest human burial in Africa. OHI and CB were both statistically significant in fresh and frozen individuals. However, looking at the human data only, there was no variability between the fresh and frozen human data. This means that these results were directly related to the human versus pig results, rather than the frozen versus fresh. More research into the effect of freezing on bone degradation is therefore necessary. The significant correlation between OHI and CB is not surprising, nor is the link with biomolecular preservation as studied with FTIR. The intensity of birefringence is dependent on the quantity and orientation of the collagen fibres and the presence of the bone mineral [19]. Low levels of bacterial attack and no changes in organic content would not affect birefringence [19, 63, 90]. The results also indicate that there was no correlation between the post-depositional interval (days of deposition) and histological degradation, as well as between time-since-deposition and bone molecular changes. It is important to mention that the time-since-deposition ranged from 463 days (1.2 years) to 1238 days (3.3 years). From a histological point of view, Yoshino et al. [26] found that the earliest bacterial degradation in bone started between 2.5 and 5 years since death by studying non-articulated experimental bones, while a study by Bell et al. [91] noticed microstructural alterations from 3 months after death in a random forensic sample of one tibia, six ribs and three teeth ranging from 3 months to 83 years of deposition. Archaeological studies using infrared spectroscopy have shown that bone collagen decreases over time and the crystal structure of bone becomes more ordered [51, 54, 57, 63, 92–94]. In addition, the extent of chemical alteration of bone will be controlled by site specific conditions [55, 63, 94]. In this study, FTIR results showed a lack of trends at both endosteal and periosteal surfaces of pig as well as human bones (Fig 10). Looking at crystallinity index, carbonate to phosphate ratio and organic bone content, there are no correlations with time-since-deposition interval. The results are in line with the study by Wang et al. [56] who did not find a correlation between PMI and FTIR results of buried and surface skull samples deposited between 76 and 552 days (about 2.5 to 18.5 months). This is in contrast with results obtained by Howes et al [55] who found decreasing organic and carbonate contents and increasing CI of buried pig bone over a period between 3 and 23 months. However, Howes et al. [55] used disarticulated pig bones while our study examined articulated pig and human bones that underwent all stages of decomposition, which is a more realistic scenario. Trueman et al. [94] suggest that the exposure of crystal surfaces follows the loss of collagen matrix. As collagen, located in the bone matrix, degrades, the surfaces of the bone crystallites are exposed leading to an increased crystallinity. This mechanism is likely consistent with the results of the present study, and the rather short time interval of the experiment. It is assumed that the decomposition process created a protective layer of lipids, shielding collagen from degradation and bone minerals from exposure. Changes in mineral composition would only happen over a longer time interval as shown by several studies on archaeological bone samples which link reorganisation of the bone mineral’s component to time and the depositional environment [e.g. 51, 54, 92–95]. It is therefore important to conduct future studies on articulated human remains in a controlled setting over a longer skeletal time scale.

Conclusion This is the first, controlled, larger scale study with human remains to validate hypotheses in histotaphonomy. We demonstrate that bioerosion in bone microstructure and biochemical properties is not due to putrefaction. The results contrast with claims in the literature that bone bioerosion assessed from histology can be used to prove in-situ decomposition and primary burial. In this study, human donors underwent all stages of decay, including putrefaction. The bones were well-preserved on a microscopic level and did not show changes in crystallinity and organic content which makes PMI prediction unreliable over the studied period of time. These results have a significant implication for the interpretation of early post-mortem practices in archaeology and forensic science. A second important outcome is that pig remains cannot be used as proxy of human remains for the study of bone histotaphonomy. Variation was noted between human and pig histology preservation despite originating from the same experimental context. Finally, this study demonstrates that optical microscopy of thin sections remains a principal technique for the evaluation of diagenetic changes, but to understand the complete suite of biological, physical and chemical changes, electron microscopy and IR spectroscopy should also be used. The main flaws in former studies on histotaphonomy are that they used unvalidated assumptions and confused several parameters, such as in-situ decay, putrefaction and bacterial origin, in their study design and interpretations. Additional factors that complicate archaeological studies of histotaphonomy are the unknown depositional timelines and contextual uncertainties. The key message from the present study is a reminder that taphonomy is multi-factorial and approaches such as histotaphonomy require rigorous experimental validation before they can be used to make conclusions about ancient patterns in funerary treatment. It is not soil versus gut bacteria, neither is bioerosion only related to early post-mortem events. These results highlight that human remains should be analysed in their broader depositional context on a case-by-case basis, and emphasises a self-critical use of methods and interpretations. This study is currently being expanded to a similar larger scale study at different locations (Canada) and different settings (buried vs surface) over a longer period of time.

Acknowledgments We are indebted to all the donors involved in research at AFTER and to the invaluable contribution they have made to forensic and archaeological sciences. We thank all the students and staff from the University of Wollongong (UOW) and the University of Technology Sydney (UTS) and for their assistance at AFTER, in particular Jodie Ward, Maiken Ueland and Mohammed Shareef. We acknowledge and pay respects to the Darug people as the Traditional Custodians of the land on which AFTER is built. Many thanks to James Wallman and Blake Dawson for providing access and sampling permission to the pig experiments. We also thank Dave McGregor and Lolita Trenchat for their support in the histology lab at the Australian National University and University of Wollongong, respectively.

[END]
---
[1] Url: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0308440

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.