| Research background:During the forensic investigation,corpses can be found in a variety of natural or artificial aquatic environments,such as ponds,rivers,lakes,seas,and water storage containers.These cases may be suicides,homicides,or accidents,with the most common death cause being asphyxiation by drowning in the water.Due to the lower temperature and the oxygen deficit,the postmortem decomposition in the aquatic environments is usually slower than that on the land,which makes the forensic identification of corpses found in the water different from that in terrestrial setting.Postmortem submersion interval(PMSI),which is defined as the period between the entry into the water and recovery of the dead body,is approximately equal to the postmortem interval(PMI)in most of the common cases of drowning.It can be used to infer the location of body falling into the water,delimiting the scope of the search for the suspect,and providing a direction for the investigation.For corpses found on land,the PMI estimation relies on the temporal changes of the postmortem decomposition.However,determining PMSI has long been problematic because the postmortem decomposition in water is affected by a variety of biotic and abiotic factors such as microbial metabolism,algal growth,and adipocere formation.For example,corpses submerged in freshwater(rivers)may show more obvious decomposition changes than those submerged in saltwater(seas),as lower temperatures and higher salinity slow down bacterial activity.It is evident that microbial metabolism is the leading factor affecting the rate of postmortem decomposition happening in the terrestrial environment.Based on this,several studies have monitored the bacterial community successions on corpses found on land and suggested that the postmortem changes of microbial communities were dramatic,measurable,and repeatable,allowing PMI to be estimated accurately even within a long time frame.However,such estimation is more complicated for carcasses found in water,especially for the cases of drowning.When drowning occurs,alien aquatic microorganisms enter the respiratory and digestive tracts along with the liquid,changing the structure of endogenous microbial communities,consequently affecting their succession patterns.Moreover,if any rupture occurs on the decomposing carcass,fluid from the environment could also enter the body and interrupt previous succession processes.All these complex scenarios are hard to evaluate in the forensic practice.As a result,to this date,limited research has been done on the succession of the microbial communities on carcasses found in water.Some previous studies noted the presence of traceable microorganism succession patterns on the submerged cadavers at the various stages of decomposition which showed promise to be used as the biological indicators to estimate PMSI,highlighting the forensic value of biofilm investigations.Experimental Background:In a number of empirical studies,the conventional sequencing methods were used to collect useful baseline information on the possible applications of naturally occurring diverse assemblies of microbes in aquatic environments to form evidentiary linkages in forensic cases among victims,suspects,and physical evidence.These studies were carried out by a number of different researchers.However,the majority of these research groups have used traditional,culture-dependent methods in their examination of bacterial flora,which has led to an incomplete analysis of species richness.Numerous studies have been conducted in an effort to uncover the intricate microbial behavior that is associated with the decomposition process;however,the majority of these research groups have used conventional methods.In recent years,there has been a rise in the accessibility of metagenomic technologies,which has made it possible to do in-depth research on the epinecrotic communities that are present during the decomposition of vertebrae.This trend in forensic science has been accompanied by a growing realization of the need of having a strong basic science foundation.High-throughput metagenomic approaches,such as Illumina Mi Seq,are examples of well-established genomic methods that have the capacity to taxonomically identify bacterial community changes throughout the process of vertebrae decomposition,accumulate large amounts of data,and characterize the sequences of newly discovered or uncultivable microbial species.Because of these characteristics,it is possible to obtain more detailed data,which in turn enable a more in-depth characterization and understanding of the spatial and temporal biofilm composition and succession on the decaying bodies.The use of next-generation sequencing(NGS)has become increasingly prevalent in the disciplines of ecological and environmental research,particularly in the investigation of biodiversity in terrestrial and aquatic ecosystems,as well as in forensic applications.It is possible to use NGS technology to detect many species simultaneously in samples that have been taken,which is a useful capability given that samples obtained at the scene of an incident frequently exhibit mixed samples.Recent research carried out by Hyde et al.,Pechal et al.,Cobaugh et al.,Metcalf et al.,Damann et al.,Can et al.,Cheol-ho Hyun et al.,and Procopio et al.used next generation sequencing(NGS)to adopt a cultureindependent methodology.Particularly noteworthy is the fact that research organizations who based their investigations on animal models were able to develop a precursor model for PMI calculations that was more accurate.However,the vast majority of investigations that have been carried out using NGS have concentrated on terrestrial examples.Studies that investigated the variety of epinecrotic biofilms during the process of aquatic decomposition found that bacteria are the best predictors of decomposition progress in marine settings,whereas diatoms are the best forecasters of decomposition progress in freshwater systems.There have been a number of studies that have noted the presence of aquatic organisms(bacteria,fungi,algae,diatoms,and aquatic insects)on cadavers at various stages of decomposition in water.These organisms have shown promise to be used as the biological indicators to estimate PMSI,proving the forensic value of different species that are found in aquatic communities.To better understand epinecrotic biofilm succession and its potential to be used as an indicator in PMSI estimation in cases of aquatic death investigations,a high-throughput metagenomic sequencing approach was used in the current study to investigate biofilm composition under varying environmental conditions on the external surfaces of decomposing pig cadavers,which were used as proxies for human bodies.In addition,epilithic biofilm was sampled from tiles that had been placed in near and distant proximity to the rotting carcasses in order to investigate specific biofilm community behavior,which had previously been documented as exhibiting obvious succession patterns.Research purposes:1)To analyse available animal models for studying PMSI and select the best proxy for human corpses to study microorganism succession patterns2)To characterize the differences between succession patterns of the bacterial communities in the epinecrotic and epilithic biofilms3)To clarify the influence of environmental factors on the microbial biofilm succession patterns of the cadaver4)To assess the repeatability of the succession patterns in the replicable settings and establish decomposition models for the use in the non-flowing aquatic environment,with consequent development of random forest regression model based on analysed data and machine learning algorithm to use for PMSI inference for cadavers in water.Experimental Design:1)Preliminary experiments with rabbit and porcine carcasses were conducted to establish the best animal model for the experiment;2)Outdoor pig cadaver models were established to replicate human body decomposition in water.3)Three types of microbial specimens were collected: epinecrotic,epilithic,and aquatic samples.The microbial community succession and the variation of dominant species in different seasons were revealed using amplicon sequencing.4)Based on the machine learning algorithm,the mathematical model was built with the screened and established dominant species for PMSI estimation in different seasons.Experimental Process:The primary purpose of utilizing animals in experiments is to replicate the process of decomposition that occurs in human bodies.The decay cycle of experimental animals is significantly impacted by factors including body weight in crucial ways.Because smaller experimental animals degrade at a faster rate than bigger experimental animals,the decay cycle of the smaller animals cannot be equivalent to that of the larger animals or represent the process of human body decay correspondingly.When compared to the environment surrounding corpses on land,the environment surrounding corpses in water is more complicated,and the decaying process of corpses in water will be more significantly impacted by elements such as currents and aquatic species.Even when the cause of death on land prior to the event is asphyxiation,it can be more challenging to infer post-mortem interval(PMI)in circumstances where the corpse appears to have been submerged in water after death.The habitat and the kinds of microorganisms that live in the water are extremely different from those that live on land,therefore the patterns of microbial community succession on dead bodies that are found in the water will likewise be distinct from those that are found on land.Therefore,in the context of forensic microbiology,such patterns of microbial succession present a great deal of chances for investigation.In general,larger experimental animals like rabbits and pigs are chosen for the underwater corpse model experiments rather than smaller animals.Several different experimental locations,including ponds,streams,rivers,small seas,and deep seas,are selected in accordance with the requirements of the inquiry.Because the tissue will gradually detach from the corpse as it decomposes,it is important to conduct experiments in water within plastic cages that have been covered with gauze or plastic sheets.This will ensure that there is a complete sample of tissue from the corpse for each sampling,and it will also make it easier to determine the weight of the corpse after it has decomposed completely.My colleague and I conducted the preliminary experiment on rabbit cadavers between April 27 and May 10,2017,a period of thirteen days,in order to determine whether or not it was appropriate to use the model for this particular investigation.In order to determine whether or not it was appropriate to use the other model for this investigation,my colleague and I conducted a follow-up preliminary experiment on pig cadavers from the3 rd of June to the 14 th of June(11 days)in June 2017.In this particular instance,a total of three pigs weighing 30 kg were utilized in order to achieve the following objectives of the experiment(one corpse that was left on the ground to rot,exposing it to insects in its entirety,two deceased bodies were placed in the water to rot(with some exposure to insects).When deciding whether to use a rabbit or a pig corpse as an animal stand-in for the purpose of researching the decomposition process,we took into account a number of important factors.Porcine cadavers are closer in size and composition to human bodies than rabbit cadavers are,which can make them a more accurate model for the decomposition of human remains.Porcine cadavers,due to their bigger size,may be more difficult to handle or transport than those of other species.Porcine cadavers may be more expensive and less readily available than rabbit cadavers,particularly in certain geographic regions.This is notably true in terms of the availability factor.In light of this,when deciding between the two animal models,we took into account both our financial and our logistical limitations.Ethical considerations: the use of animal models gives rise to ethical concerns,and we made sure that the animal model is appropriate both morally and legally for the research by ensuring that it meets all of the requirements.The study offered particular research questions that we were attempting to answer,and determining whether a rabbit or porcine cadaver would be more suitable for our research objectives was based on whether or not those questions were answered.Insect colonization patterns,for instance,were thought to be better studied using rabbits,whereas the effects of aquatic habitats on decomposition were thought to be better studied using pig cadavers.Rabbits,on the other hand,were thought to be more ideal for researching the effects of aquatic environments on decomposition.The ultimate decision will have an impact on the experimental design of our study,which will include aspects such as the number of duplicates,sampling procedures,and data analysis strategies.The choice of animal model can have an impact on the overall design of the experiment as well as the reliability of the findings.It was determined that the nature of rabbit cadavers was too different from that required for an accurate research of microbial biofilm succession patterns,despite the fact that they offered a number of benefits,such as their ease of handling.The rate at which their bodies decomposed was also incredibly quick,which became especially clear when researchers examined the rate at which rabbit corpses decomposed when left on land at a rather high temperature and in the company of insects.Most notably,the nature of the skin is very different from that of a human in terms of its thickness,amount of subcutaneous fat,and other characteristics.As a result,a porcine cadaver was chosen to serve as the subject of the experiment.Moving hefty(about 30 kg),cages with porcine cadavers during the experiment to follow required a high degree of control to ensure that cadavers were not harmed and that sample collection was not jeopardized.This was made clear by the constraints that were posed by the nature of the experiment,which were discussed above.To tackle the issues of sampling and cadaver protection during the decomposition process,I made several device designs that would help us to proceed with the experiment in a safe and reliable manner.It took all of my interdisciplinary knowledge and group teamwork to assemble the device which could enable one person with minimal assistance to move cadavers in and out of water.Afterwards,two drowning experiments were carried out in Changsha,China,inside the Xiangjiang River watershed from the 31 st of October to the 26 th of December 2017(winter trial)and from the 21 st of July to the11 th of August 2018(summer trial),respectively.Each experiment was carried out in three freshwater ponds that were close to one another.After receiving anesthesia,the six female pigs that we used in our research were sent to a nearby farm and weighed between 7.05 and 11.50 kg apiece before being put down by drowning.Carcasses were individually placed on a fine mesh nylon pad inside the plastic cages(0.75 x 0.55 x 0.25 m3)to facilitate weighing as the carcass disarticulated and to prevent removal of the carcasses by scavengers such as fish,shrimps,and crabs.In addition,bricks were fastened to the bottom of the cages in order to prevent the carcasses from emerging from the water when they were in the swollen stage.This,in turn,prevented the carcasses from being colonized by insects that live on land.Because of these precautions,naturally occurring microbial biofilms will be able to grow on the surfaces of the carcass without being disrupted by the random activity of aquatic and terrestrial scavengers.Each cage was submerged approximately one meter below the surface of the water in its own individual pond.A simultaneous sample of two tiles that had been placed in the pond0.5 meters away from different sides of each corpse in order to investigate the influence of carcass decomposition on the production of epilithic biofilm was carried out.The tiles were placed on opposite sides of each carcass.During the summer trial,additional two sterile tiles served as a negative control by being submerged one meter below the surface of the water in the surrounding lake.This was done approximately one hundred meters away from the sunken carcasses.The postmortem alterations were photographed and recorded every day as they occurred during the course of decay.The length of time spent in each stage of decay was measured using the framework developed by Zimmerman and Wallace,and the results were used to create a visual representation of the decomposition processes at work on the corpses.The total aquatic decomposition scoring(TADS)system,as defined by Daalen et al.,was used to conduct the analysis that determined the visual body score.An aquatic decomposition scoring(ADS)system was created by Daalen et al.in order to obtain a standardized method that could quantify the degree of decomposition that had occurred in remains that had been recovered from water.The process of aquatic decomposition follows a pattern that may be described as sequential,and the scoring phases were constructed in accordance with this pattern using an ascending point based system.A data logger took readings of the temperature and relative humidity of the air once every hour.The Water Quality Checker was used to collect data on a variety of water quality parameters,including temperature,dissolved oxygen,p H,conductivity,and salinity.These readings were taken four times per day at a distance of one meter from each carcass and at the site designated as the negative control.The water temperature readings that were taken every six hours during each day of the experiment were used to determine the average daily water temperature.Accumulated degree days(ADD)were determined by adding up the average daily water temperature above the lower development threshold(LDT).The validity of the pig membrane as an in vitro model was strongly reinforced by comparison of the results with those that had previously been reported for human skin in vivo.A total of 131 samples obtained from the winter and summer trials.The cotton tip of each swab and the fragment cut from each filter membrane were put in a bead tube for genome DNA extraction.The succeeding steps were performed as dictated by the manufacturer’s specifications of Mo Bio Power Soil DNA Isolation Kit.The V4 region of the 16 S r DNA gene has been amplified using dual indexed primers(515F/806R: 5′-GTGYCAGCMGCCGCGGTAA-3′;5′-GGACTACNVGGGTWTCTAAT-3′).The 16 S protocol was designed to amplify prokaryotes(bacteria and archaea)using paired-end 16 S community sequencing on the Illumina platform.Primers 515F–806R target the V4 region of the 16 S r RNA.All amplicons in the size range of200～300 bp were purified using the Axy Prep DNA Gel Extraction Kit and pooled into equal concentrations.Sequencing libraries were generated using the Tru Seq(?) Nano DNA LT Library Prep Kit following the manufacturer’s recommendations,and index codes were added.The library quality was assessed on the Agilent Bioanalyzer 2100 system using Agilent High Sensitivity DNA Kit and the Promega.Finally,the library was sequenced on an Illumina Mi Seq platform,which generated 300 bp pairedend reads.After sequencing,the paired-end reads were assigned to the samples based on their unique barcode,truncated by cutting off the barcode and primer sequence,and merged using FLASH(version 1.2.7).The merged reads containing ambiguous bases(N)or low-quality bases were filtered out using QIIME filter(version 1.8.0),and chimeras were removed using USEARCH(version 5.2.236).After the removal of singleton sequences,operational taxonomic units(OTUs)were classified with the threshold of 97% similarity using the UCLUST in QIIME.A representative sequence was picked by selecting the longest sequence that had the largest hit number to other sequences in each OTU.Representative sequences of16 S OTUs were respectively aligned and annotated using the Greengenes database.To avoid biases of biodiversity data generated by the number of sequences,the data were rarefied to 90% of the minimum library size.Statistical analysis was conducted using a web-based tool,Microbiome Analyst.The results were visualized as a heat map to show the temporal changes in the taxonomic clusters.The alpha diversity indices of Chao1 and Shannon were calculated to evaluate the species richness and evenness of a sample.The statistical significance of differences in alpha diversity indices between experimental groups/decay stages was estimated using the analysis of variance(ANOVA).Weighted and unweighted Uni Frac distances were calculated with QIIME software.Principal coordinate analysis(PCo A)diagrams were drawn using the vegan software package of R software.The statistical significance of the clustering pattern in the ordination plot was evaluated by permutational multivariate analysis of variance(PERMANOVA).The taxonomic composition of each sample was visualized in a stacked bar plot in chronological order.Linear discriminant analysis effect size(LEf Se)was used to identify taxa with significantly differential abundance across experimental groups.Random forest(RF)analysis has been employed for the regression of the OTU relative abundances against the ADDs using the “random Forest” R package.The OTUs were ranked in the order of their feature importance and selected to generate predictive biomarker sets.The mean absolute error(MAE)and goodness-of-fit(R2)were used to evaluate the performance of the RF regression models,which established for predicting the PMSI based on generated biomarker sets.Results:1)Preliminary experiments resulted in porcine cadaver being selected as proxy for human corpse in this study.A device for weighty cadaver extraction from water was designed during the preliminary experiment with porcine cadavers that can facilitate research of PMSI estimation.2)Regression training was performed on the relative abundances and accumulated degree days(ADDs)of all operational taxonomic units(OTUs)using the random forest(Random forest)machine learning algorithm.The succession inferred PMSI model based on the experimental data from winter and summer trials explained 98.90% and 95.80% of ADD,with the mean absolute errors of 21.59±3.56 ADD and 27.21±5.75 ADD,respectively,meaning,the model inferred PMSI errors in winter and summer at 1.5 days and 0.8 days respectively.3)Similar to the findings of the previous terrestrial carcass decomposition studies,prominent bacteria successions occurred in a predictable and reproducible manner on the surfaces of the carcasses submerged in water.In addition,these successive changes were highly repetitive between carcasses within coincident experimental settings.4)During Winter,Proteobacteria and Firmicutes were the dominating communities before the body had entered the water.During the decay stages in the water,Proteobacteria remained dominant throughout,while Firmicutes had first decreased,and then increased in the floating decay stage.Bacteroidetes,representing a minor part of the antemortem communities,had increased first,followed with the short decrease,and then kept increasing up until the sunken remains stage.In addition,Cyanobacteria,Acidobacteria and Fusobacterium were scarce and occurred opportunistically throughout the decomposition.Microorganism community was more diverse in Summer but had similar dominant species with varied occurrences.5)Seasonal conditions,such as temperature,light,and water quality,were confirmed to be the critical factors to community successions and need to be studied further.The disparate community structures and succession patterns in different seasons have been observed,meaning that existing findings were not sufficient for the use in diverse environments.Conclusions:1)Successfully identified the structural changes in the surface necrotic bacterial community of bodies submerged in non flowing ponds during the entire decomposition process.This further emphasizes the suitability of pig carcasses as substitutes in human cadaver experiments,providing recommendations for future similar studies.2)We found significant differences in the community composition of the biofilms(inorganic biofilm and carrion biofilm)between these two types of corpses.Considering the influence of seasons and other environmental factors,the succession pattern of biofilms on non biological objects coexisting with necrotic biofilms can be used as one of the time reference variables for PMSI estimation in most aquatic habitats.3)This study found that temperature significantly affects the structure and succession changes of microbial membranes in aquatic corpses,which suggests that we need to consider seasonal and temperature changes when inferring the time of death using surface microorganisms of aquatic corpses.4)Our study identified prominent bacterial succession patterns in the necrotic biofilm formed on the solid liquid surface of aquatic corpses,providing new insights for accurate estimation of PMSI.28 Figures,4 Tables,138 References... |
PDF Full Text Request