Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jun 2;39(17):e10071. doi: 10.1002/rcm.10071

Advantages of Incomplete Digestion in Human Hair Shaft Proteomics, a Focus on Cuticular Keratins

Phonchawit Kiatthitinan 1,2, Darian H Yee 2, Huu M Q Tran 2, Bingyun Sun 1,✉
PMCID: PMC12130615  PMID: 40457564

ABSTRACT

Rationale

The proteome of the hair shaft has been increasingly studied by mass spectrometry for sensitive, accurate, and comprehensive characterization of major hair proteins such as cuticular keratins for biomedical and forensic applications. As an external appendage of human skin, the shaft of scalp hair is formed by dead keratinized cells that are biologically and chemically stable. The extraction and digestion of hair shaft proteins have been bottlenecks in hair proteomics.

Methods

We present a straightforward and reliable sample preparation procedure using a commercial Precellys homogenizer in mild basic conditions. We further shortened the sample preparation procedure by implementing an overnight tryptic digestion for partial proteolysis instead of a 3‐day complete digestion.

Results

Using this method, we achieved over 75% protein extraction efficiency from the shaft of human scalp hair, and the limited proteolysis improved keratin sequence coverage. The robustness of our method was confirmed by high reproducibility, with R 2 values exceeding 0.95 in pairwise quantitative comparisons via spectra counting across different operators, processes, and laboratories.

Conclusions

We developed a facile and robust sample preparation strategy for human hair shafts. The improved sequence coverage in cuticular keratins by shortened and incomplete proteolysis is critical for the identification of genetically variant peptides in keratins.

Keywords: cuticular keratin sequence coverage, genetically variant peptides, hair proteomics, human scalp hair, incomplete digestion, keratins, stoichiometry of Type I and Type II keratins

1. Introduction

Human hair proteins have been historically studied in relation to hair‐related diseases [1], and recently, increasing interest has also focused on hair proteins for forensics [2]. As an external appendage, human scalp hair is naturally shed owing to its growth cycle [1]. Therefore, it is a readily obtained sample. Traditional forensic hair examinations rely on microscopic morphology for the identification of individuals [3, 4, 5, 6]. However, microscopic morphology is not as reliable as protein composition in individual identification. For example, morphologically identical hair could have largely varied protein compositions in pedigrees [7]. The hair shaft outside of the skin is formed by dead cells with little water but rich in proteins [8]; hair proteomes from trace hair samples obtained from crime scenes and suspects can provide accurate sequence information for the identification of single‐amino acid polymorphisms (result of nonsynonymous single‐nucleotide polymorphisms [SNPs]) [9, 10, 11]. A piece of information can be used for human identification through unique genetic sequence variances (i.e., genetically variant peptides [GVPs]) [11]. These GVPs have been detected by mass spectrometry (MS)–based proteomics in the past [12], and recently, this approach was demonstrated to be an effective tool for individual identification [11].

In forensics, DNA‐ is frequently used for individual identification by unique genetic sequence variances [13]. However, reports have shown that human hair obtained from crime scenes is less likely to contain robust DNA‐ for genetic identification of variance [14]. In comparison, the rich proteins remaining in the hair, even a single 2‐cm‐long strain of hair, provide sufficient proteins for GVP identification in individuals [10].

Human hair has a filamentous structure that grows out of the epidermis with a deep root in the dermis [15]. The shaft extruding of the epidermis is formed by dead keratinized cells [2, 15]. Close to the root, the hair fiber has a medulla [16]. As the hair fiber is pushed far from the epidermis during growth, the central medulla can disappear and be replaced with peripheral cortex cells [16], as shown in Figure S1A. The outer scale‐like cuticle cells are the hardest, followed by the cuticle cells, with the medulla cells having the loosely packed softest texture [16]. All three hair layers are formed by keratinized cells. Keratins are the most abundant proteins in the hair shaft and account for more than 70%–85% of the total hair weight [8]. In cortex cells, Type I acidic keratin protein binds with Type II basic keratin to form a coiled‐coil dimer, and two such dimers coil together to form tetramers [17], as shown in Figure S1B. Keratin tetramers further aggregate to form protofilaments, and multiple protofilaments bundle together to form intermediate filaments. In cortex cells, intermediate filaments with their surrounding matrix further form macrofilaments [15]. These filamentous structures provide the tensile strength of human hair [18]. Keratin profiling of the hair shaft would be an easy and reliable target for identifying genetic variance in individual verification. For example, the 2 cm of human hair provides more than 50 μg of keratins that are suitable for such analysis [10], and ~70 proteins have been detected from as short as 1 mm of the human hair shaft [19]. Hair proteins recovered from explosive damage can still be used to identify individuals by GVPs [9].

Human hair fibers are highly resistant to biodegradation, physically robust, and chemically stable owing to dense interpeptide crosslinking, including disulfide and isopeptide bonds, which strengthen the abovementioned integumentary structures. Methods have been developed to maximize hair solubility and protein extraction efficiency. Further analysis revealed that greater hair solubility does not translate into better peptide and GVP identification, potentially because of in vitro modification of peptides introduced during prolonged sample preparation [10].

Hair keratin proteins possess many repeats, which pose a unique challenge to shotgun‐based proteomic protein identification. Here, we introduce a relatively quick method with high keratin sequencing coverage to suit the forensic analysis of GVPs and hair diseases. By comparing our results with those obtained from the gold standard, mature, and high‐protein identification protocols, we discuss a potential solution for better coverage of hair proteins with relaxed sample preparation for future rapid, reliable, hair protein and keratin profiling in applications related to criminal justice and human health.

2. Materials and Methods

2.1. Materials

Sodium dodecyl sulfate (SDS), beta‐mercaptoethanol (β‐MeSH), ethylenediaminetetraacetic acid (EDTA), N‐ethylmaleimide (NEM), tris(2‐carboxyethyl)phosphine (TCEP), and iodoacetamide (IAA) were obtained from Sigma‐Aldrich (St. Louis, MO, USA). RapiZyme trypsin (Cat# 186010106) and MCX cartridges (Cat# 186000252) were obtained from Waters (Milford, MA, USA). Bicinchoninic acid (BCA) protein assay (Cat# 23227) and all the other chemicals without specification were obtained from Pierce Thermo Fisher Scientific (Mississauga, ON, Canada).

2.2. Protein Extraction

Human scalp hair samples from two donors were collected from home hair cut with a protocol approved by the Simon Fraser University Human Research Ethics Board (application number: 30001840). Approximately 3 mg of hair was first rinsed in ethanol for 5 min to remove surface lipids and particulates. The sample was subsequently boiled with 500 μL of protein extraction solution containing 0.2 M NaOH, 1% SDS, 2% β‐MeSH, and 10 mM EDTA for 10 min and homogenized with a commercial bead beater: Precellys (Bertin Technologies, Montigny‐le‐Bretonneux, France) using 2 mL vial and ceramic beads (product name: Tissue grinding CKMix50‐R, Cat# 000922‐LYSK0‐A.0, from Bertin Technologies) operated at 6500 rpm for 2 × 20 s cycles. The boiling and Precellys homogenization steps were repeated once. The resulting suspension was centrifuged to obtain a clear solution. Acetone precipitation (1:4 sample to acetone volume and chilled at −20°C overnight) was applied to the hair protein solution, and the obtained protein pellet was resuspended in 300 μL of 0.2% SDS in 40 mM Tris–HCl buffer (pH 8). The protein concentration of each diluted sample was determined via a BCA assay.

2.3. Protein Digestion

Hair protein of 100 μg was mixed with 100 mM Tris–HCl (pH 6.8) for a final volume of 100 μL and heated at 80°C for 15 min to denature the proteins. Blocking of potentially free cysteine residues in hair proteins was performed by reacting with 20 mM NEM for 5 min [20]. The obtained protein was precipitated with acetone again following the same procedure described above to remove excess NEM. The protein precipitate was then resuspended in 50 μL of lysis buffer with 0.3% SDS, 20 mM TCEP, and 10 mM EDTA in 40 mM Tris–HCl buffer (pH 8.0) before being boiled for 10 min. After cooling to room temperature, 24 mg of ultrapure urea powder was added to the sample, and the mixture was incubated at 37°C for 30 min with continuous end‐to‐end rotation. The reduced cysteine residues were alkylated by 60 mM iodoacetamide at room temperature in the dark, and 30 mM dithiothreitol was added to quench excess iodoacetamide at room temperature for 15 min. Subsequently, 450 μL of PBS was added to the sample, and trypsin was added at 1:50 enzyme‐to‐protein ratio to digest the sample overnight at 37°C with end‐to‐end rotation (Process 1) or to digest the sample for 3 days with supplementation of equal amounts of trypsin at the end of the first and second day of incubation at 37°C with end‐to‐end rotation (Process 2) [21]. Finally, the sample was acidified to halt enzymatic activity. The digestion efficiency was confirmed by SDS‐PAGE. The resulting peptides were purified on an MCX column with the procedure recommended by the manufacture and dried in a SpeedVac (Thermo Scientific, Mississauga, ON, Canada) prior to LC‐MS/MS analysis.

2.4. LC‐MS/MS Analysis

The cleaned and dried sample was resuspended in 0.1% formic acid (FA) prior to analysis by an Easy nLC 1000 coupled to a Q‐Exactive HF Orbitrap mass spectrometer through a nano‐EASY‐Spray source (Thermo Fisher Scientific, Mississauga, ON, Canada). The separation was carried out using a 200 nL/min constant flow rate at a gradient of 60, 120, or 180 min in 2%–35% Buffer B (0.1% FA in acetonitrile) running on a 2 cm trap column and a 10 cm analytical column (both packed with Magic C18aq from Michrom Bioresources Inc.) with 75 μm ID and 5 μm resin with 100 Å pore size. After the gradient, both columns were applied a short 10 min gradient of 35%–65% Buffer B for 10 min, before rising from 65% to 85% Buffer B for 2 min and held at 85% for 20 min and subsequently reconditioned back to 0% Buffer B and 100% Buffer A (0.1% FA) for the next injection. The data‐dependent acquisition method was similar to that in previous reports [22, 23]. A voltage of 2.3 kV was used for electrospray, and the ion transfer tube that guided ions from the spray into the MS was heated at 320°C. For the Q‐Exactive HF MS/MS analysis, the data‐dependent acquisition (DDA) method was applied. The top 10 most abundant precursor ions were selected from the MS1 scan and subjected to higher energy collisional dissociation (HCD) fragmentation with a dynamic exclusion of 10 s. The MS1 scans were acquired in the m/z range of 400–2000 at 120 000 mass resolution, and the MS2 resolution was 30 000. The automatic gain control (AGC) values for the MS1 and MS2 scans were 1 × 106 and 2 × 105, respectively. Precursors with charges of 1+ or > 5+ were excluded from the MS2 scans, and the default charge was 2+. The ion injection (IT) times for MS1 and MS2 were 200 and 100 ms, respectively. The mass isolation window for MS2 was 2.0 Th offset by 0.5 Th to include more isotopic peaks of the target peptides.

2.5. Data Analysis

Xcalibur (Thermo Fisher Scientific, Mississauga, ON, Canada) was used to acquire the mass spectra. The obtained raw spectra files were further searched by Proteome Discoverer 2.1 (Thermo Fisher Scientific, Mississauga, ON, Canada) using the Sequest HT algorithm against databases of the human UniProt proteome and common contaminants. The monoisotopic precursor mass tolerance and the monoisotopic fragment mass tolerance were set at ±10 ppm and ±0.02 Da, respectively. Other search criteria included full tryptic ends, a minimum peptide length of six amino acids, and two miscleavages. Cysteine carbamidomethylation (+57.051 Da) was selected as a static modification; NEM modification, methionine oxidation (+15.995 Da), glutamine and asparagine deamidation (+0.984 Da), and acetylation at the protein N‐terminus (+42.011 Da) were selected as dynamic modifications in the search parameters. The reversed sequences of the human UniProt database were used to assess the false discovery rate (FDR). A strict FDR cutoff of 0.01 and Percolator were used to filter the confident identifications.

3. Results and Discussion

3.1. Protein Extraction

A rapid and mild alkaline extraction approach was reported by Wong et al. [24]. We tested the integration of the published alkaline protein extraction solution with our Precellys homogenization step. After two cycles of heating and homogenization (~30 min), we achieved an average protein extraction efficiency exceeding 75%, which nearly doubled the yield reported for heating and mechanical stirring [24]. Precellys can effectively break hair fibers into small particles as shown in Figure S2A in the smooth pellet formed after centrifugation, which is crucial for high protein extraction.

3.2. Alkylation of Free Thiols by NEM

We examined a NEM alkylation step [20] to label the free thiols formed during protein extraction. The obtained MS results suggested that few free thiols from cysteine residues were labeled. In our best results, for more than 700 detected peptides, only 28 NEM modifications were observed, accounting for less than 4% of the total identification. In contrast, 182 IAA alkylations, which are 6.5‐fold greater than the number of NEM modifications, were observed in the same analysis. The results suggested that the reduced disulfide bonds during protein extraction might have been reformed after acetone precipitation and resuspension in Tris–HCl buffer at pH 8 [25].

3.3. Protein Digestion and LC‐MS/MS Identification

The efficiency of the enzymatic digestion of hair protein was examined by SDS‐PAGE (Figure S2B). The complete removal of higher molecular weight proteins ‐ after digestion is shown in Figure S2B. Certain degradation of the sample occurred during the preparation step, reflected by the less effective separation of proteins in the gel, and was confirmed by the semi‐tryptic search in the obtained MS results. More than 150 proteins can be robustly identified in single LC‐MS runs (Table S1). We also varied the LC elution time from 60 to 180 min to determine the optimal separation conditions. Interestingly, the number of proteins identified did not significantly increase when we extended the LC gradient (Figure 1A), likely because of the large dynamic range in the sample owing to the dominant keratin proteins. Therefore, we maintained a 60‐min gradient for higher analysis efficiency.

FIGURE 1.

FIGURE 1

Open in a new tab

Comparison of the identified hair proteomes from different gradients. (A) Comparison of the number of detected proteins from chromatography gradients of 1, 2, and 3 h of the same hair sample. (B) Comparison of PSM% in the 1 and 2 h chromatography results. (C) Comparison of PSM% in the 1 and 3 h chromatography results. (D) Comparison of PSM% in 2 and 3 h chromatography results. PSM, peptide spectrum matching.

To evaluate the reproducibility of the sample analysis procedure, the same human hair sample was repeatedly analyzed with technical replicates and by different operators on different dates via the same protocol (Process 1). To compare the results quantitatively, we used the spectra counting method [26, 27] with normalized peptide spectra matches (PSMs), that is, the percentage of the PSM value of a particular protein to the total PSM value of all the detected proteins (PSM%). Figure 1B–D shows pairwise comparisons of the PSM% for technical replicates of the same sample analyzed with different LC gradients from 1 to 3 h. Extremely high reproducibility was observed, with R 2 > 0.99 among all three pairwise comparisons. When the results from two different operators were compared using the same sample processing protocol (Process 1) and analyzed on different dates, Figure 2A shows a slight decrease to R 2 > 0.98 from the previous R 2 > 0.99 (Figure 1B–D) in the pairwise quantitative comparison, but high reproducibility remained. When we changed the protocols and the hair samples, regardless of the operator, the pairwise quantity comparison results, as shown in Figure 2B,C, markedly decreased to R 2 ~ 0.95.

FIGURE 2.

FIGURE 2

Open in a new tab

Quantitative pairwise comparisons of the hair proteome among different operators, different procedures, and different hair samples. (A) Comparison of PSM% in results from two different operators processing the same sample using the same protocol. (B) Comparison of PSM% in results from the same operator processing two different hair samples using two procedures. (C) Comparison of two different operators processing two samples with two different processes (Process 1: 1 day digestion and Process 2: 3 day digestion) from the same laboratory. (D–F) Comparison of different hair proteomic results from different labs using different procedures (Process 3: 3‐day digestion by Rice group [21]) on different hair samples.

We also downloaded the raw MS results from ProteomeExchange (https://www.proteomexchange.org/) obtained by the Rice group from their published studies [21] and searched them via our Discoverer software and the human protein database. Therefore, the search results of the Rice group can be compared with our results. We performed a similar pairwise comparison between our results (Data 1–3) and one Rice result (Data 4) searched by us as shown in Figure 2D–F. The R 2 value was maintained in the range of 0.95–0.96. These results suggested the robustness of the hair proteomics results.

Although the quantitative comparison results revealed high similarity, the number of identified proteins differed from our results and those from the Rice group. In a typical result from the Rice group, over 600 human proteins were detected, whereas we detected only 150–200 proteins. When we combined three of our results (Data 1–3), the total number of identified human hair proteins was greater than 300. This number is equivalent to that of published proteins [28] but significantly lower than that of proteins reported by Rice et al. [21].

To identify the differentially detected proteins, we compared the protein IDs among the datasets analyzed in Figure 2, as shown in the Venn diagram in Figure 3A. For the 514 proteins that were uniquely detected in the Rice results, we analyzed their enriched Gene Ontologies (Figure 3B–D). The major enriched GO terms focused on cytosolic proteins in “translation,” “cytosolic ribosome,” and “RNA binding.” The potential causes can be several, such as a higher amount of the sample loading to boost the detection of low abundance proteins, or the inclusion of the part of the hair roots in the sample, which contain living cells [29] and have full protein complements instead of the dead and fully differentiated keratinocytes in the hair shaft. We sampled the hair shaft far from the root to minimize the inclusion of roots, and we loaded only 1–2 μg of sample for each LC‐MS/MS analysis.

FIGURE 3.

FIGURE 3

Open in a new tab

Comparison of detected hair proteomes from our study as well as published results. (A) Venn diagram comparison of hair proteins detected by us (Data 1–3) to published results from the Rice group (Data 4). (B–D) Gene Ontology enrichment of the biological process (B), cellular component (C), and molecular process (D) terms of 514 proteins uniquely identified in results from the Rice group (Data 4).

To further investigate the data, we isolated all keratins from our results. In total, we identified all reported cuticular and cytoskeletal keratins [30]. Using the detected spectra to quantify the relative quantity of the detected keratins, we discovered that cuticular keratins were the dominant proteins and were 21‐fold more abundant than cytoskeletal keratins (Figure 4). Cytoskeletal keratins were present in both our study and the Rice study and were also reported in recent hair proteome publications [28]. Historically, cytoskeletal keratins were considered to be expressed in epithelial cells [31]. Using antibody‐based staining, selected cytoskeletal keratins were also detected in the hair medulla [32]. We therefore focused our study on the robustly detected cuticular keratins for analysis.

FIGURE 4.

FIGURE 4

Open in a new tab

Quantitative comparison of Type I and Type II keratins in cuticular and cytoskeletal keratins detected in our study (Data 1–3) and results from the Rice group (Data 4).

First, we compared the quantity of two types of cuticular keratins in human hair. It is known that acidic Type I keratin and basic Type II keratin form a coiled‐coil structure at a 1:1 ratio [15]. Our MS results revealed an approximately one‐fold greater quantity of basic type II cuticular keratins than acidic Type I cuticular keratins (Figure 4, Data 1–3), which is also consistent with the results of Rice (Figure 4, Data 4). Further analysis of the amino acid composition of two representative cuticular keratins: O76009, Type I, and P78386, Type II, we identified the cause of such detection bias. The basic keratin P78386 carries a total of 66 basic residues of K and R, which trypsin recognizes for proteolytic digestion in our shotgun proteomics. In contrast, acidic keratin O76009 has only 42 such basic residues. Higher number of enzymatic cleavage sites and better ionization efficiency for a greater percentage of basic residues [33] in Type II keratins render these proteins more likely to be identified in shotgun proteomics with higher PSM% values.

Because in forensic analysis, the use of GVPs to identify individuals requires higher protein sequence coverage, we further analyzed the sequence coverage of cuticular keratins. Cuticular keratins are the most abundant proteins in the hair proteome. A comparison of the detected peptide groups from O76009 (Figure 5A) and P78386 (Figure 5C) between our results (Data 1–3) and the Rice results (Data 4) (Table S2) showed that we detected more peptides with greater protein coverage from both proteins using Process 1 (Data 2 and 3). We were intrigued by this result, because we identified quite fewer proteins. An in‐depth comparison of each identified peptide (Table S2) suggested that we had more miscleavages in the detected peptides (Figure 5B,D). These miscleavages are not ideal in typical shotgun proteomics, but they can increase protein coverage, when complete cleavages resolve smaller peptides and are ignored by the search algorithm (length < 6 amino acids). In the case of cuticular keratins, such as P78386, the average length of tryptic peptides is 7.7 amino acids, and more than 1/3 of tryptic peptides have a length of five amino acids or fewer that will not be detected owing to their short length and unreliable sequence match to the existing proteins. Miscleavages that occurred in our limited proteolysis extended the length of these short peptides and allowed them to meet the research criteria, which contributed to an increase in protein sequence coverage.

FIGURE 5.

FIGURE 5

Open in a new tab

Comparison of peptide groups detected from Type I acidic cuticular keratin, O76009, and Type II basic cuticular keratin, P78386 among our Data 1–3 and Rice Data 4. Data 1 and 4 were from 3‐day digestion, and Data 2 and 3 were from overnight digestion. (A) Venn diagram of the peptide groups detected from O76009. Three sections are boxed by dotted lines, that is, unique to Data 2–4, common to all, and unique to Data 1 and 4. (B) The average miscleavage sites in a peptide detected in the sections highlighted in (A). (C) Venn diagram of the peptide groups detected from P78386. Two sections are boxed by dotted lines, that is, the top and the bottom sections. (D) The average miscleavage sites in a peptide detected in the sections highlighted in (C).

Complete trypsin digestion as Rice group demonstrated can be achieved through extensive digestion for 3 days. In our studies, Process 1 (Data 2 and 3) included only an overnight incubation. When we extended our digestion time to three nights in Process 2 (Data 1), similar to Rice protocol, we achieved comparable digestion efficiency and lower keratin coverage as well (Figure 5 and Table S2). For analyses requiring higher protein sequence coverage and faster analysis speed, our findings suggest that it is possible to use a short digestion time with incomplete digestion for better hair keratin sequence coverage. Foreseeable disadvantages of too little proteolysis include less solubility of the sample and less efficiency in C18‐based column chromatography.

4. Conclusions

In summary, we developed a facile hair proteomics procedure built on a previously developed protocol and incorporated a Precellys homogenization step to extract human hair shaft proteins for a better than 75% yield. In addition, we employed a shorter overnight tryptic digestion than the 3‐day digestion reported previously [21], which resulted in incomplete keratin digestion. However, the miscleaved tryptic peptides improved keratin sequence coverage. Our results suggest that for GVP analysis of the human hair proteome, our method can provide better efficiency with a shortened analysis time.

Author Contributions

Phonchawit Kiatthitinan: writing – original draft, writing – review and editing, investigation, visualization. Darian H. Yee: investigation, writing – original draft, writing – review and editing. Huu M. Q. Tran: investigation, writing – review and editing. Bingyun Sun: conceptualization, methodology, supervision, resources, funding acquisition, writing – original draft, writing – review and editing.

Peer Review

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer‐review/10.1002/rcm.10071.

Supporting information

Table S1 Supporting information.

RCM-39-e10071-s001.xlsx (43.2KB, xlsx)

Table S2 Supporting information.

RCM-39-e10071-s003.xlsx (12.8KB, xlsx)

Figure S1. Illustration of human scalp hair structure. (A) The three major structural components of human hair. (B) A four‐stage expansion of a single cortical cell, including a macrofilament, an intermediate filament, a protofilament, and a keratin tetramer.

Figure S2. Images of Precellys processed hair debris (A) and SDS‐PAGE result (B) of before and after tryptic digestion of the hair protein lysate.

RCM-39-e10071-s002.pdf (198.2KB, pdf)

Acknowledgments

This work is supported by the Natural Sciences and Engineering Research Council of Canada (RGPIN06073), the Canada Foundation of Innovation, and the British Columbia Knowledge Development Fund.

Funding: This work was supported by the Natural Sciences and Engineering Research Council of Canada (RGPIN06073), Canada Foundation for Innovation, and British Columbia Knowledge Development Fund.

Phonchawit Kiatthitinan, Darian H. Yee, and Huu M. Q. Tran contributed equally.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

References

  • 1. Kligman A. M., “Pathologic Dynamics of Human Hair Loss. I. Telogen Effluvium,” Archives of Dermatology 83 (1961): 175–198. [DOI] [PubMed] [Google Scholar]
  • 2. Liu Z., Simayijiang H., Wang Q., et al., “DNA and Protein Analyses of Hair in Forensic Genetics,” International Journal of Legal Medicine 137, no. 3 (2023): 613–633. [DOI] [PubMed] [Google Scholar]
  • 3. Brace C. L., “Region Does Not Mean “Race”–Reality Versus Convention in Forensic Anthropology,” Journal of Forensic Sciences 40, no. 2 (1995): 171–175. [PubMed] [Google Scholar]
  • 4. Houck M. M. and Budowle B., “Correlation of Microscopic and Mitochondrial DNA Hair Comparisons,” Journal of Forensic Sciences 47, no. 5 (2002): 964–967. [PubMed] [Google Scholar]
  • 5. Kim B. J., Na J. I., Park W. S., Eun H. C., and Kwon O. S., “Hair Cuticle Differences Between Asian and Caucasian Females,” International Journal of Dermatology 45, no. 12 (2006): 1435–1437. [DOI] [PubMed] [Google Scholar]
  • 6. Koch S. L., Shriver M. D., and Jablonski N. G., “Variation in Human Hair Ultrastructure Among Three Biogeographic Populations,” Journal of Structural Biology 205, no. 1 (2019): 60–66. [DOI] [PubMed] [Google Scholar]
  • 7. Yu J., Yu D. W., Checkla D. M., Freedberg I. M., and Bertolino A. P., “Human Hair Keratins,” Journal of Investigative Dermatology 101, no. 1 Suppl (1993): 56S–59S. [DOI] [PubMed] [Google Scholar]
  • 8. Bradbury J. H., “The Structure and Chemistry of Keratin Fibers,” Advances in Protein Chemistry 27 (1973): 111–211. [DOI] [PubMed] [Google Scholar]
  • 9. Chu F., Mason K. E., Anex D. S., Jones A. D., and Hart B. R., “Proteomic Characterization of Damaged Single Hairs Recovered After an Explosion for Protein‐Based Human Identification,” Journal of Proteome Research 19, no. 8 (2020): 3088–3099. [DOI] [PubMed] [Google Scholar]
  • 10. Goecker Z. C., Salemi M. R., Karim N., Phinney B. S., Rice R. H., and Parker G. J., “Optimal Processing for Proteomic Genotyping of Single Human Hairs,” Forensic Science International: Genetics 47 (2020): 102314. [DOI] [PubMed] [Google Scholar]
  • 11. Parker G. J., Leppert T., Anex D. S., et al., “Demonstration of Protein‐Based Human Identification Using the Hair Shaft Proteome,” PLoS ONE 11, no. 9 (2016): e0160653. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Pan S., Aebersold R., Chen R., et al., “Mass Spectrometry Based Targeted Protein Quantification: Methods and Applications,” Journal of Proteome Research 8, no. 2 (2009): 787–797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Melton T., Holland C., and Holland M., “Forensic Mitochondrial DNA Analysis: Current Practice and Future Potential,” Forensic Science Review 24, no. 2 (2012): 101–122. [PubMed] [Google Scholar]
  • 14. Linch C. A., “Degeneration of Nuclei and Mitochondria in Human Hairs,” Journal of Forensic Sciences 54, no. 2 (2009): 346–349. [DOI] [PubMed] [Google Scholar]
  • 15. Yang F. C., Zhang Y., and Rheinstadter M. C., “The Structure of People's Hair,” PeerJ 2 (2014): e619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Kunchi C., Venkateshan K. C., and Adusumalli R. B., “Nanoindentation of Hair Cortex and Medulla Regions,” Fibers and Polymers 20, no. 7 (2019): 1538–1545. [Google Scholar]
  • 17. Fraser R. D. B., MacRae T. P., Sparrow L. G., and Parry D. A. D., “Disulphide Bonding in Alpha‐Keratin,” International Journal of Biological Macromolecules 10 (1988): 106–112. [Google Scholar]
  • 18. Tohmyoh H. and Futada K., “Division of Force Among Layers Constituting Human Hair During Bending and Tension,” Journal of the Mechanical Behavior of Biomedical Materials 133 (2022): 105346. [DOI] [PubMed] [Google Scholar]
  • 19. Carlson T. L., Moini M., Eckenrode B. A., Allred B. M., and Donfack J., “Protein Extraction From Human Anagen Head Hairs 1‐Millimeter or Less in Total Length,” BioTechniques 64, no. 4 (2018): 170–176. [DOI] [PubMed] [Google Scholar]
  • 20. Paulech J., Solis N., and Cordwell S. J., “Characterization of Reaction Conditions Providing Rapid and Specific Cysteine Alkylation for Peptide‐Based Mass Spectrometry,” Biochimica et Biophysica Acta 1834, no. 1 (2013): 372–379. [DOI] [PubMed] [Google Scholar]
  • 21. Plott T. J., Karim N., Durbin‐Johnson B. P., et al., “Age‐Related Changes in Hair Shaft Protein Profiling and Genetically Variant Peptides,” Forensic Science International: Genetics 47 (2020): 102309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Adams S. L., Chang G., Fouda M. A., Kumar S., and Sun B., “Absolute Quantification of Nav1.5 Expression by Targeted Mass Spectrometry,” International Journal of Molecular Sciences 23, no. 8 (2022): 4177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Chang G., Aroge F. A., Venkateshappa R., Claydon T. W., and Sun B., “Development of an Absolute Quantification Method for hERG Using PRM With Single Isotopologue In‐Sample Calibration,” ACS Omega 9, no. 31 (2024): 33972–33982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Wong S. Y., Lee C. C., Ashrafzadeh A., Junit S. M., Abrahim N., and Hashim O. H., “A High‐Yield Two‐Hour Protocol for Extraction of Human Hair Shaft Proteins,” PLoS ONE 11, no. 10 (2016): e0164993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Jiang Y., Rex D. A. B., Schuster D., et al., “Comprehensive Overview of Bottom‐Up Proteomics Using Mass Spectrometry,” ACS Measurement Science Au 4, no. 4 (2024): 338–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Liu H., Sadygov R. G., and J. R. Yates, 3rd , “A Model for Random Sampling and Estimation of Relative Protein Abundance in Shotgun Proteomics,” Analytical Chemistry 76, no. 14 (2004): 4193–4201, 10.1021/ac0498563. [DOI] [PubMed] [Google Scholar]
  • 27. Sun B., Ma L., Yan X., et al., “N‐Glycoproteome of E14.Tg2a Mouse Embryonic Stem Cells,” PLoS ONE 8, no. 2 (2013): e55722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Wu X., Zhang T., Mao M., Zhang Y., Zhang Z., and Xu P., “A Methodological Exploration of Distinguishing Hair Quality Based on Hair Proteomics,” Proteome Science 22, no. 1 (2024): 5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Schneider M. R., Schmidt‐Ullrich R., and Paus R., “The Hair Follicle as a Dynamic Miniorgan,” Current Biology 19, no. 3 (2009): R132–R142. [DOI] [PubMed] [Google Scholar]
  • 30. Schweizer J., Bowden P. E., Coulombe P. A., et al., “New Consensus Nomenclature for Mammalian Keratins,” Journal of Cell Biology 174, no. 2 (2006): 169–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Langbein L., Rogers M. A., Winter H., et al., “The Catalog of Human Hair Keratins. I. Expression of the Nine Type I Members in the Hair Follicle,” Journal of Biological Chemistry 274, no. 28 (1999): 19874–19884. [DOI] [PubMed] [Google Scholar]
  • 32. Langbein L., Yoshida H., Praetzel‐Wunder S., Parry D. A., and Schweizer J., “The Keratins of the Human Beard Hair Medulla: The Riddle in the Middle,” Journal of Investigative Dermatology 130, no. 1 (2010): 55–73. [DOI] [PubMed] [Google Scholar]
  • 33. Schnier P. D., Gross D. S., and Williams E. R., “On the Maximum Charge State and Proton Transfer Reactivity of Peptide and Protein Ions Formed by Electrospray Ionization,” Journal of the American Society for Mass Spectrometry 6, no. 11 (1995): 1086–1097. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1 Supporting information.

RCM-39-e10071-s001.xlsx (43.2KB, xlsx)

Table S2 Supporting information.

RCM-39-e10071-s003.xlsx (12.8KB, xlsx)

Figure S1. Illustration of human scalp hair structure. (A) The three major structural components of human hair. (B) A four‐stage expansion of a single cortical cell, including a macrofilament, an intermediate filament, a protofilament, and a keratin tetramer.

Figure S2. Images of Precellys processed hair debris (A) and SDS‐PAGE result (B) of before and after tryptic digestion of the hair protein lysate.

RCM-39-e10071-s002.pdf (198.2KB, pdf)

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Articles from Rapid Communications in Mass Spectrometry are provided here courtesy of Wiley

RESOURCES