Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Mar 23;48(5):990–1005. doi: 10.1111/ics.70092

Bio‐active designed peptide for hair keratin strengthening

Chan‐Su Rha 1,✉, Byeong Gyu In 1, Kilsun Myoung 1, Chaeyeon Song 1, Eun‐Soo Lee 1, Heung Soo Baek 1, Won Seok Park 1
PMCID: PMC13636109  PMID: 41872001

Abstract

Objective

Environmental and chemical stressors can degrade the integrity and mechanical properties of hair, necessitating innovative approaches for effective hair care solutions. This study aimed to computationally design targeted tripeptides for hair protein strengthening and experimentally validate their efficacy for cosmetic applications.

Methods

We employed rational peptide design, molecular docking and comprehensive functional testing to identify optimal tripeptide sequences. Computational screening evaluated binding interactions with hair keratins, followed by experimental validation using tensile strength measurements, breakage analysis, surface morphology assessment and confocal microscopy for penetration studies.

Results

Computational screening identified KCV (Tripeptide‐132) as the most promising candidate, demonstrating notable binding energy (approximately −45 kcal/mol), whereas palmitoylated KCV (pal‐KCV) exhibited enhanced binding energy (−65 kcal/mol). Hair fibres treated with pal‐KCV showed the highest tensile strength improvement compared to untreated controls, representing a 44% increase. Additionally, KCV treatment reduced mechanical hair breakage by 48%–52% and demonstrated significant improvement in surface smoothness and cuticle integrity. Confocal microscopy revealed that KCV thoroughly penetrated hair cross‐sections, with intense fluorescence in both the cuticle and cortex regions. Molecular dynamics simulations confirmed robust KCV‐keratin interactions through 100 distinct noncovalent interactions, predominantly hydrogen bonds (75%).

Conclusion

These findings establish that KCV and its palmitoylated derivatives are highly effective agents for comprehensive hair protein strengthening applications, offering promising potential for advanced hair care formulations with superior performance characteristics.

Keywords: computer modelling, hair treatment, keratin, peptides, protein strengthening, spectroscopy, tensile strength

Bioactive tripeptides through in silico screening revealed KCV (Tripeptide‐132) as the most promising keratin‐targeting candidate. Experimental studies confirmed its superior hair‐strengthening performance with 44% tensile strength enhancement, 50% breakage reduction, improved surface integrity and fibre penetration. This computational‐experimental approach provides a robust framework for developing next‐generation hair care formulations with enhanced efficacy.

graphic file with name ICS-48-990-g003.webp

INTRODUCTION

The human body's outermost protective barrier represents a sophisticated biological defence system designed to shield internal tissues from environmental stressors and maintain physiological homeostasis [1]. This protective interface comprises the stratum corneum, hair and nail structures, each serving critical functions in environmental adaptation and survival [2]. Among these protective elements, scalp hair plays particularly diverse roles that extend beyond simple aesthetic considerations [3]. Hair contributes to thermoregulation by providing a degree of insulation [4], and offers partial indirect protection against ultraviolet radiation on the scalp [5]. In evolutionary contexts, hair serves important reproductive signalling functions, conveying information about health, genetic fitness and hormonal status [6]. Additionally, hair functions as an environmental filter, trapping airborne particles and preventing their entry into respiratory pathways [7].

At the molecular level, the structural integrity and functional performance of these protective appendages depend fundamentally on keratin proteins, which form the primary structural matrix of hair, nails and the outermost skin layers [8]. Keratin is one of the most abundant and structurally important proteins in the human body, providing the mechanical strength, flexibility and resilience necessary for effective barrier function [9]. The hierarchical organization of keratin, from individual protein molecules to complex fibrillar structures, creates a remarkably robust yet adaptable material that can withstand significant mechanical stress while maintaining its protective properties [10]. This protein matrix undergoes continuous remodelling and repair processes that are essential for maintaining barrier integrity throughout an individual's lifetime [11].

However, constant exposure to environmental stressors poses significant challenges to keratin stability and function. Ultraviolet radiation, chemical treatments, mechanical stress and oxidative damage can progressively degrade keratin structures, leading to compromised barrier function and reduced protective capacity [12]. This degradation manifests as increased hair brittleness, reduced tensile strength, loss of elasticity and diminished aesthetic appeal. The cumulative effects of such damage can significantly impact both the functional and cosmetic aspects of hair, creating a need for targeted therapeutic interventions that can restore and enhance keratin integrity [13, 14, 15].

Recent advances in peptide science have revealed the tremendous potential of small bioactive peptides for addressing keratin degradation and promoting structural repair. Tripeptides, consisting of three amino acid residues, represent particularly promising candidates due to their optimal balance of molecular size, stability and biological activity [16]. Their relatively small molecular weight enables effective penetration through hair cuticles and into the cortical regions where keratin repair is most needed. Simultaneously, their specific amino acid sequences can be designed to interact selectively with keratin structures, promoting targeted binding and functional restoration [16].

Molecular docking simulations have emerged as powerful computational tools for predicting and optimizing these interactions, enabling researchers to screen large numbers of potential peptide candidates and identify those with the highest binding affinity and therapeutic potential. These computational approaches can significantly accelerate the discovery process while reducing the costs and time associated with purely experimental screening methods [17].

Despite significant progress in peptide‐based hair care research, substantial gaps remain in our understanding of the precise molecular mechanisms governing tripeptide‐keratin interactions. Previous studies have often focused on empirical testing of limited peptide libraries without systematic analysis of structure‐activity relationships [18]. The application of advanced computational modelling techniques, combined with comprehensive experimental validation, offers unprecedented opportunities to develop highly targeted and effective therapeutic interventions [19].

This study addresses these knowledge gaps by employing a systematic computational‐experimental approach to identify and validate optimal tripeptide sequences for keratin strengthening. Through rational peptide design, molecular docking simulations and comprehensive functional testing, we sought to develop scientifically validated solutions for hair protein repair and strengthening [20, 21, 22]. Our integrated methodology combines predictive modelling with rigorous experimental validation to establish clear structure‐activity relationships and demonstrate therapeutic efficacy [23]. The research presented here represents an advancement in the field of peptide‐based hair care, providing both fundamental insights into peptide‐keratin interactions and practical solutions for addressing hair damage and degradation. By establishing clear connections between molecular structure, binding affinity and functional outcomes, this work lays the foundation for the development of next‐generation hair care products with enhanced efficacy and targeted therapeutic action [24, 25, 26].

EXPERIMENTAL SECTION

Chemicals

Tripeptides and palmitoyl tripeptides were purchased from Caregen Ltd. (Anyang, Republic of Korea) and Wellpep Co. Ltd. (Incheon, Republic of Korea). Free amino acids were purchased from Sigma‐Aldrich (St. Louis, MO, USA). All other chemicals used in this study were of American Chemical Society grade or higher.

Tripeptide candidate sequence development and binding affinity assessment

Potent amino acids were identified based on principal component analysis (PCA) and a neural machine learning model following docking score evaluation of the entire 8000‐compound library (data not shown). A library of candidate tripeptide sequences was constructed using the following criteria. Three amino acid groups were defined: group 1 (C), consisting of amino acids effective in disulfide bonding, such as cysteine (Cys, C); [16] group 2 (HPO), comprising hydrophobic amino acids with enhanced penetration properties, such as alanine (Ala, A), leucine (Leu, L), phenylalanine (Phe, F), proline (Pro, P), valine (Val, V) and tryptophan (Trp, W); [27] and group 3 (POS), which included positively charged amino acids that promote tensile strength, such as arginine (Arg, R), lysine (Lys, K) and histidine (His, H) [28, 29, 30]. Combinations of these groups were arranged into nine cases based on the importance of groups 1 and 2, resulting in 216 tripeptide candidates (Tables 1 and S1). Sequences were translated into three‐dimensional (3D) peptide molecular files (.sd) using Discovery Studio (DS) software 2024 (Dassault Systèmes, Paris, France). To determine binding affinity, three keratin proteins (Figure S1) from Homo sapiens, including keratin 85 (K85; UniProt P78386, AF‐P78386‐F1‐v4), keratin 33b (K33b; UniProt Q14525, AF‐Q14525‐F1‐v4) and keratin‐associated protein 3–1 (KAP3‐1; UniProt Q9BYR8, AF‐Q9BYR8‐F1‐v4), were obtained from the AlphaFold Protein Structure Database (TEMBL‐EBI; Hinxton, Cambridgeshire, UK) [31]. The binding energies between the tripeptides and keratin proteins were calculated using the CDocker module in DS software (Table S2). The docking sphere in each keratin was automatically selected using DS software. Tripeptides with binding energies of <0 kcal/mol were selected for subsequent analyses.

TABLE 1.

Sequence combinations of tripeptide candidates based on amino acid groupings.

Case Sequence composition1 1st AA 2nd AA 3rd AA No. Tripep
1 Group 1 (C2) → Group 2 (HPO3) → Group 2 (HPO) 1 6 6 36
2 Group 2 (HPO) → Group 1 (C) → Group 2 (HPO) 6 1 6 36
3 Group 2 (HPO) → Group 2 (HPO) → Group 1 (C) 6 6 1 36
4 Group 1 (C) → Group 2 (HPO) → Group 3 (POS4) 1 6 3 18
5 Group 2 (HPO) → Group 1 (C) → Group 3 (POS) 6 1 3 18
6 Group 2 (HPO) → Group 3 (POS) → Group 1 (C) 6 3 1 18
7 Group 1 (C) → Group 3 (POS) → Group 2 (HPO) 1 3 6 18
8 Group 3 (POS) → Group 1 (C) → Group 2 (HPO) 3 1 6 18
9 Group 3 (POS) → Group 2 (HPO) → Group 1 (C) 3 6 1 18
Open in a new tab

Note: 1Classification of tripeptide candidates based on the sequential arrangement of amino acid groups. Each case represents a unique combination of 2cysteine (C), 3hydrophobic (HPO) and 4positively charged (POS) amino acids along with the respective number of tripeptide sequences generated. The sequences generated are listed in Table S1.

Amino acid combinations, tripeptides and fatty acid‐conjugated tripeptides preparation

Mixtures were prepared to assess the functional impact of the tripeptide‐derived amino acids. The two mixtures were based on CKW and KCV (INCI Monograph ID: 39901, Tripeptide‐132 HCl) (high docking score), and the other was based on PCP (low docking score), each with specified molar ratios. The synthesis of five key tripeptides (CKF, CWK, KCL, KCV and PCP) aligns with established protocols, beginning with the activation of a resin such as 2‐Cl‐(Trt)‐Cl resin. Fmoc‐protected amino acids were introduced sequentially using standardized methodologies to ensure that each residue was added in a controlled manner [32]. After the desired sequence was constructed, the tripeptides were cleaved from the resin. This process often involves acidic treatments that release peptides while leaving the ancillary groups intact [33]. Tripeptides can be modified to enhance their functional properties by conjugation with fatty acids. For instance, palmitic acid can be conjugated to the N‐terminus using reagents such as hexafluorophosphate‐benzotriazole‐tetramethyluronium (HBTU) and N‐methylmorpholine (NMM) in dimethylformamide (DMF), as previously reported [34].

Hair tress preparation and determination of denaturation enthalpy

Detailed methods can be found in Supporting Information S1.1 and S1.2.

Tensile strength of hair

Bleached hair tresses (1 g) were immersed for 30 min in 100 mL of a 0.1% (w/w) solution of amino acids or synthesized tripeptides. The treated tresses were rinsed under running water for 1 min and then air‐dried for 12 h in a controlled environment at 25 ± 1°C and 55 ± 5% relative humidity. For the untreated control samples, bleached tresses were washed with a basic shampoo free of conditioning agents, rinsed for 1 min and dried for 2 min. Fifty individual hair fibres from the middle section of each treated tress were randomly selected, ensuring a consistent mean diameter of approximately 70–90 μm across all groups. The tensile strength (break load) of each fibre was measured under the same controlled conditions using an MTT175 device (Dia‐Stron; Andover, UK). As preliminary tests confirmed that the peptide treatments did not induce significant changes in fibre diameter, break load was used as a direct and valid indicator of changes in mechanical strength attributable to the treatments.

Combing‐induced hair breakage

Four replicates (n = 4) of Indian hair samples (30 cm long, 6 g) from Beaulax were secured using Sirtaki hair clips to assess hair breakage. The hair was combed continuously for 8 and 16 h using a Sirtaki comb (Bossanova Vision, Los Angeles, CA, USA) at a speed of 25 rpm (Figure S2). The area of the detached hair fragments was measured using image analysis, and the data were quantitatively analysed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). This area‐based measurement (Sirtaki method) was chosen as it correlates with the total mass of broken fibres, providing a more accurate reflection of the true magnitude of hair damage than counting individual fragments. This method appropriately weights longer, more structurally significant fractures over minor tip breakages. The results are presented as the mean ± standard deviation, and statistical significance was determined using a one‐way analysis of variance (ANOVA) followed by Tukey‐Kramer Honestly Significant Difference (HSD) test.

Hair surface imaging and fluorescent binding assay

Detailed methods can be found in Supporting Information S1.3 and S1.4.

Molecular dynamics simulations of K85 and KCV

Ten KCV peptides were randomly allocated alongside a fragmented K85 molecule (Cys185‐Asp273, 89 amino acids in total) using DS software. The Nanoscale Molecular Dynamics (NAMD) conditions were as follows: simulation time, 300 ps; production, canonical ensemble (NVT); time step, 2 fs; target temperature, 300 K; and nonbonded distance, 14 Å. All other boundary conditions were the same as the default values used for the NAMD protocol. The end pose of the simulation was captured, and the ligand‐residue interactions were analysed.

Molecular docking simulation for multi‐tripeptides library

To compare the performances of the selected tripeptides with those of other peptides known for their hair repair or related effects, commercial tetrapeptides (obtained from personalcarecouncil.org) and decapeptides (University of Minho, Braga, Portugal) were used [15]. The peptides were normalized to similar amino acid numbers as follows: our tripeptides were elongated to 12 amino acids by adding the same or different corresponding tripeptides (CKF, CWK, KCL and KCV) for a total of 1296 dodecapeptides. Similarly, commercial tetrapeptides (Tetrapeptide‐97; INCI Monograph ID 37801; synthetic peptide consisting of Arg, Cys and Gly) were elongated to 12 amino acids by adding the same tetrapeptides (36 tetrapeptides and 36 dodecapeptides) and nine decapeptides comprising 10–13 amino acids (PepA to PepG and two KPs) (Supplementary data file: Multi_peptides.csv). The unknown amino acid composition of tetrapeptides was randomly mixed on the basis of possible amino acids given by the Personal Care Products Council (https://incipedia.personalcarecouncil.org/infobase/ingredient‐custom‐search/, Washington, DC, USA). All the additional peptides described above were coded using the R script and exported to a csv file. The 3D coordinate molecule files were generated using DS software (Dassault Systèmes).

Statistical analysis

Data are expressed as mean ± standard error of the mean for a set of samples. One‐way analysis of variance was performed using the Tukey‐Kramer HSD test for comparisons for all pairs at the significant level of α = 0.05 using JMP 18 (SAS Institute Inc., Cary, NC, USA). The –CDocker energy of docking simulation and amino acid datasets (amino acid type and position, that is 1st, 2nd and 3rd in tripeptides) were implemented using SIMCA 18 (Umetrics AB, Umeå, Sweden) for PCA and partial least squares (PLS) analysis. The models were fitted using the autofit option and optimized by eliminating outliers in the score and normal probability plots until the highest R2 and Q2 values of the fitted models were obtained. Amino acids were identified using variable importance in the projection (VIP) and loading plots, where the cut‐off was VIP>0.8. Predictive statistical modelling was performed using the neural model of KFold (5) and 0.3 portion of the validation method from the dataset. The number of hidden nodes of the neural model was set to three.

RESULTS

Computational screening and optimization of hair‐targeting tripeptides

Optimal keratin‐binding candidate among tripeptides

The binding affinities of the 216 tripeptide candidates for three keratin proteins (K85, K33b and KAP3‐1) were assessed using molecular docking simulations. Table S2 lists the docking conditions and the representative binding modes. Using the 50 highest –CDocker energies for each peptide candidate, a PLS analysis was conducted for K85. Based on the tripeptide composition, the input variables were as follows: first, second, third, first two amino acids, and last two amino acids. Distinctive clusters are shown in the score plot (Figure 1a), resulting in proline (P)‐rich peptides positioned on the left side of panel A, and cysteine (C)‐ and lysine (K)‐rich peptides in the upper right corner of panel A. The PLS model was validated using a residual normal probability plot (Figure 1b). The most influential amino acid and the sequential position of amino acids were identified as P and K at every position, negatively and positively, respectively (Figure 1c,d). Finally, predictive neural modelling was performed using a combined dataset of the three keratin docking results to select several top‐ranked peptides (Figure 1e and Figure S3). Tripeptides, such as CKF (or KCF), CWK (or CKW), KCL and KCV, demonstrated binding energies in the top 3% of 216 (approximately −50 kcal/mol), indicating strong interactions with the binding site of keratin proteins (Table S3). Notably, the KCV (Figure S4) exhibited the most favourable docking results, not only in the neural model but also in the bootstrap forest model (data not shown). Conversely, PCP showed the weakest interaction with a binding energy above 0 kcal/mol. Interestingly, all peptides containing P showed fewer interactions within similar groups. Therefore, CKF, CWK, KCL, KCV and PCP were selected and synthesized for comparison, as described below.

FIGURE 1.

FIGURE 1

Open in a new tab

Multivariate analysis of tripeptide‐keratin binding interaction and molecular docking analysis of peptide classes and modifications. Partial least squares (PLS) and neural network analyses of 216 tripeptide candidates bound to keratin proteins K85, K33b and KAP3‐1. (a) Principal component analysis (PCA) score plot showing distinct clustering patterns based on amino acid composition. Each point represents one tripeptide candidate, coloured according to the amino acid composition: A (alanine), C (cysteine), F (phenylalanine), H (histidine), K (lysine), L (leucine), P (proline), R (arginine), V (valine) and W (tryptophan). (b) Normal probability plot of residuals validating PLS model quality (R2Y = 0.709, Q2 = 0.707). The linear distribution of the residuals confirms the validity of the model and the absence of systematic bias. (c) Variable importance in projection (VIP) scores identifying the most influential amino acid positions. AA1, AA2 and AA3 represent the first, second and third amino acid positions in the tripeptide sequence, respectively. VIP scores of >0.8 indicate significant variables. (d) PLS loading plot showing the relationship between amino acid identity and binding energy. P showed a negative correlation (lower left quadrant), whereas K showed a positive correlation (upper right quadrant) with binding affinity. (e) Predictive neural network model performance showing the relationship between predicted and experimental –CDocker energies. The model demonstrated a strong predictive capability with a high correlation coefficient. Higher negative values indicate stronger binding affinity. Data scaling: UV scaling was applied to all variables. (f) Class A‐s: Original single tripeptides (CKF, CWK, KCL, KCV, PCP); Class A‐c: Palmitoyl‐conjugated tripeptides (pal‐CKF, pal‐CWK, pal‐KCL, pal‐KCV, pal‐PCP); Class A‐m: Identically extended 12‐mers from tripeptides (e.g. KCVKCVKCVKCV); Class A‐r: Randomly extended 12‐mers from tripeptides; Class C: Commercial tetramers from Caregen and their extended 12‐mers (RGCR, GGRC and CGRR series; the selected tetramers and 12‐mers were chosen among 1440 in total); Class K: Decapeptide 13‐mers including KP variants (GGVCGPSPPCITT and GGVCGFSFFCITT). Each box represents the interquartile range (25th–75th percentile), with the median line and whiskers extending to 1.5× IQR. Outliers are excluded. Lowercase letters in the frame indicate statistically significant differences between groups, as determined by the HSD test (p < 0.05). Different letters indicate statistically significant differences between the groups; n = 10 top scores per group, depending on peptide class.

Docking results of conjugated and elongated peptides

According to our docking results, the corresponding tripeptides were conjugated by palmitoyl moieties, such as pal‐CKF, pal‐CWK, pal‐KCL, pal‐KCV and pal‐PCP. Furthermore, we pooled two other types of oligopeptides, short oligomers (tetramers) and long oligomers (decamers) and subjected them to an additional docking test as described above. As shown in Figure 1f, palmitoylation further reduced the binding energies, with pal‐KCV achieving the highest –CDocker energy (approximately 65 kcal/mol) among the five palmitoylated tripeptides, indicating the strongest binding affinity. Further analysis of the docking results (Figure 1f) revealed several key insights into the binding affinities of various peptide classes. Among the non‐palmitoylated tripeptides (Class A‐s), PCP exhibited the lowest –CDocker energy value (approximately 0 kcal/mol), indicating the weakest binding affinity within this group. Conversely, KCL and KCV showed the highest mean –CDocker energy (approximately 45 kcal/mol), indicating the strongest binding among the Class A‐s tripeptides. When palmitoyl moieties were introduced into these tripeptides (Class A‐c), the binding affinity generally increased compared with that of their non‐palmitoylated counterparts. pal‐KCV exhibited the highest –CDocker energy. Despite its better binding ability than its non‐palmitoylated form, pal‐PCP exhibited a relatively lower –CDocker energy (approximately 18 kcal/mol) than other palmitoylated peptides in this class. The extension of tripeptides into 12‐mers (Class A‐m), such as KCVKCVKCVKCV, significantly enhanced binding affinity, with –CDocker energies ranging from 120 to 140 kcal/mol. This represents a substantial 2.5–3‐fold increase in binding affinity compared with that of the original tripeptide forms (e.g. KCV at approximately 45 kcal/mol). Randomly extended 12‐mers (Class A‐r) exhibited binding affinities similar to or slightly higher than those of identically extended 12‐mers, reaching 140–150 kcal/mol. However, when highly effective tripeptide candidates are combined, determining the sequence specificity of the resulting 12‐mers is challenging. Peptides from other sources were also evaluated. Commercial tetramers (represented by Class C peptides, such as RGCR, GGRC and CGRR) displayed binding affinities comparable to those of Class A tripeptides (e.g. 40–60 kcal/mol). When these tetramers were extended to 12‐mers (e.g. RGCRRGCRRGCR), their binding affinities increased by approximately 100–120 kcal/mol, an improvement of approximately two‐fold. However, 12‐mers from Class C showed lower binding affinities than the 12‐mers derived from Class A‐s tripeptides (Class A‐m and A‐r). Regarding the decapeptide 13‐mers designated as KP (Class K), the sequence GGVCGPSPPCITT, as specified in the results of Cruz et al. [15], showed a relatively low binding affinity (approximately 70 kcal/mol). By contrast, the GGVCGFSFFCITT sequence disclosed in PCPC exhibited a characteristically high binding affinity (approximately 115 kcal/mol). Nevertheless, neither of these KP variants had a better binding affinity than the extended 12‐mers from Class A‐m or A‐r. Although these results compare binding at specific keratin sites, the repetitive amino acid sequence of keratin fibres indicates that these findings can be extrapolated to peptide binding throughout the keratin region.

Functional and structural effects of tripeptides on hair properties

Tensile strength enhancement by amino acid combinations

The effects of amino acid mixtures derived from CKW, KCV and PCP on hair tensile strength were evaluated. The tensile strength of the treated hair was significantly higher than that of the untreated hair (Figure 2a). The average break loads for CKW‐ and KCV‐treated hair were 127 and 126 gmf, respectively, showing a 44% improvement over untreated hair and a significant improvement over CPP‐treated hair.

FIGURE 2.

FIGURE 2

Open in a new tab

Tensile strength, thermal stability and breakage resistance analysis of hair. (a) Effects of the amino acid mixture on hair tensile strength. CKW: Mixture containing C, K and W in equimolar ratios; KCV: Mixture containing K, C and V in equimolar ratios; PCP: Mixture containing P and C in an equimolar ratio (negative control); data presented as mean ± SEM (n = 50 hair strands per group). (b) Effects of unmodified tripeptides on hair tensile strength. Con: Untreated control; CKF: Cys‐Lys‐Phe; CWK: Cys‐Trp‐Lys; KCL: Lys‐Cys‐Leu; KCV: Lys‐Cys‐Val; PCP: Pro‐Cys‐Pro. (c) Effects of palmitoylated tripeptides compared to the control and individual components. Con: Untreated control; aminomix: Amino acid mixture; palmitic acid: Palmitic acid alone; pal‐CKF: Palmitoyl‐Cys‐Lys‐Phe; pal‐CWK: Palmitoyl‐Cys‐Trp‐Lys; pal‐KCV: Palmitoyl‐Lys‐Cys‐Val. One gmf (gram‐force) is equivalent to 9.80665 × 10−3 N. Data are presented as box plots, showing the median, interquartile range and individual data points. Lowercase letters on the bar indicate statistically significant differences between groups, as determined by the HSD test (p < 0.05). (d1–d4) Grayscale images of detached hair collected after combing: (d1) Control 8 h, (d2) KCV 8 h, (d3) Control 16 h and (d4) KCV 16 h. (e1–e4) Corresponding binary images processed using ImageJ for quantitative area analysis. White areas represent detached hair fragments. Control, untreated hair; KCV, Lys‐Cys‐Val tripeptide‐treated hair. Quantitative results (% area of detached hair in mm2): Control 8 h = 20.6%, Control 16 h = 40.5%, KCV 8 h = 10.7%, KCV 16 h = 19.4%.

Tensile strength enhancement by synthesized tripeptides

The tensile strengths of the synthesized tripeptides CKF, CWK, KCL, KCV and PCP were tested. CKF‐, CWK‐, KCL‐ and KCV‐treated hairs showed significant improvements, reaching break loads of 120 gmf (Figure 2b). In contrast, PCP‐treated hair exhibited no significant changes compared with the control (Con). To assess the impact of hydrophobic modification, N‐terminal palmitoylation was applied to selected tripeptides (CKF, CWK and KCV) and their effects on hair tensile strength were compared with those of the unmodified peptides and control treatments. All palmitoylated tripeptides significantly enhanced the break load compared with Con, aminomix and palmitic acid alone (Figure 2c). pal‐KCV exhibited the highest tensile strength improvement, with a median break load of 130 gmf, followed by pal‐CWK and pal‐CKF (126 and 125 gmf, respectively). Hair treated with pal‐CWK showed a 34% increase in tensile strength compared with untreated hair, surpassing that of its non‐palmitoylated counterpart (CWK), indicating that palmitoylation improves the interaction of the peptide with hair keratin. By contrast, no significant enhancement was observed with palmitic acid treatment alone. These findings demonstrate that lipid modification via palmitoylation substantially strengthens tripeptides in hair fibres.

Thermal stability enhancement by synthesized tripeptides

Differential scanning calorimetry (DSC) was used to evaluate the effects of peptide treatment on the thermal stability of hair proteins. DSC analysis indicated that palmitoylated peptides, particularly pal‐CWK, pal‐KCV and pal‐CKF, exhibited higher denaturation enthalpies and temperatures compared to controls, suggesting improved keratin structural integrity. These preliminary observations support the hypothesis that palmitoylation may enhance the ability of tripeptides to stabilize hair proteins. Detailed DSC data are provided in Supporting Information (Figure S5).

Mechanical resistance enhancement by tripeptide KCV

The grayscale images (Figure 2d1–4) show the amount of detached hair collected after automated combing. The corresponding binary images (Figure 2e1–4) were processed using ImageJ software to highlight and quantify the area of detached hair. These binary images visually confirm the grayscale images, where the white areas represent detached hair. The white areas in the images of the KCV‐treated samples (e2 and e4) are clearly less dense than those of the control samples (e1 and e3), visually confirming the reduced hair breakage with KCV treatment. The results shown in Figure 2d,e clearly demonstrate that KCV treatment significantly reduces hair breakage caused by automated combing. Both visual inspection of the detached hair and quantitative analysis of the detached hair area consistently showed that KCV‐treated hair samples experienced considerably less breakage than untreated control samples after 8 and 16 h of combing. This treatment effectively reduced hair breakage by 48%–52%, indicating that KCV protects hair by enhancing its resistance to mechanical stress.

Localization of tripeptide KCV

The penetration and localization of the three Fluorescein 5(6)‐isothiocyanate (FITC)‐labelled tripeptides (CKF, CWK and KCV) into damaged hair fibres were evaluated using confocal microscopy (Figure 3a,b). Untreated hair (Figure 3a2) showed negligible fluorescence, indicating no significant background signal or non‐specific binding of the fluorescent label. The CKF‐treated hair samples (Figure 3b2) displayed strong green fluorescence, primarily along the periphery of the hair cross‐section, corresponding to the cuticle region. Fluorescence was also observed within the cortex and cell membrane complex (CMC), indicating that these peptides could penetrate the surface; however, their distribution remained largely concentrated in the outer layer. Both CWK‐ (Figure 3c2) and KCV‐treated hair (Figure 3d2) exhibited intense and widespread green fluorescence not only at the periphery but also throughout the cross‐section, including the cortex and CMC. This indicates that these two tripeptides have superior penetration abilities, allowing them to diffuse deeply and bind extensively within the hair fibre. The fluorescence intensity profiles across hair diameters (Figure 3e) quantitatively support the images. The control group showed only the baseline fluorescence levels. The CKF treatments resulted in sharp peaks at the edges (cuticle), with a rapid decrease toward the centre. By contrast, CWK and KCV treatments produced high and broad fluorescence intensities across both the periphery and interior, confirming their enhanced penetration and uniform distribution within the hair structure. Interestingly, many medium‐sized peaks in the ranges of 15–55 and 70–130 μm are distinguishable, demonstrating that KCV can strongly influence the cortex area, including CMC. These results indicate that the tripeptide KCV can penetrate deeply and distribute widely within damaged hair fibres, localizing both at the cuticle and cortex. The superior penetration and binding abilities of KCV validated its potential as an effective agent for hair repair and protection.

FIGURE 3.

FIGURE 3

Open in a new tab

Confocal microscopy analysis of tripeptide penetration and distribution and SEM analysis of hair surface morphology. (a1–d1) Bright field and (a2–d2) corresponding fluorescence confocal microscopy images of hair cross‐sections: (a) untreated control, (b) FITC‐CKF‐treated, (c) FITC‐CWK‐treated and (d) FITC‐KCV‐treated hair, each incubated with 0.1% FITC‐labelled tripeptides (1000 ppm). Strong green fluorescence indicates peptide localization. Scale bar: 20 μm. (e) Quantitative fluorescence intensity profile analysis across the horizontal cross‐section (a–d) from left to right of the hair image via the centre of the hair medulla. (f1) Untreated control hair, (f2) CKF‐treated hair (Cys‐Lys‐Phe), (f3) CWK‐treated hair (Cys‐Trp‐Lys), (f4) KCV‐treated hair (Lys‐Cys‐Val). Scale bar: 30 μm for all images. Hair fibre diameters: (f1) 73, (f2) 82, (f3) 85 and (f4) 87 μm.

Smoothness enhancement of hair surface by tripeptide KCV

Scanning electron microscopy (SEM) revealed the effect of different tripeptide treatments on the surface morphology of human hair fibres (Figure 3 and Figure S6). These images highlight the ability of the peptides to enhance the smoothness and integrity of hair cuticles. The untreated control hair fibres (Figure 3f1) clearly possessed the typical characteristics of damaged hair under harsh conditions. The surface appeared rough, and the cuticle scales were visibly lifted, chipped and irregular. This indicates a compromised hair surface, which can lead to increased friction, tangling and dullness. Compared with the untreated control, hair fibres treated with CKF displayed a noticeably smoother surface morphology (Figure 3f2). The cuticle scales appeared flattened and intact with less lifting and damage. This indicates that CKF treatment helps to repair and realign the cuticle layers, contributing to a smoother hair surface. Similar to CKF, hair fibres treated with CWK exhibited improved cuticle integrity and reduced surface roughness (Figure 3f3). The cuticle scales adhered more tightly to the hair shaft and the overall surface appeared smoother than that of the control. This indicates that CWK helps to restore the protective outer hair layer. Among all hair treatments, the KCV‐treated hair fibres clearly improved the most in terms of surface smoothness and cuticle integrity (Figure 3f4). The cuticle scales appear remarkably flat, smooth and tightly sealed, with virtually no lifting or damage, even under harsh conditions. The hair surface in this image appears to be exceptionally uniform and healthy. This visual evidence supports the potential of these tripeptides in improving hair health and appearance by restoring the natural protective barrier of hair.

Molecular dynamics simulation of the KCV tripeptide

An analysis of 89 amino acid residues in the K85 segment (Figure 4, positions 185–273) revealed diverse compositions with distinct distribution patterns. Glutamic acid (E) was the most abundant residue, comprising 18.0% of the total composition (16 residues), followed by leucine (L) at 12.4% (11 residues) and valine (V) at 11.2% (10 residues). This composition reflects the characteristics of intermediate filament proteins with a significant proportion of charged and hydrophobic residues [35]. Notably, the protein segment contained more polar residues (59.6%, 53 residues) than hydrophobic residues (40.4%, 36 residues). This polarity bias is consistent with the role of the protein in forming stable yet dynamic cytoskeletal structures that interact with the cellular environment [36].

FIGURE 4.

FIGURE 4

Open in a new tab

Molecular dynamics simulation of K85‐KCV interactions. (a) Ribbon representation of the K85 protein structure (grey) with bound KCV tripeptides (stick representation) at the simulation endpoint. (b) Surface‐filled representation showing the K85 protein surface (white) with KCV peptides (red), highlighting binding site accessibility and peptide positioning. (c) Detailed view of noncovalent interactions between K85 residues (thin sticks) and KCV peptides (thick sticks). (d, e) Magnified interaction schemes showing specific amino acid residues involved in binding. Atom colour coding: Grey (carbon), red (oxygen), orange (sulphur), white (hydrogen) and blue (nitrogen). All the panels are model schemes captured at the end of molecular dynamics simulation.

When the amino acids were grouped according to their chemical properties, clear patterns emerged for hydrogen bond participation. Positively charged amino acids showed the highest overall participation rate, at 78.6% (11 out of 14 residues), followed by negatively charged amino acids at 52.4% (11 out of 21 residues). Polar‐uncharged amino acids accounted for 33.3% (6 of 18 residues), whereas hydrophobic amino acids showed the lowest participation rate, at 5.6% (2 of 36 residues). A comprehensive analysis of the noncovalent interactions between K85 and the KCV ligand identified 100 distinct interactions. Hydrogen bonds dominated the interaction landscape, accounting for 75% of all interactions. Conventional hydrogen bonds were more frequent (48%) than carbon‐hydrogen bonds (27%). An analysis of amino acid participation in the noncovalent interactions revealed distinct patterns correlated with the chemical properties. Glutamic acid residues were the most frequent participants, being involved in 24 interactions (32.0% of all amino acid participation). This high participation rate reflects the abundance of glutamic acid in the protein sequence and its strong hydrogen‐bonding and electrostatic interaction capabilities. R residues participated in 15 interactions (20.0%), followed by K residues in 13 interactions (17.3%). The high participation rates of these positively charged residues highlight their importance in stable interactions with the negatively charged components of the KCV ligand.

DISCUSSION

Rationale for target keratin protein selection

The selection of K85, K33b and KAP3‐1 as target proteins for tripeptide binding studies was based on their documented susceptibility to physicochemical stress and their critical roles in hair structural integrity. This strategic selection addresses a fundamental gap in current haircare research, in which target protein selection often lacks mechanistic justification. Further comprehensive discussion is provided in Supporting Information S3 [37, 38, 39, 40, 41, 42, 43, 44].

Molecular docking results and binding affinity analysis

The molecular docking results revealed clear structure‐activity relationships that aligned with the established principles of protein‐peptide interactions. The superior performance of cysteine‐containing tripeptides (CKF, CWK, KCL and KCV) with binding energies below −50 kcal/mol reflects the importance of disulfide bonding potential in hair protein interactions. This finding is consistent with that of Cruz et al. [15], who demonstrated that cysteine‐rich peptides show enhanced affinity for keratin structures due to their ability to form both covalent and noncovalent bonds. The poor performance of proline‐containing peptides, particularly PCP, with a binding energy above 0 kcal/mol, can be attributed to the rigid five‐membered ring structure of proline, which introduces steric hindrance and reduces conformational flexibility [29]. This structural constraint limits the ability of the peptide to adopt optimal binding conformations within the keratin‐binding sites, resulting in unfavourable binding energies. The PLS analysis revealed that K residues consistently enhanced the binding affinity across all positions in the tripeptide sequence, whereas proline residues decreased the binding affinity. This pattern reflects the importance of electrostatic interactions between positively charged K residues and negatively charged regions of keratin proteins, particularly the glutamic acid‐rich domains identified in our molecular dynamics simulations.

The significant improvement in binding affinity observed upon palmitoylation highlights the importance of hydrophobic interactions in peptide‐keratin binding. This finding challenges the conventional wisdom that water‐soluble peptides are optimal for hair applications [14]. Instead, our results indicate that strategic lipophilic modification can enhance both binding affinity and penetration into the hair matrix. The mechanism of palmitoylation enhancement likely involves multiple factors: (1) increased hydrophobic interactions with lipid‐rich CMC regions of hair, (2) enhanced penetration through the cuticle barrier and (3) improved stability of the peptide‐keratin complex through additional van der Waals interactions [24]. This multimodal enhancement explains the superior performance of the palmitoylated tripeptides in both computational and experimental assays.

Functional validation and hair property enhancement

The improvement in hair tensile strength achieved with CKW and KCV treatments substantially exceeds that of many commercial hair‐strengthening products. This improvement is particularly significant when compared with the typically modest effects of single amino acid treatments or conventional conditioning agents [28]. The mechanism underlying this tensile strength enhancement likely involves multiple pathways: (1) direct reinforcement of damaged disulfide bonds through cysteine‐mediated cross‐linking, (2) electrostatic stabilization of keratin‐KAP interactions through K residues and (3) hydrophobic stabilization of protein structures through tryptophan and phenylalanine residues [30]. The synergistic effects of these interactions explain why tripeptides outperform individual amino acid mixtures. The superior performance of palmitoylated tripeptides demonstrates the importance of delivery and penetration for hair‐treatment efficacy. The lipophilic palmitic acid moiety likely facilitates deeper penetration of the hair cortex, where the majority of structural proteins are located [27].

The DSC analysis provided preliminary evidence for the structure‐stabilizing effect of tripeptide treatment. The observed trends in denaturation enthalpy and temperature suggest potential structural reinforcement, which may reflect the formation of additional cross‐links and the stabilization of existing protein structures through peptide‐protein interactions [45, 46]. While these preliminary observations are encouraging, further studies with larger sample sizes and statistical validation are warranted to confirm these findings. Detailed DSC data are provided in Supporting Information.

Penetration and distribution analysis: Confocal microscopy insights

Confocal microscopy images showing the penetration of FITC‐labelled tripeptides into both cuticle and cortex regions validated the proposed mechanism of action. Localization to the CMC region is particularly significant, as Rogers [38] noted that CMC consists of a central protein layer flanked by lipid layers, and serves as a pathway between cells in the hair structure. The preferential accumulation in the CMC regions indicates that tripeptides follow established transport pathways rather than creating new damage‐induced channels. This selective distribution pattern indicates that the peptides interacted with specific binding sites. Cruz et al. [16] demonstrated that peptide‐keratin interactions are driven by specific hydrophobic interactions and disulfide bonds, rather than by non‐specific surface coating. The depth of penetration observed for palmitoylated tripeptides supports the hypothesis that lipophilic modifications enhance transport through lipid‐rich CMC regions. This finding aligns with the established principles of transdermal drug delivery, in which lipophilic modifications are commonly used to enhance drug penetration through biological barriers [47].

Molecular dynamics simulation insights: Protein‐peptide interaction networks

The molecular dynamics simulation provided detailed insights into the specific interactions driving tripeptide‐keratin binding. The identification of 100 distinct interactions between K85 and KCV, with hydrogen bonds accounting for 75% of all interactions, demonstrated the complexity and specificity of these molecular recognition events [21]. The predominance of glutamic acid residues in peptide interactions (32% of all amino acid participation) reflects the importance of electrostatic interactions in stabilizing peptide‐protein complexes. This finding is consistent with the positive contribution of K residues observed in the PLS analysis, as lysine‐glutamate salt bridges are known to significantly stabilize protein structures [48]. The high participation rate of positively charged residues (R: 20%; K: 17.3%) in noncovalent interactions highlights the importance of electrostatic complementarity in peptide design. This information can guide future peptide optimization efforts by emphasizing the inclusion of positively charged residues at strategic positions [20].

Comparison with existing literature and commercial products

The tripeptide approach described in this study offers several advantages over existing hair‐strengthening technologies. Compared with protein hydrolysates commonly used in commercial products, the defined tripeptide sequences provide a consistent composition and predictable interactions [49]. Furthermore, enhanced binding affinity to keratin can be achieved either by simultaneously binding multiple short‐sequence peptides to keratin fibres or by utilizing oligopeptides with extended repetitive sequences. However, considering the molecular size (tripeptides: typically <500 Da) required for effective penetration into the keratin matrix, smaller low‐molecular‐weight peptides may be more advantageous than larger oligopeptides [14, 50, 51]. The computational design approach used in this study is a significant advancement over the empirical screening methods traditionally used in cosmetic development. By combining molecular docking, experimental validation and mechanistic analysis, this study provides a rational framework for peptide optimization that can be applied to other haircare applications [25].

Although the current study provides strong evidence for the efficacy of tripeptides, several limitations should be acknowledged. While providing controlled conditions for mechanistic studies, the in vitro nature of testing may not fully capture the complexity of real‐world haircare scenarios. Long‐term stability studies under various environmental conditions are needed to assess the durability of these improvements [52]. The focus on Asian and Indian hair types, although relevant to a significant global population, limits the generalizability of our findings to other hair types with different structural characteristics. Future studies should examine the efficacy of these tripeptides across diverse hair types under various damage conditions [53]. The current study examined only a limited subset of possible tripeptide sequences. Expanding the sequence space to include other amino acid combinations, particularly those incorporating non‐natural amino acids or modified residues, may yield more effective formulations [32].

Implications for haircare product development

The demonstrated efficacy of palmitoylated tripeptides, particularly pal‐KCV, has significant implications for developing targeted hair repair treatments. The ability to selectively interact with damaged keratin structures while preserving healthy protein domains represents a major advancement in precision hair care [54]. The rational design approach validated herein provides a robust framework for optimizing peptide formulations for specific hair damage conditions. The integration of computational screening with experimental validation proved highly effective, with a strong correlation between predicted binding energies and measured functional improvements. This methodology can serve as a template for accelerating hair care product development while reducing empirical testing requirements. Detailed methodological considerations, commercial implications and expanded future research directions are provided in Supporting Information S3.2.–3.3.

Future directions and broader impact

This study establishes important principles for protein‐peptide interaction design that extend beyond hair care applications. The computational framework developed herein can be adapted for targeting other structural proteins and keratin‐related disorders. Future studies should explore expanded peptide sequences, alternative lipophilic modifications and advanced delivery systems to further enhance efficacy. The mechanistic insights gained provide a foundation for personalized hair care approaches tailored to specific hair types and damage patterns. Comprehensive discussion of future research opportunities and broader scientific implications is available in Supporting Information S3.4.–3.5. While our computational‐experimental approach successfully demonstrates the potential of rationally designed tripeptides for hair keratin enhancement, we acknowledge several limitations that warrant further investigation (Supporting Information S3.6).

CONCLUSION

This comprehensive study demonstrated the successful application of computational design principles to the development of effective hair‐strengthening formulations. The rational selection of target proteins (K85, K33b and KAP3‐1) based on their documented vulnerability to physicochemical stress, combined with systematic peptide optimization and multimodal experimental validation, yielded tripeptide formulations with better efficacy than those produced using existing approaches. The 45% improvement in hair tensile strength achieved with palmitoylated tripeptides, supported by enhanced thermal stability and detailed mechanistic understanding, represents a significant advancement in haircare science. The integration of computational and experimental approaches provides a robust framework for future peptide optimization and establishes new standards for evidence‐based haircare product development. The broader implications of this study extend beyond cosmetic applications to fundamental principles of protein engineering, targeted therapeutics and precision medicine. The methodological innovations and mechanistic insights developed here provide a foundation for continued advancements in protein‐based therapeutics and personalized care approaches.

CONFLICT OF INTEREST STATEMENT

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: C.‐S. Rha and B. G. In has patent #Composition for hair protection or hair strengthening, 102 745 546, KR issued to AMOREPACIFIC Corp. C.‐S. Rha and B. G. In has patent #Composition for hair protection or hair strengthening, PCT‐KR2024‐019085, WO pending to AMOREPACIFIC Corp. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Supporting information

Data S1:

ICS-48-990-s001.pdf (2.2MB, pdf)

ACKNOWLEDGEMENTS

We are sincerely grateful to those who assisted in the docking simulation: S. J. Lim and K. H. Kim from Insilico Corp., Republic of Korea. No external funding was received for this study.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available in the Supporting Information of this article.

REFERENCES

  • 1. Proksch E, Brandner JM, Jensen JM. The skin: an indispensable barrier. Exp Dermatol. 2008;17(12):1063–1072. [DOI] [PubMed] [Google Scholar]
  • 2. Fuchs E, Horsley V. More than one way to skin. Genes Dev. 2008;22(8):976–985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Paus R, Cotsarelis G. The biology of hair follicles. N Engl J Med. 1999;341(7):491–497. [DOI] [PubMed] [Google Scholar]
  • 4. Schneider MR, Schmidt‐Ullrich R, Paus R. The hair follicle as a dynamic Miniorgan. Curr Biol. 2009;19(3):R132–R142. [DOI] [PubMed] [Google Scholar]
  • 5. Dario MF, Baby AR, Velasco MVR. Effects of solar radiation on hair and photoprotection. J Photochem Photobiol B Biol. 2015;153:240–246. [DOI] [PubMed] [Google Scholar]
  • 6. Grammer K, Fink B, MØLler AP, Thornhill R. Darwinian aesthetics: sexual selection and the biology of beauty. Biol Rev. 2003;78(3):385–407. [DOI] [PubMed] [Google Scholar]
  • 7. Aumond S, Bitton E. The eyelash follicle features and anomalies: a review. Journal of Optometry. 2018;11(4):211–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Schweizer J, Bowden PE, Coulombe PA, Langbein L, Lane EB, Magin TM, et al. New consensus nomenclature for mammalian keratins. J Cell Biol. 2006;174(2):169–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Langbein L, Schweizer J. Keratins of the human hair Follicle11This article is dedicated with gratitude to Werner W. Franke on the occasion of his 65th birthday. His pioneering work on epithelial and hair keratins has been pivotal to our own investigations in this field. Int Rev Cytol. 2005;243(J. Invest. Dermatol.1222004):1–78. [Google Scholar]
  • 10. Bragulla HH, Homberger DG. Structure and functions of keratin proteins in simple, stratified, keratinized and cornified epithelia. J Anat. 2009;214(4):516–559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Candi E, Schmidt R, Melino G. The cornified envelope: a model of cell death in the skin. Nat Rev Mol Cell Biol. 2005;6(4):328–340. [DOI] [PubMed] [Google Scholar]
  • 12. Ferrucci L, Fabbri E. Inflammageing: chronic inflammation in ageing, cardiovascular disease, and frailty. Nat Rev Cardiol. 2018;15(9):505–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Yu Y, Yang W, Wang B, Meyers MA. Structure and mechanical behavior of human hair. Mater Sci Eng C. 2017;73:152–163. [DOI] [PubMed] [Google Scholar]
  • 14. Malinauskyte E, Shrestha R, Cornwell PA, Gourion‐Arsiquaud S, Hindley M. Penetration of different molecular weight hydrolysed keratins into hair fibres and their effects on the physical properties of textured hair. Int J Cosmet Sci. 2021;43(1):26–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Cruz CF, Martins M, Egipto J, Osório H, Ribeiro A, Cavaco‐Paulo A. Changing the shape of hair with keratin peptides. RSC Adv. 2017;7(81):51581–51592. [Google Scholar]
  • 16. Cruz CF, Azoia NG, Matama T, Cavaco‐Paulo A. Peptide‐protein interactions within human hair keratins. Int J Biol Macromol. 2017;101:805–814. [DOI] [PubMed] [Google Scholar]
  • 17. Sliwoski G, Kothiwale S, Meiler J, Lowe EW, Barker EL. Computational methods in drug discovery. Pharmacol Rev. 2014;66(1):334–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Fields GB. The rebirth of matrix metalloproteinase inhibitors: moving beyond the dogma. Cells. 2019;8(9):984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Macalino SJY, Gosu V, Hong S, Choi S. Role of computer‐aided drug design in modern drug discovery. Arch Pharm Res. 2015;38(9):1686–1701. [DOI] [PubMed] [Google Scholar]
  • 20. Kurcinski M, Badaczewska‐Dawid A, Kolinski M, Kolinski A, Kmiecik S. Flexible docking of peptides to proteins using CABS‐dock. Protein Sci. 2020;29(1):211–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Zhou P, Jin B, Li H, Huang SY. HPEPDOCK: a web server for blind peptide‐protein docking based on a hierarchical algorithm. Nucleic Acids Res. 2018;46(W1):W443–W450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Hollingsworth SA, Dror RO. Molecular dynamics simulation for all. Neuron. 2018;99(6):1129–1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Lionta E, Spyrou G, Vassilatis DK, Cournia Z. Structure‐based virtual screening for drug discovery: principles, applications and recent advances. Curr Top Med Chem. 2014;14(16):1923–1938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Sanders JM, Coscia BJ, Fonari A, Misra M, Mileo PGM, Giesen DJ, et al. Exploring the effects of wetting and free fatty acid deposition on an atomistic hair fiber surface model incorporating keratin‐associated protein 5‐1. Langmuir. 2023;39(15):5263–5274. [DOI] [PubMed] [Google Scholar]
  • 25. Weiand E, Ewen JP, Koenig PH, Roiter Y, Page SH, Angioletti‐Uberti S, et al. Coarse‐grained molecular models of the surface of hair. Soft Matter. 2022;18(9):1779–1792. [DOI] [PubMed] [Google Scholar]
  • 26. Fosgerau K, Hoffmann T. Peptide therapeutics: current status and future directions. Drug Discov Today. 2015;20(1):122–128. [DOI] [PubMed] [Google Scholar]
  • 27. Kuzuhara A. Raman spectroscopic analysis of L‐phenylalanine and hydrolyzed egg white protein penetration into keratin fibers. J Appl Polym Sci. 2011;122(4):2680–2689. [Google Scholar]
  • 28. Baus RA, Leichner C, Steinbring C, Bernkop‐Schnurch A. Strategies for improved hair binding: keratin fractions and the impact of cationic substructures. Int J Biol Macromol. 2020;160:201–211. [DOI] [PubMed] [Google Scholar]
  • 29. Donald JE, Kulp DW, DeGrado WF. Salt bridges: geometrically specific, designable interactions. Proteins. 2011;79(3):898–915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Rogacheva ON, Izmailov SA, Slipchenko LV, Skrynnikov NR. A new structural arrangement in proteins involving lysine NH(3)(+) group and carbonyl. Sci Rep. 2017;7(1):16402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583–589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Mei Y, Beers KL, Byrd HCM, VanderHart DL, Washburn NR. Solid‐phase ATRP synthesis of peptide−polymer hybrids. J Am Chem Soc. 2004;126(11):3472–3476. [DOI] [PubMed] [Google Scholar]
  • 33. Gauthier MA, Klok H‐A. Peptide / protein – polymer conjugates: synthetic strategies and design concepts. Chem Commun. 2008;23:2591–2611. [DOI] [PubMed] [Google Scholar]
  • 34. Lee YS, Park JS, Jung SM, Kim SD, Kim JH, Lee JY, et al. Inhibition of lethal inflammatory responses through the targeting of membrane‐associated toll‐like receptor 4 signaling complexes with a Smad6‐derived peptide. EMBO Mol Med. 2015;7(5):577–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Scoville D, Stamm JD, Toledo‐Warshaviak D, Altenbach C, Phillips M, Shvetsov A, et al. Hydrophobic loop dynamics and actin filament stability. Biochemistry. 2006;45(45):13576–13584. [DOI] [PubMed] [Google Scholar]
  • 36. Pace CN, Scholtz JM, Grimsley GR. Forces stabilizing proteins. FEBS Lett. 2014;588(14):2177–2184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Harland DP, Popescu C, Richena M, Deb‐Choudhury S, Wichlatz C, Lee E, et al. The susceptibility of disulfide bonds to modification in keratin fibers undergoing tensile stress. Biophys J. 2022;121(11):2168–2179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Rogers GE. Known and unknown features of hair cuticle structure: a brief review. Cosmetics. 2019;6(2):32. [Google Scholar]
  • 39. Chamcheu JC, Siddiqui IA, Syed DN, Adhami VM, Liovic M, Mukhtar H. Keratin gene mutations in disorders of human skin and its appendages. Arch Biochem Biophys. 2011;508(2):123–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Breakspear S, Nöcker B, Popescu C. Chemical bonds and hair behaviour—a review. Int J Cosmet Sci. 2024;46(5):806–814. [DOI] [PubMed] [Google Scholar]
  • 41. Wu X, Zhang T, Mao M, Zhang Y, Zhang Z, Xu P. A methodological exploration of distinguishing hair quality based on hair proteomics. Proteome Sci. 2024;22(1):5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Shimomura Y, Wajid M, Petukhova L, Kurban M, Christiano AM. Autosomal‐dominant woolly hair resulting from disruption of keratin 74 (KRT74), a potential determinant of human hair texture. Am J Hum Genet. 2010;86(4):632–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Matsunaga R, Abe R, Ishii D, Watanabe S‐i, Kiyoshi M, Nöcker B, et al. Bidirectional binding property of high glycine–tyrosine keratin‐associated protein contributes to the mechanical strength and shape of hair. J Struct Biol. 2013;183(3):484–494. [DOI] [PubMed] [Google Scholar]
  • 44. Blanco A, Blanco G. Chapter 3 – Proteins. In: Blanco A, Blanco G, editors. Medical biochemistry (Second Edition). Cambridge, MA: Academic Press; 2022. p. 21–75. [Google Scholar]
  • 45. Gama RM, Balogh TS, França S, Dias TCS, Bedin V, Baby AR, et al. Thermal analysis of hair treated with oxidative hair dye under influence of conditioners agents. J Therm Anal Calorim. 2011;106(2):399–405. [Google Scholar]
  • 46. Wortmann FJ, Springob C, Sendelbach G. Investigations of cosmetically treated human hair by differential scanning calorimetry in water. J Cosmet Sci. 2002;53(4):219–228. [PubMed] [Google Scholar]
  • 47. Prausnitz MR, Langer R. Transdermal drug delivery. Nat Biotechnol. 2008;26(11):1261–1268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Kumar S, Nussinov R. Relationship between ion pair geometries and electrostatic strengths in proteins. Biophys J. 2002;83(3):1595–1612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Adav SS, Subbaiaih RS, Kerk SK, Lee AY, Lai HY, Ng KW, et al. Studies on the proteome of human hair ‐ identification of histones and deamidated keratins. Sci Rep. 2018;8(1):1599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Molisso S, Williams DR, Ces O, Rowlands LJ, Marsh JM, Law RV. Molecular interaction and partitioning in α‐keratin using 1H NMR spin‐lattice (T1) relaxation times. J R Soc Interface. 2021;18(185):20210698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Robbins CR. Chemical and physical behavior of human hair. 5th ed. Heidelberg: Springer Berlin; 2012. XXIII, 724 p. [Google Scholar]
  • 52. Nigeria ACS, Canapés C, Charnels C, Gasper A, Jokes IS. Hair color changes caused by dyeing and thermal treatments. Int J Cosmet Sci. 2005;27(1):35. [PubMed] [Google Scholar]
  • 53. Franbourg A, Hallegot P, Baltenneck F, Toutaina C, Leroy F. Current research on ethnic hair. J Am Acad Dermatol. 2003;48(6):S115–S119. [DOI] [PubMed] [Google Scholar]
  • 54. Lazarus BS, Chadha C, Velasco‐Hogan A, Barbosa JDV, Jasiuk I, Meyers MA. Engineering with keratin: a functional material and a source of bioinspiration. iScience. 2021;24(8):102798. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Data S1:

ICS-48-990-s001.pdf (2.2MB, pdf)

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.

Articles from International Journal of Cosmetic Science are provided here courtesy of Wiley

RESOURCES