Lipid distribution on ethnic hairs by Fourier transform infrared synchrotron spectroscopy

Received: 12 March 2021 Revised: 25 May 2021 Accepted: 31 July 2021 DOI: 10.1111/srt.13093 ORIGINAL ARTICLE Lipid distribution on ethnic hairs by Fourier transform infrared synchrotron spectroscopy Clara Barba1Marc Adria Oliver1Meritxell Martí1Martin Kreuzer2 Luisa Coderch1 1Department of Surfactants and Nanobiotechnology, IQAC-CSIC, Barcelona, Spain 2ALBA Synchrotron, Cerdanyola del Vallès, Barcelona, Spain Correspondence Clara Barba, Department of Surfactants and Nanobiotechnology, IQAC-CSIC, Barcelona, Spain. Email: [email protected] Abstract Background: A synchrotron-based Fourier transform infrared micro-spectrometer (µ- FTIR) allows the spatial determination of lipids across the different layers of ethnic hairs and differentiates between the lipid order arrangement and quantity. Materials and methods: The three ethnic fibers were delipidized, the lipid extracts were characterized, and the delipidized fibers were studied by dynamic vapor sorption experiments (DVS) and FTIR-synchrotron techniques. Results: The average spectra from the different hair regions exhibited the most intense CH2sym peaks on the medulla, followed by those from the cuticle and cortex for all hairs of different ethnicities. Differences in the lipid fraction of the three hair types have been observed, and they can explain some barrier properties. African virgin hair was demonstrated to have more lipids mainly in the medulla, which implies an important hydrophobicity with low hysteresis between absorption and desorption water vapor processes. In addition, these lipids are highly disordered, mainly in the cuticle, which can be related to its high water vapor diffusion. Asian and Caucasian virgin hairs presented a similar lipid order in all regions, with similar diffusion coefficients. Results indicate that the higher order of the lipid bilayer hinders water permeation kinetics in some way. Conclusion: The differences in the presence and organization of the lipids in the different regions of the African hair can account for its differentiation with regards to moisturization and swelling from the other types of fibers. KEYWORDS hair, lipids, permeability, synchrotron, µ-FTIR 1INTRODUCTION FTIR spectroscopy is based upon the absorption of IR light by the vibration of all transitions in covalent bonds. The IR spectrum of complex biological samples contains all of the contributions of different vibrating molecules including proteins, lipids, nucleic acids, and so on, depending on the natural composition of the probed samples. IR analysis is performed to extract information on specific This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2021 The Authors. Skin Research and Technology published by John Wiley & Sons Ltd. molecular components and their changes within the representative spectra.1 In IR spectra, one can detect, identify, and quantify many molecular species within a biological sample. This IR spectroscopy capability has been combined with microscopy, which allows one to conduct chemical analysis and map the distribution of chemical species with fine spatial resolution. The resultant spectromicroscopy techniques applied to biological systems have been reviewed many times.2Many common Skin Res Technol. 2022;28:75–83. wileyonlinelibrary.com/journal/srt 75 76 BARBA ET AL. biomolecules, such as nucleic acids, proteins, and lipids, among others, have characteristic and well-defined IR-active vibrational modes.3 Biomedical and pharmaceutical fields have certainly benefited from this imaging methodology.1,4 Infrared microspectrometry is an ideal tool to investigate the chemical nature of biological tissues, particularly human hair.5 A synchrotron radiation source may provide mid-IR radiation 2−3 orders of magnitude brighter than a conventional Globar through a small aperture, obtaining IR spectra with high signal-to-noise ratio values and images with a spatial resolution down to the diffraction limit in thick samples.6Fourier transform infrared (FTIR) spectromicroscopy with a synchrotron radiation-based (SR) source is a newly emerging bioanalytical and imaging tool that can monitor biochemical events within different compartments of an individual living cell without the need for fixing, staining, or labelling.7Recent uses of synchrotron infrared spectromicroscopy include the examination of biological samples such as individual living cells,8tissue samples,9 microbial–chemical interactions in environmental settings,10 protein conformations,11 and plant-soil interactions.12 A synchrotron is a highenergy electron storage ring optimized for the production and collection of intense light radiated by electrons upon acceleration. The high brightness of the synchrotron IR source plays an important role in studying localized biochemical phenomena in living single cells.2 Human hair cross-sections reveal three major identifiable regions. The medulla is the center-most portion of the hair. It is composed of loosely packed cells. The medulla can be either continuous or discontinuous along the hair length and is often completely absent. The cortex makes up the bulk of a hair and determines the strength of a hair. It is composed of long embedded cortical cells; it also contains the hair pigment melanin. The outermost layer of a hair strand is the cuticle, a dense layer of fat keratinized cells, which protects the hair fiber.5 Virgin human hair accounts for an important portion of the phenotypic variation among the different human races. Typically, three different types of ethnic hair can be categorized: African, Caucasian, and Asian. There are numerous studies analyzing some of the characteristics related to human skin but very few on hair morphology across populations. Although the study of hair properties has largely addressed the fraction of protein within the fibers, only slight differences in the amino acid composition of hairs from humans of different ethnicities have been found, which cannot explain the physical differences that have been previously demonstrated.13 Recently, the importance of lipids within hair fibers has become apparent,14 even though lipids comprise less than 9% of the fiber dry weight.15 Some studies16,17 have been conducted to identify the effects of lipids on the properties of fibers and have found that lipids play a key role in maintaining adequate water permeability and the hydrophobic character of the fiber surface. These effects were primarily related to the lipid amounts and the order within the cuticle region. Changes in the lipid quantity or composition could modulate the dynamics of water in the fiber and thereby generate regions with distinct resistance.18 To determine the presence, distribution, and function of lipids of each ethnicity in depth, the three ethnic fibers were delipidized to study the amount and structure of the lipid fraction in the fibers, their effect on water vapor sorption (DVS) and the calorimetric properties of the extracted lipids. A synchrotron-based Fourier transform infrared microspectrometer (µ-FTIR) was used in this work to study the lipid fraction distribution in hairs of different ethnicities. µ-FTIR allows the spatial determination of the lipids across the different hair layers and differentiates between the lipid order arrangement and quantity.19,20 In particular, this study analyses three virgin ethnic hair fibers and delipidized fibers with a combination of transmittance microscopy FTIR with an ALBA synchrotron infrared source. In addition, the vapor sorption properties of the fibers were determined, and solvent extracted lipids were analyzed and thermically characterized (thermogravimetric (TGA) and differential scanning calorimetric (DSC) analyses) to discern between the different lipid families and lipid properties across the different ethnicity hair regions. 2MATERIALS AND METHODS 2.1 Hair samples The hair samples used in this study were natural virgin brown Caucasian hair and natural dark brown Asiatic hair without pretreatment along with natural kinked Afro hair with thermal pretreatment. All hair samples were provided by De Meo Brothers (Passaic, New Jersey, USA). The hair samples were washed with 3% diluted Pantene Pro- V commercial shampoo (Procter & Gamble, USA) at a hair/surfactant solution 1/30 bath ratio followed by a water rinse and drying under ambient conditions. 2.2 Lipid extractions Lipids were removed by extraction with different solvent mixtures of chloroform/methanol (2:1 v/v, 1:1 v/v, and 1:2 v/v) for 2 h for every mixture and 100% methanol overnight at room temperature in a stirring system using the same hair sample. The different extracts were then combined, concentrated and dissolved in chloroform/methanol (2:1) before analysis. To evaluate the total amount of lipids extracted, 1 mL of each extract was evaporated to dryness in a desiccator under a P2O5 atmosphere and weighed to a constant weight. 2.3 Lipid analyses Lipid analyses of the different extracts were performed with thinlayer chromatography coupled to an automated flame ionization detector (TLC/FID) and an Iatroscan MK-5 analyser (Iatron, Tokyo, Japan). The lipid extracts were directly spotted on silica gel-coated Chromarods (type S-III) by means of a precision Hamilton 2 µL syringe coupled to an SES 3202/IS-02 sample spotter (NiederOLm, Germany). The determination of the lipid content was performed using an optimized TLC/FID protocol14,16 using a methodology where rods (in sets 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BARBA ET AL.77 of 10) were developed using the following mobile phases: 1. chloroform/methanol/water (57:12:0.6, v/v/v) to a distance of 2.5 cm twice, 2. n-hexane/ethyl ether/formic acid (50:20:0.3, v/v/v) to 8 cm, and 3. nhexane/benzene (35:35, v/v) to 10 cm. Finally, a total scan (100%) to quantify the most polar lipids was performed. 2.4 Thermogravimetric analysis Thermogravimetric analyses were performed with a TGA instrument (Model TGA/SDTA 851; Mettler Toledo, Barcelona, Spain). Approximately 5 mg of extract was packed into an aluminum pierced pan (100 µl) and heated under a nitrogen stream from 25 to 550◦Cata heating rate of 10◦C/min. 2.5 Differential scanning calorimetry The samples of lipid extracts (5-10 mg) were placed in a hermetically sealed aluminium pan (40 µL), cooled under a stream of nitrogen from 25 to −100◦C at a cooling rate of 10◦C/min and immediately heated from −100 to 100◦C at a heating rate of 10◦C/min. Differential scanning calorimetry was performed using a Mettler Toledo DSC Model DSC-281 equipped with a Dewar containing liquid nitrogen. 2.6 Dynamic water vapor sorption A thermogravimetric balance, Sorption Analyser Q5000SA (TA Instruments, New Castle, USA) with a controlled humidity chamber, was applied to determine water absorption and desorption in the hair fibers. Experiments were conducted in triplicate on each hair sample (10 ±1 mg) with a total gas flow of 200 mL/min at 25◦C following the protocol described elsewhere.16,17 The method applied by Vickerstaff21 to study the diffusion of dyes within fibers was used to obtain the diffusion coefficient. It is represented by an expression derived from Fick’s equation applied to moisture diffusion. Satisfactory results for the early stages of moisture absorption were obtained with this expression, as in the case of dye diffusion. The fraction of absorbed water plotted against the square root of the absorption time should lie on a straight line in which the slope is the square root of the apparent diffusion coefficient, DA. The apparent diffusion coefficient is measured in min−1over the sample mass. R(t)∕Rf=√DA√t 2.7 Synchrotron-based Fourier transform infrared microspectroscopy (μ-FTIR) and data acquisition Hair tufts of 1 cm were embedded into an optimal cutting temperature (OCT) compound (Bright Instruments, Bedforshire, UK) and immediately frozen using liquid N2.Sampleblockswerecutinto 5µm cross-sections using a Cryostat CM3050 S (Leica Biosystems Nussloch, Germany). Sections were placed on CaF2circular windows with a 1 mm thickness and 13 mm diameter (Cystran, Dorset, UK). µ-FTIR was performed with a MIRAS beamline at the ALBA synchrotron7(Cerdanyola del Vallès, Spain) using a Hyperion 3000 Microscope (Bruker, USA) equipped with a 36×magnification objective and condenser coupled to a Vertex 70 spectrometer (Bruker, USA) with a 50 µm HgCdTe (MCT) detector that was purged with N2gas. The measurement range was 4000–900 cm−1,and spectral collection was carried out in the transmission mode with 4cm −1resolution, 10 µm×10 µm aperture dimensions, a step size of 10 µm and 128 co-added scans. The background spectra were collected from a clean area of the CaF2window every 20 scans. Cross-sections (n =2) of hairs of different ethnicities were simultaneously studied at identical single points within each region from different cross-sections (n =10 per region). Measurements were taken with a 10 ×10 µm aperture to avoid supersaturation, keeping in mind that a large aperture could mean additional noise on the narrow regions (cuticle and medulla). All of the spectra that were obtained were grouped based on the different hair regions from which they were obtained. The average spectra for each hair region were calculated to analyze the CH2sym peak in terms of the peak position and intensity. OPUS software (Version 7.5, Bruker, USA) was used to obtain individual spectra from each region. Classification by precedence (cuticle, cortex or medulla) was performed manually by overlapping the optical microscope image with the measurement points. Due to the small width of the cuticle layer, all spectra from areas inside the hair and close to the border have been considered to be derived from cuticle regions. Cortex and medulla spectra were easily distinguished by their position within the cross-section. Spectra with high saturation in the lipid zone (3000-2800 cm−1) were discarded as well as spectra that contained no signal. Chemical maps were generated using the same software by calculating the second derivatives of the spectra using the Savitsky- Golay algorithm with a second polynomial order and 21 smoothing points to avoid saturated peaks of the map. The integrated areas of the lipid CH2peak symmetric stretching are shown22,23 (CH2sym., 2860-2840 cm-1). Unscrambler X software (Version 10.5, CAMO Software, Norway) was used to perform the statistical analysis. Spectra were corrected using the baseline offset and linear baseline correction algorithms provided by the software.24 Average spectra were calculated with the reduce algorithms provided by the software using the baseline corrected spectra. Score and loading values as well as the mean values were used for data interpretation. Average spectra of all hair types were analyzed using UnscramblerX software. The peak position and amplitude of each spectrum (685 in total) were determined using Gaussian/Lorentzian fitting using a self-written Python code. 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 78 BARBA ET AL. FIGURE 1 Cross-sections observed by optical microscopy for Caucasian hair selected to analyze by µ-FTIR, regions were manually determined 2.8 Statistical analysis Statistical analyses were performed with StatGraphics software (version 5.0). Differences were tested for statistical significance using nonparametric tests (Kruskal-Wallis) with p<.05 considered significant. 3RESULTS AND DISCUSSION An in-depth study of the lipid fraction of different ethnic hairs is proposed in this investigation. For this, the lipid distribution across the structure of the different hairs will be followed; furthermore, the extracted lipids will be analyzed and thermically characterized. Three different types of ethnic hair have been studied: Caucasian, Asian, and African hair. µ-FTIR experiments allowed detection of the vibrations of the alkyl chains of hair lipids. All major conclusions are extracted from the vibrational fingerprint differences in the two frequency domains of the IR spectra. The CH2asymmetric stretching (CH2asym) and the CH2symmetric stretching (CH2sym) peaks, with maxima at approximately 2920 and 2850 cm−1, respectively, are the main peaks that reflect the phases of the lipids. These peaks allow for the measurement of the amount, conformational order and packing of the lipids on hair structure.25 The lipid order plays an important role, as does the lipid content. The fluidity of the lipids can modulate the rates of penetration and departing of substances. Optical microscopy images of hair cross-sections allowed the observation of hair pigmentation, and the three hair layers, cuticle (the outermost dark layer), cortex, and medulla (dark spot on the center), were easily distinguished (Figure 1). Synchrotron-based FTIR measurements were used to obtain FTIR chemical maps of the hair regions. After conversion of the spectra within the map based on the second derivative, FIGURE 2 Chemical map of second derivative obtained at 2850 cm−1(CH2symmetric stretching) of Caucasian virgin hair (a) and Caucasian delipidized hair (b) the integrated values of the CH2sym bands were used to qualitatively map the distribution of the lipids in the different regions. The highest concentration of lipids was found in the cuticle and medulla layers in hair, while lower concentrations of lipids were found in the cortex (Figure 2A). To confirm the attribution of the above-described signals to lipids, lipid extraction was performed on the different ethnic hair samples. FTIR chemical maps of hair regions after lipid extraction were also obtained (Figure 2B). The latter led to a clear disappearance of the lipid bands in all of the different hair regions, thus supporting our previous observations. Our results are in concordance with previous investigations in which higher concentrations of lipids on the medulla and cuticle compared with the hair cortex were demonstrated.26 The average spectra from the different hair regions exhibited the most intense CH2sym peaks on the medulla, followed by those from the cuticle and the cortex for all hairs of different ethnicities (Table 1). This enabled us to understand the hair lipid distribution within the hair fibres. The differences between the lipid peak amplitudes can be 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BARBA ET AL.79 TABLE 1 Mean values ±SD for wavelength values (cm−1) and amplitude (in arbitrary units, a.u) of the CH2symmetric stretching peak (CH2 Sym.) from Caucasian, Asian and African fibres in the cuticle, cortex and medulla Cuticle λ(cm−1) Amount (a.u) Cortex λ(cm−1) Amount (a.u) Medula λ(cm−1) Amount (a.u) Caucasian 2851.57 ±0.52 0.662 ±0.12 2851.56 ±0.52 0.504 ±0.18 2851.55 ±0.70 2.367 ±2.25 Asian 2852.27 ±1.44 0.426 ±0.29 * 2851.66 ±1.23 0.580 ±0.31 2851.44 ±0.93 1.183 ±0.52 * African 2852.68 ±0.61 * 0.946 ±0.20 * 2852.16 ±0.60 * 0.900 ±0.25 * 2851.66 ±1.27 2.894 ±0.71 *p<.05, statistics between the Caucasian hair and the other ethnics hairs (Asian and African). related to the different amounts of lipids present on the studied hair structures. Our results demonstrated more lipids for the African hair samples, followed by the Caucasian hair, with the lowest amount of lipids found in Asian hair. The 2850 cm−1peak is more sensitive, with values between 2847 and 2855 cm−1; a maximum at <2850 cm−1indicates the presence of an orthorhombic (OR) lipid chain conformation, while a hexagonal (HEX) chain conformation produces a maximum between 2850 and 2852 cm−1and a liquid crystalline (LIQ) chain conformation produces a maximum at a wavenumber greater than 2852 cm−1. For that reason, the exact positions and amplitude of the CH2sym peak for each spectrum were analysed (Table 1). The peak positions and amplitudes were obtained from Gaussian-Lorentzian fitting of the CH2sym peak spectra. A lower wavelength value indicated the presence of more highly ordered lipids, and lower amplitude values indicated a lower lipid amount. A wavelength shift between the different ethnic hairs was also observed. Lipids were found to be more structurally ordered in Caucasian hairs, with African hair lipids being the most disordered. This was found in both cuticle and cortex regions, with no difference in the lipid conformational order for the different ethnic hairs in the medulla region (Table 1). Comparing the three regions for each ethnic fiber individually, it can be observed that no differences in lipid order were found when the Caucasian hair was analyzed. However, the CH2sym amplitudes obtained for Caucasian hair showed differences between the three regions; the highest amount of lipids was present in the medulla (2.367 a.u.) followed by the cuticle (0.662 a.u.) and finally the cortex (0.504 a.u.). Lipids in Asian hair cuticle exhibited a slightly lower order based on the CH2sym values (2852.27 cm−1) than in the cortex and medulla (2851.66 and 2851.44 cm−1, respectively). The Asian hair medulla also has the highest amount of lipids based on the CH2sym amplitude (1.183 a.u.), in contrast to the cuticle and cortex (0.426 and 0.580 a.u., respectively). The same behavior was found for African hairs, with the most disordered lipids present in the cuticle (2852.68 cm−1)and the highest amount of lipids located in the medulla region (2.894 a.u) (Table 1). The study of the lipid peaks observed in the synchrotron-based IR spectra enabled the determination of the lipid distribution and order on the different hair regions. Differences in the lipid fraction on the three ethnic hair types were observed. The African hair structure showed the greatest lipid content in all of the hair regions. Asian hair has the lowest amount of lipids, and Caucasian hair has a similar lipid order in all regions, which is higher than that of cuticles of Asian and African types, in accordance with previous studies.14,27,28 TABLE 2 Total amount of lipids obtained by extraction with chloroform/methanol mixtures of Caucasian, Asian, and African hairs Caucasian Asian African Total Lipids (% owf) 1.88 ±0.20 1.48 ±0.14 3.39 ±0.34 FIGURE 3 Quantification of hair lipids in Caucasian (C), Asian (AS) and African (AF) using TLC/FID. Apolar Lipids (APOLAR), free fatty acids (FFA), cholesterol (CHOL) and polar lipids (POLAR) To investigate the lipid presence and conformational order, the three ethnic fibers were delipidized as described in the experimental section. The lipid extracts were characterized, and the delipidized fibers were studied by FTIR-synchrotron. The analyses of the lipid extracts might provide interesting knowledge of the amounts and types of lipids that can be obtained from each hair sample analyzed. Hairs were first washed with shampoo and then extracted with different solvent mixtures of chloroform/methanol and 100% methanol at RT. The different lipid extracts were concentrated and dissolved in chloroformmethanol (2:1) before analysis. The results are expressed as the percentage of the total lipid extract in Table 2. The results of TLC/FID analyses as a percentage of lipids can be seen in Figure 3. It is important to note the highest amount of lipids for African hair compared with Asian and Caucasian hair (Table 2). Although the extracted lipids account for more than 3% of African hair, they account for only approximately 1.5% of Asian hair and 1.9% of Caucasian hair. These results are in accordance with the amplitudes obtained by FTIR in Table 1. To determine the differences in lipid composition between hair ethnicities, the lipid extracts were analyzed in triplicate using TLC/FID (Figure 3). This technique has been recently used to analyze lipids from 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 80 BARBA ET AL. TABLE 3 DSC and TGA lipids analysis results of extracted lipids from Caucasian (C), Asian (AS) and African (AF) hair fibers DSC [T (◦C); ΔH(J/g)] TGA[T( ◦C); mass (%)] C AS AF C AS AF -41.75; 3.42 -31.34; 1.29 -27.72; 34.47 131.41; 0.71 145.43; 2.65 179.46; 6.12 16.85; 71.62 14.44; 2.65 16.87; 199.49 238.86; 45.76 238.76; 60.87 273.11; 66.52 32.65; 17.62 25.51; 163.04 36.17; 17.25 320.66; 39.56 343.46; 21.73 310.52; 22.20 41.29; 9.95 456.86; 5.95 457.26; 7.98 462.86; 1.86 50.52; 22.65 52.77; 0.79 495.16; 0.25 TABLE 4 Maximum moisture regain at 95%RH, mean water regain, mean hysteresis, total time to reach equilibrium (tT) and apparent diffusion coefficient (DA) for native and external extracted ethnic hair fibers Fiber Evaluation Caucasian Asian African NT Lip. Ext NT Lip. Ext NT Lip. Ext Regain at 95% HR (%) 26.5 26.3 23.8 24.4 24.8 26.2 Media Reg (g water/g sample) 0,114 0,115 0,105 0,108 0,105 0,112 Hysteresis (g water/g sample) 0,0243 0,0266 0,0245 0,0248 0,0231 0,0274 tT(min) 3459,2 3556,4 3423,5 3496,1 3223,6 3319,3 DA(min−1×10−3)0,0187 0,0159 0,0188 0,0157 0,0212 0,0208 hair.16 Great differences in the lipid analyses between the three kinds of ethnic hair fibers were found. Related to the whole fiber, the African hair has the greatest proportion of apolar lipids (sterol esters and squalene), with a lower amount of polar lipids. Caucasian and Asian hair fibers have similar amounts of apolar lipids, those of Caucasians have a significantly higher amount of free fatty acids, and those of Asians have more polar lipids. The phase transitions of the extracts could be important for determining the physicochemical properties of the lipids (Table 3). The DSC results indicate that the African lipid extract has main phase transitions at low temperatures (−27 and 16◦C), while the Asian lipid extract has the main phase transition at 25◦C with a small phase transition at 52◦C. This extra phase transition at high temperatures is also found and is more significant for the Caucasian lipid extract at 50◦C. Moreover, TGA evaluation of the three extracts presents two main degradation steps, one more important from 237 to 270◦Candthesecond from 320 to 340◦C. As in the DSC, the Caucasian and Asian lipids have new peaks at higher temperatures, 455 and 490◦C. These results are in accordance with the DSC results. The presence of these peaks at high temperature for Caucasian and Asian lipids can be related to less unsaturation or less fluidity than African lipids.16 Hair water sorption properties can also reflect structural modification of the fibers due to lipid extraction. The water absorption and desorption actions of ethnic hairs with and without lipid extraction were evaluated to determine not only the amount of water absorbed and desorbed in all processes but also their kinetic behavior, which could be related to the lipid order in the bilayers of the hair fibers (Table 4). In accordance with previous results,16,27 a higher moisture content was found for Caucasian hairs than for the other two types of hair. Total lipid extraction clearly increases the water regain on the Asian and the African fibers, in which the extension of the extraction was more accused. Moisture sorption hysteresis occurs between the sorption and desorption processes. Hysteresis is known to occur due to fiber conformation changes with variations in its degree of hydration.29 Keratinized tissues such as hair fibers are known to change significantly when moisture or desorbs from the tissue.30,31 The degree of hysteresis was calculated by subtracting the mass at desorption from that at absorption at the same humidity (Figure 4and Table 4). Virgin Asian fibers were demonstrated to be able to have fewer lipids and have a more important degree of hysteresis than Caucasian and African fibers.27 It is important to remark that the increase in hysteresis is related to the amount of lipids extracted, with African fibers exhibiting much greater modification with lipid depletion. Total lipid depletion, which is more important for Caucasian and African fibers, could make the inner fiber more polar. The hair medulla has been shown to have a rich lipid content fraction mainly for Caucasian and African fibers. Cassie studied the importance of mechanical constraints acting upon sorbed water in wool fibres.32,33 The vapor pressure experienced by the absorbed water is influenced by the mechanical constraint exerted on the water as the fiber swells. Lipid depletion favors water absorption to different degrees depending on the ethnic fiber, with subsequent fiber swelling and fiber constraints inducing a higher degree of hysteresis. In addition, kinetic evaluation of the absorption/desorption process is a good strategy to determine the structural integrity of the fibers. Differences in the keratinized structure of extracted hairs can also be confirmed by evaluation of the diffusion coefficients (Table 4). African hairs present high diffusion coefficients (DA), which is in accordance with the most disordered lipids present in all regions of the fiber (Table 1). There is a marked decrease in diffusion, especially in 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BARBA ET AL.81 FIGURE 4 Hysteresis curves for Caucasian (C), Asian (AS) and African (AF) virgin and lipid extracted hair fibres FIGURE 5 (A) Lipid peak amplitude and (B) Lipid peak position (mean values ±SD) of µ-FTIR analysis before and after lipid extraction for Caucasian (C), Asian (AS) and African (AF) hairs in the cuticle (CU), cortex (CO) and medulla (M). (*p<.05, statistics between hairs non-extracted and lipid extracted) desorption for Caucasian and Asian fibers, with little modification for African fibers. These results can be related to the modification of the bilayer lipid structure, which will be followed by frequency changes based on the CH2sym peak of the lipids. µ-FTIR experiments were performed with the delipidized fibers. Hair lipid extraction ends up with all ethnic fibers with lower amounts of lipids in the three hair regions. Differences before and after lipid extraction in the three hair regions for all of the hairs studied can be graphically visualized in Figures 5A and 5B. The results obtained for Caucasian hair showed that lipid extraction mainly affects cuticle lipids (Figure 5A) with a clear decrease in the obtained µ-FTIR peak amplitude when compared with the other hair regions. When the Caucasian hair is lipid extracted, the resulting extract is mainly composed of FFAs (Figure 3), thus indicating the main localization of this type of lipid on the external part of the Caucasian hair structure. In addition, the lipid chain conformation can also be analyzed before and after lipid extraction (Figure 5B), demonstrating that when the lipids are extracted from the Caucasian hair cuticle, the remaining hair structure presents a more disordered cuticle lipid distribution. Different behaviors were found for the cortex and medulla region, where lipid extraction led to a slightly more ordered lipid distribution. Different behaviors can be observed when analyzing the other ethnic hair samples. When Asian hair is submitted to a lipid extraction procedure, the resultant extract, even though it is low, is mainly obtained from the medulla region (Figure 5A). The Asian lipid extract was richer mainly in polar lipids and FFAs, with a minor content of nonpolar lipids (Figure 3). Differences in the lipid chain conformational order after lipid extraction were also observed; when Asian hair was lipid extracted, the remaining lipids on the hair structure were more orderly distributed in the three hair regions (Figure 5B). In addition, the analysis of lipid extraction from African hair showed an important decrease in the lipid content in the three different hair regions, which was very important in the medulla, followed by the cuticle and the cortex (Figure 5A). More extracted lipids were obtained for African hair than for Caucasian and Asian hair (Table 2), with this extract being rich mainly in apolar lipids (Figure 3). For Caucasian and Asian hair, the conformational order of the remaining lipids on African hair after the extraction procedure was followed, and the results showed very similar lipid distributions in all regions not affected by delipidization. In summary, the study of the lipid peaks observed in the synchrotron-based IR spectra enabled the determination of the lipid distribution and order on the different hair regions. Differences in the lipid fraction on the three ethnic virgin hair types were observed. The African hair structure showed the greatest lipid content in all of the hair regions. Asian hair had the lowest amount of lipids, and Caucasian hair had a similar lipid order in all regions, with it being higher than those of cuticles of Asian and African types. Moreover, the three ethnic fibers were delipidized, the lipid extracts were characterized, and the delipidized fibers were studied by DVS 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 82 BARBA ET AL. and FTIR-synchrotron. In accordance with the amplitudes obtained by FTIR, it is important to remark that the highest amount of lipids was extracted from African hair compared with Asian and Caucasian hair. African hair has the greatest proportion of apolar lipids (sterol esters and squalene), Caucasian hair has a significantly higher amount of free fatty acids, and Asian hair has more polar lipids. The TGA and DSC studies suggests the presence of peaks at high temperature for Caucasian and Asian lipids, indicating higher lipid saturation than that corresponding to African lipids, which could have more unsaturation and more fluidity.16 The determination of the physicochemical properties of the extracted lipids revealed differences in the degree of unsaturation of the extracted lipids from the different ethnic hairs. In accordance with previous works,26,27 differences in the degree of unsaturation of the lipid extracts can be related to the fluidity of the lipids and therefore related to the fiber diffusion properties. Lipid bilayers are known to have fluid and non-fluid zones depending on the nature of the lipids that form them, which account for permeability differences. African extracts have proven to be formed by more unsaturated lipids, and these kinds of lipids are responsible for leading to more fluid lipid zones. A synchrotron-based Fourier transform infrared microspectrometer (µ-FTIR) was used in this work to study the lipid fraction distribution in different virgin and delipidized ethnic hairs. The spatial determination of the lipids across the different hair layers allows differentiation between the lipid order arrangement and quantity. Its relationship with the vapor sorption properties of the fibers and the chemical and thermic characterization of the extracts could investigate the role of lipids in the hair and ethnic differentiation in depth. The FTIR-synchrotron study indicates the preferential extraction of cuticular lipids for Caucasian hairs with a disorganization of its structure and a lower lipid extraction of Asian lipids in all regions without structural disorganization, in contrast to a tendency to increase its organization and a maximum extraction of the three regions of African hair with little modification of the lipid order in any region. The lack of order modification and water diffusion for the African fibers from which most lipids are extracted is difficult to explain. Perhaps the high degree of fluidity of the lipids diminishes the main barrier effect of the lipid bilayers for water permeability. In contrast, the low fluidity of the more saturated lipids from Caucasian and Asian fibers has more implications for the water barrier function, and their extraction somehow leads to a more packed lipid structure, inducing lower water diffusion. Then, the higher order on the lipid bilayer of Caucasian and Asian delipidized fibers hinders water permeation kinetics in some way. The different presence and organization of the lipids in the different regions of the African hair can account for its differentiation with regards to moisturization and swelling compared to the other types of fibers. 4CONCLUSIONS The lipid amount and organization in the different hair regions of the studied ethnic fibers can explain some of their barrier properties. African virgin hair was demonstrated to have more lipids mainly in the medulla, which implies an important hydrophobicity with low hysteresis between the absorption and desorption of water vapor processes. In addition, these lipids are highly disordered, mainly in the cuticle, which can be related to their high water vapor diffusion. Asian and Caucasian virgin hairs presented a similar lipid order in all regions, with similar diffusion coefficients. Analyses of delipidized hairs indicate a maximum extraction of the medullar region of the African hair. This is supported by an increase in moisture of the African lipid-depleted fiber, which favors the degree of hysteresis. This also happens for the Caucasian fibers to a lesser extent. The increase in water content at the interior of the fiber promotes swelling and fiber constraints, inducing a higher degree of hysteresis. Moreover, the absence of modification of the lipid order in any region of the delipidized African fiber is reflected in the similar water diffusion to the African virgin diffusion. In contrast, the increase in the order in most regions of the delipidized Asian and Caucasian fibers is supported by the marked decrease in diffusion. This result indicates that the higher order of the lipid bilayer hinders water penetration kinetics. There is a wide gap in the understanding of the structure of ethnic and particularly African hair. The study of the presence and bilayer organization of the lipids in the different regions of the fibers can open new insights to explain their different properties including moisturization, swelling, and so on. The results from this work can have particular relevance to chemists and new-product development specialists who work in the ethnic hair-care market. ACKNOWLEDGMENTS Support for this work from CTQ CTQ2018-094014-B-100 is gratefully acknowledged. Moreover the authors acknowledge the MIRAS beamline at ALBA Synchrotron for the grant and the ALBA staff for their expert assistance. ORCID Clara Barba https://orcid.org/0000-0002-3084-2925 REFERENCES 1. Amigo JM. Practical issues of hyperspectral imaging analysis of sòlid dosatge forms. Anal Bioanal Chem. 2010;398:93-109. 2. Holman H, Martin M, McKinney W. Tracking chemical changes in a live cell: Biomedical applications of SR-FTIR spectromicroscopy. Spectroscopy. 2003;17; 139–59. 3. Stuart B, Ando DJ. Biological Applications of Infrared Spectroscopy, Chichester, New York, Published on behalf of ACOL (University of Greenwich) by John Wiley, 1997. 4. Holman H, Martin M, McKinney W. Synchrotron-Based FTIR Spectromicroscopy: Cytotoxicity and Heating Considerations. Journal of Biological Physics. 2003;29; 275–86. 5. Wetzel DL, LeVine SM. Imaging molecular chemistry with infrared microscopy. Science. 1999;285;1224-5. 6. Marcelli A, Cricenti A, Kwiatek WM, et.al. Biological applications of synchrotron radiation infrared spectromicroscopy. Biotechnology Advances. 2012;30; 1390–404 7. Yousef I, Ribó L, Crisol A, et.al. MIRAS: The Infrared Synchrotron Radiation Beamline at ALBA. Synchrotron Radiation News. 2017; 30. 4–6. 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BARBA ET AL.83 8. Holman H, Martin M, Blakely E, et.al. IR spectroscopic characteristics of cell cycle and cell death probed by synchrotron radiation based Fourier transform IR spectromicroscopy, Biopolymers: Biospectroscopy. 2000;57;329. 9. Miller LM, Tibrewala J, Carlson CS, Examination of bone chemical composition in osteoporosis using fluorescenceassisted synchrotron infrared microspectroscopy, Cellular and Molecular Biology. 2000;46;1035 10. Briki F, Busson B, Kreplak L, et.al. Exploring a biological tissue from atomic to macroscopic scale using synchrotron radiation: Example of hair, Cellular and Molecular Biology. 2000;46; 1005 11. Geller JT, Holman HY, Su G, et.al. Flow dynamics and potential for biodegradation of organic contaminants in fractured rock vadose zones, Journal of Contaminant Hydrology. 2000; 43; 63. 12. Xie AH, He Q, Miller L, et.al. Chance, Low frequency vibrations of amino acid homopolymers observed by synchrotron far-IR absorption spectroscopy: Excited state effects dominate the temperature dependence of the spectra, Biopolymers. 1999; 49; 591 13. Raab TK, Martin MC. Visualizing rhizosphere chemistry of legumes with mid-IR synchrotron radiation, Planta. 2001; 213; 881. 14. Wolfram LJ. Human hair: a unique physicochemical composite. J Am Acad Dermatol. 2003; 48: 106-14 15. Martí M, Barba C, Manich AM, et.al. The influence of hair lipids in ethnic hair properties. Int J Cosmet Sci. 2016; 38(1): 77-84. 16. Masukawa Y, Narita H, Imokawa G. Characterization of the lipid composition at the proximal root regions of human hair. J Cosmet Sci. 2005; 56: 517-16. 17. Coderch L, Oliver MA, Martínez V, et.al. Exogenous and endogenous lipids of human hair. Skin Res Technol. 2017; 23(4): 479-85. 18. Barba C, Martí M, Carilla J, et.al. Moisture sorption/desorption of protein fibres. Thermochim Acta. 2013; 552: 70-6 19. Ji JH, Park TS, Lee HJ, et al. The ethnic differences of the damage of hair and integral hair lipid after ultra violet radiation. Ann Dermatol. 2013; 25: 54-60. 20. Kreplak L, Briki F, Duvault Y, et.al. Profiling lipids across Caucasian and Afro-American hair transverse cuts, using synchrotron infrared microspectrometry. International Journal of Cosmetic Science. 2012; 23; 369–74. 21. Mansour R, Sallam A, Hamdan I, et.al. Elucidation of penetration enhancement mechanism of Emu oil using FTIR microspectroscopy at EMIRA laboratory of SESAME synchrotron. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2017;185. 22. Vickerstaff T. The physical chemistry of dyeing. London: Oliver and Boyd; 1954 23. Roke S, Schins J, Müller M, et.al. Vibrational Spectroscopic Investigation of the Phase Diagram of a Biomimetic Lipid Monolayer. Physical Review Letters. 2003; 90, 128101. 24. Martin FL, Kelly JG, Llabjani V, et.al. Distinguishing cell types or populations based on the computational analysis of their infrared spectra. Nature Protocols. 2010; 5; 1748–60. 25. AS, CAMO SOFTWARE. [Online] [Cited: November 20, 2018.] https: //www.camo.com/downloads/usermanuals.html 26. Nishimura K, Nishino M, Inaoka Y, et.al. Interelationship between the hair lípids and their moisture. J. Cosmet Sic Soc Jpn. 1989; 13: 134–9 27. Zhang G, Senak L, Moore DJ. Measuring changes in chemistry, composition, and molecular structure within hair fibers by infrared and Raman spectroscopic imaging. J. of Biomedical Optics. 2011; 16(5); 056009. 28. Oliber MA, Coderch L, Carrer V, et.al. Ethnic hair: Thermoanalytical and spectroscopic differences Skin Res Technol. 2020;00:1–10. 29. Leerunyakul K, Suchonwanit P, Asian hair: a review of structures, properties, and distinctive disorders. Clinic, Cosmetic and Investigational Dermatology. 2020; 13; 309–18 30. Anderson, RB, Hall, WK. Modifications of the Brunauer, Emett and Teller equationII. J. Am. Chem. Soc., 1948; 70: 1727–34 31. Syed A, Kuhajda A, Ayoub H, Ahmad K, Frank EM, African-American hair: its physical properties and differences relative to Caucasian hair. Cosmet Toil. 1995; 110:39-48 32. Hel Heldman DR, Hall CW, Hedrick I. Vapor equilibrium relationships of dry milk. Dairy Sci. 1965; 48: 845–52 33. Cassie A.B.D. Trans Faraday Soc. 1945, 41,458 How to cite this article: Barba C, Oliver MA, Martí M, Kreuzer M, Coderch L. Lipid distribution on ethnic hairs by Fourier transform infrared synchrotron spectroscopy. Skin Res Technol. 2022;28:75–83. https://doi.org/10.1111/srt.13093 16000846, 2022, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/srt.13093 by Readcube (Labtiva Inc.), Wiley Online Library on [14/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License