Effects of peptide treatments on thermal properties of hair After seeing notable changes in mechanical properties, we tested thermal properties of hair expecting to observe significant changes in DSC parameters. The denaturation temperature and enthalpy of virgin hair were 149.7 Æ 0.2°C and 19.9 Æ 1.3 J gÀ1, respec- tively. The damaging effect of relaxing treatments was significant: all of the relaxed hair peaks flattened. The dramatic change in the curves shape made it difficult to define the baseline limits which are required to calculate the enthalpies. The denaturation enthalpy of relaxed hair varied between 3 and 5 J gÀ1. The denaturation temperature results are presented in Fig. 9. Statistical analysis did not show a difference in denaturation temperature of differently treated relaxed hair, with the exception of the mid-MW peptide treated sample. Here, a statistically significant (but relatively small) increase in denaturation temperature (1.7°C) was observed when comparing mid-MW peptide treated hair with the relaxed hair control. Discussion Fluorescence microscopy Fluorescence microscopy techniques, including confocal fluores- cence microscopy, have been used for many years for the charac- terization of hair and for examining the penetration of fluorescently tagged materials [22, 23]. Unfortunately, there is a large amount of natural autofluorescence that can arise from chromophores in the hair, including amino acids with aromatic side chains: trypto- phan, tyrosine and phenylalanine. Autofluorescence can also arise from the breakdown products of these amino acids and, of course, melanin [24]. McMullen et al. have created a series of excitation– emission matrices for hair fibres and have shown that pigmented hair fibres have two major excitation/emission peaks. One peak that is related to tryptophan has an excitation wavelength at 290 nm and an emission wavelength 335–345 nm (the exact posi- tion depending on the level of pigmentation). Furthermore, they report another peak with an excitation wavelength of 366 nm and an emission wavelength of 433 nm this is related to tryptophan degradation products [25]. Background autofluorescence can be a major problem for pene- tration studies such as this one, as it can create a false impression of the presence of labelled active in the images. For this reason, the present study used a red fluorescent label that emitted at 579 nm, well above the wavelengths seen with hair autofluores- cence. Background autofluorescence was further reduced by look- ing at treated samples just at 570–640 nm. Despite these efforts, a small amount of background autofluorescence was still observed, and care was taken to take images of control and trea- ted sections with the same microscope and camera settings. Differ- ences in fluorescence were then certainly related to the absorbed peptides. Penetration of keratin peptides Hair fibres have, through evolution, developed a natural barrier to the absorption of chemicals and micro-organisms called the cell membrane complex (CMC). The CMC runs between all the cells in the hair and comprises two lipid layers sandwiching a protein-rich layer, known as the d-layer [26]. The CMC is the only continuous structure in the hair fibre. Recent studies suggest that the CMC between the cuticle and the cortex contains an additional, highly resistant structure that prevents penetration of materials into the cortex [27] therefore, many compounds pass very slowly through this barrier, unless the hair is damaged. The routes of penetration of materials into the hair have been visualized by a number of studies [22, 23, 28]. All these studies show that penetration through the CMC is the main pathway for delivery of actives into the hair cortex. Consistent with these find- ings, kinetic studies investigating uptake of materials into the hair also show that small molecules penetrate the hair best when they are less ionized, and so, more able to travel through and along the lipid bilayers in the CMC [29]. SEM examination of virgin and relaxed hair revealed an additional pathway that may be common in textured hair: large cracks and pores as a result of pre-existing mechanical damage (Fig. 5). The penetration to cortex via mechan- ically damaged surface may be significantly more effective than just sodium hydroxide affected cuticles and CMC. The results of the dimension measurement study show that only low-MW actives penetrated into the relaxed hair cortex in large Figure 9 Denaturation temperature as a function of treatment and hair condition (means Æ standard errors). Results that are statistically different in compar- ison with relaxed hair control (DI water) are labelled with the letter a, and results that are significantly different from virgin hair control (DI water) are labelled with the letter b. © 2020 The Authors. International Journal of Cosmetic Science published by John Wiley & Sons Ltd on behalf of Society of Cosmetic Scientists and Societe Francaise de Cosmetologie 9 International Journal of Cosmetic Science, 1–12 Larger hydrolyzed keratins reduce hair breakage E. Malinauskyte et al. 153.0 --------------------- c ~ 150.0 a g e 141.0 ___________________ __, ~ 2 144.0 ab ::::, ro ro al 141.0 b b b C C.. £3 ~ 138.0 - i-. . 132.0 Relaxed Relaxed Relaxed Relaxed Hair DI Hair 1 % Hair 1 % Hair 1 % Water Leucine Low Mw Mid Mw Keratin Peptide Keratin Peptide b Relaxed Virgin Hair Hair 1 % DI Water High Mw Keratin Peptide