An ex vivo comparison of the tensile strengthening properties of protein derivatives on damaged hair

An ex vivo comparison of the tensile strengthening properties of protein derivatives on damaged hair Nicholson Sherine 2, Daniels Gabriela 1, Grant-Ross Peter 1, Tamburic Slobodanka 1 1 School of Management and Science, London College of Fashion, University of the Arts London, UK 2 Primp. Nails & Beauty Ltd, London, UK Abstract Conventional conditioning agents, cationic surfactants and polymers, are often limited in their ability to repair and strengthen the hair fibre, while protein-derived actives have been found to enhance the tensile strength of damaged hair. This study investigates the effects of keratin, wheat and collagen hydrolysates on the tensile strength of three types of damaged hair: bleached, permed and thermally treated. The investigated actives were Hydrolysed Keratin (HK; 600Da), Hydrolysed Wheat Protein (and) Wheat Starch (HWP/WS; 1,500Da) and Hydrolysed Collagen (HC; 9,000Da). The actives were formulated into an emulsion base, including Cetrimonium Chloride at 1% w/w and non-ionic emulsifiers. Further to a preliminary concentration response study, the actives were formulated at the following concentration levels: HK - 0.5%, HWP/WS - 2% and HC- 2%w/w. Virgin Caucasian brown hair tresses were used and damaged following standardised bleaching, perming and repetitive thermal straightening protocols. Three test formulations and a control conditioner were applied to each type of damaged hair tresses, respectively. The treatment time for each tress was 5 minutes at room temperature, followed by rinsing and blow-drying for 5 minutes at 50 C. The tensile strength of wet and dry hair fibres was measured before and after treatment using the TA.XT Plus Texture Analyser (Stable Micro Systems, UK). Three-way ANOVA (damage type, type of active, wet/dry state) and two-way ANOVA (type of active, wet/dry state) were used for statistical analysis of the data followed by Tukey Honest Definition Test (THDT) where appropriate. All actives elicited improvement in hair tensile strength, irrespective of the nature of the damage. HK produced the highest significance in tensile strengthening in all cases, whilst HWP/WS and HC were most efficient in permed and heat-damaged hair. Bleached hair showed statistically significant tensile strength improvement in both wet and dry state, with HK and HWP/WS as the most efficient actives. The improvement in tensile strength of permed and thermally treated hair was only statistically significant in the wet state. The results have shown that hydrolysed proteins are effective in improving the tensile strength of wet and dry damaged hair fibres. 1. INTRODUCTION Chemical processing of hair (e.g. bleaching, permanent waving/straightening and colouring) is known to damage the hair fibre by reducing its structural integrity and strength. In addition, a frequent use of thermal styling tools has also been proven to lead to the surface and structural degradation. The mechanical resilience of the hair fibre at molecular level is due to complex protein structures enforced by multiple types of bonding: covalent bonds (disulfide and isopeptide bonds), ionic bonds, hydrophobic forces, and hydrogen bonds. The latter are responsible for maintaining the 1

water content of the hair fibre, offering elasticity and plasticity (Feughelman, 1997a, Swift, 1997). In the presence of heat, which can reach 200 C, internally bound water evaporates and, under mechanical strain, hydrogen bonds and other weak interactions between the protein chains break and reconfigure. A conformational change from α-helix to β-helix, detected at high temperatures, also contributes to the changes in hair structure. At a macrostructural level, the mechanical properties of the hair largely reflect the integrity of the intermediate filaments and the surrounding matrix of the cortex. Since chemical and heat processing largely affect these regions, tensile testing could be employed to assess the degree of damage caused to the hair fibre. The mechanism involves stretching a fibre of a known length at a fixed rate to a breaking point via load-elongation. The degree of hair damage is reflected as a decrease in the force required to break the hair fibre, relative to undamaged hair. Robbins (2012) has pointed out to cuticle separation as the first stage of damage when stretching fibres, suggesting several different fracture patterns of human hair, specifically in relation to wet/dry states. Breaking wet hair generally involves fractures in the hydrophilic layers of Cellular Membrane Complex and is clear cut, whilst dry fractures involve hydrophobic layers. This suggests that the efficacy of treatments in not limited to the cortex. Conventional conditioning agents, such as cationic polymers, are considered too large to penetrate into the cortex (Ruetsch and Kamath, 2005). Protein-derived conditioning actives, on the other hand, share similarities with the amino acid structure of the hair and have been found to enhance tensile strengths of damaged hair (Teglia and Secchi, 1999). Protein hydrolysates and peptide mixes are therefore technically preferred choice for hair treatments. The natural affinity of these ingredients to the hair keratin is due to the numerous bonding sites and reactive groups in their structure. The penetration of external materials into the hair occurs mainly through diffusion, its rate pending on factors such as hair surface porosity, the ratio of acidic to basic amino acids at the hair surface and the cross-link density of the protein layers in the cuticle and cortex (Robbins, 2012). The ph of the treatment and the isoelectric point of the hair also play a fundamental role. Proteins which in acidic environment show an excess of positive charge, will have a greater affinity to hair. Thus, the subtle differences in the amino acid profiles of the treatment actives could result in differences in their conditioning effects. This study aims to compare the effects of three protein-based conditioning actives (keratin, wheat and collagen hydrolysates) on the tensile strength of three types of damaged hair: bleached, permed and thermally treated. 2. MATERIALS AND METHODS 2.1. Materials Protein hydrolysates with different amino acid profiles, reflecting their different origins (plant and animal), and different molecular weights were selected for this investigation. The investigated materials INCI designations, molecular weights and optimal ph ranges are listed in Table 1. 2

Table 1: Protein-derived conditioning actives Source/ Derived from* INCI Name (% active as supplied) Average Molecular Weight * Animal/ Wool Hydrolyzed Keratin (20%) 600 4.0-5.0 (Da) ph * Plant/Wheat Hydrolyzed Wheat Protein (and) Wheat Starch (85%) 1,500 4.0-5.0 (in 10% water) Animal (Bovine hide) /Collagen * Information as provided by respective suppliers Hydrolyzed Collagen (95%) 9,000 5.5-6.5 (in 10% water) Each investigated material was added to the stable cationic emulsion base (Table 2), at the following active concentration levels: hydrolised keratin (HK) - 0.5%w/w; hydrolised wheat protein (and) wheat starch (HWP/WS) - 2%w/w; hydrolised collagen (HC) - 2%w/w. These active levels were identified in a preliminary concentration response study, employing a testing method identical to this investigation. Table 2. Formulation of the cationic base emulsion (control) INCI Function % w/w Aqua Solvent to 100 Glycerin Humectant 3.00 Cetrimonium Chloride (30%) & Aqua Mono-alkyl cationic conditioner and emulsifier 1.00 Cetearyl Alcohol Non-ionic emulsifier 5.00 Ceteareth-20 Non-ionic emulsifier 2.00 Propylene Glycol (and) Diazolidinyl Urea (and) Methylparaben (and) Propylparaben Preservative (Carried in propylene glycol solvent) 0.50 ric Acid (20%) ph regulator q.s. The ph of all formulation variables was adjusted to 4.0(+/-2). Caucasian virgin hair tresses, weighting 1.5g (+/-1g) and of 150mm in length, were used as treatment substrates. Professional bleaching and permanent waving products were used to damage the hair. 2.2. Methods Hair damage treatments Bleaching: four hair tresses were treated with a commercial professional hair bleaching product, containing 12%H2O2, for 60 minutes, according to the manufacturer s instructions. Permanent waving: four hair tresses were treated with a commercial professional hair perming kit comprising a perm lotion and a neutraliser for normal and resistant hair. The active levels of Ammonium Thioglycolate and H2O2 was not indicated by the supplier. Repetitive thermal straightening: four tresses were treated following a protocol for heat treatment 3

developed by McMullen and Jachowicz (1998) comprising four consecutive cycles of washing, blow drying and intermittent applications of flat straightening ceramic iron for a total of 3 minutes each. Ghd ceramic straighteners (210 C) were used. Six wet single hair fibres were selected randomly from each damaged wet tress (following the final rinse) and removed for immediate wet tensile strength testing. The hair tress was then blow dried at 50 C for 5 minutes, and six single fibres were removed for dry tensile strength testing. Figure 1: Dark brown virgin hair tress before (left) and after bleaching with 12% H 2 O 2 for 60 minutes (right) Figure 2: Dark brown virgin hair tress before (left) and after permanent waving (right) Conditioning treatments of the damaged hair 2ml of Sodium Laureth Sufate was applied to a damaged tress, massaged for 30 seconds and the hair was rinsed under running water (35 C) for 1 minute. A conditioning treatment (0.7g per 1g of hair) was then applied, massaged for 30 seconds and left on the hair for 5 minutes. The tress was rinsed off for 1 minute under running water (35 C). Six wet single hair fibres were selected randomly from the tress and removed for immediate wet tensile strength testing. The hair tress was then blow dried at 50 C for 5 minutes, and six single fibres were removed for dry tensile strength testing. Four treatments (control and three active formulations) were applied to tresses representing each hair damage type, respectively. In total, 12 variations of hair damage and respective treatments were measured for each state (wet and dry). Conditioning treatment of bleached hair under elevated temperature 2ml of Sodium Laureth Sufate was applied to a bleached tress, massaged for 30 seconds and the hair was rinsed under running water (35 C) for 1 minute. A conditioning treatment (0.7g per 1g of hair) was applied to the hair tress and massaged for 30 seconds. The tress was then placed in a towel steamer (104 C) for 5 minutes (Beauty Pro Hot Towel Steamer HTS I). The tress was rinsed off for 1 minute under running water (35 C). This treatment was adapted from ISP Hair Care Applications Test Method (Laryea, 2001). Six wet single hair fibres were selected randomly from each wet tress and removed for immediate wet tensile strength testing. The hair tress was then blow dried at 50 C for 5 minutes, and six single more fibres were removed for dry tensile strength testing. All three protein-based active treatments were tested. 4

Tensile Strength Testing Single fibre tensile testing, was carried out with TA.XTPlus Texture Analyser and the results were recorded by the Texture Exponent Software (Stable Microsystems, UK). A single hair fibre (length=55mm) was stretched at constant speed of 15mm/s. The applied force (F), measured in Newton, which caused the hair to break was recorded. The results were analysed using Analysis of Variance (ANOVA) for multiple factors, followed by a Tukey Honest Significant Difference (THSD) test, using the R programming language. Probability values p <0.05 were considered statistically significant. Table 3 outlines the number of factors of analysis per each stage of the investigation. Table 3: Statistical tests used for each experimental stage Stage Type of ANOVA test Factors Variables within factors 1 2-way ANOVA Hair damage treatments vs. virgin hair tress damage type tress state bleached, permed, heat tress state: wet/dry 3-way ANOVA treated damage types bleached, permed, heat Treatments of damaged hair vs. corresponding damage type tress state protein active type tress state: wet/dry 2 2-way ANOVA Protein active treatment vs conditioning base treatment treated damage types protein active type protein active type (wet and dry combined ) vs condition base 3-way ANOVA tress state (wet/dry), tress state (wet/dry), 3 Bleached hair protein active treatments room temperature vs. elevated temperature protein active type (wet and dry combined) temperature protein active type (wet and dry combined) temperature In addition to the statistical analysis of break force, the tensile strength increase for each treatment was calculated using the following formula: % tensile strength increase = (Fd Ft) / Fd Ft=break force of treated damaged hair; Fd=break force of damaged hair 3. RESULTS & DISCUSSION Tensile strength of damaged hair Analysis of damaged hair types (wet and dry) was carried out in order to identify the comparative level of tensile strength reduction and to verify the sensitivity of the test. The comparison of the mean tensile break-strengths for each type of damaged hair are presented in Figure 4. Two-way ANOVA confirmed significant difference amongst hair damage types (p<0.05); the THDT paired comparison test indicated significant differences between each damage type and virgin hair, in both wet and dry state. 5

Figure 4. Comparison of the mean tensile strengths of virgin, bleached, permed and thermally straightened (thermal) hair in wet and dry state. () p<0.001, (**) p <0.01 Statistically significant differences in tensile break strength between the wet and dry state of each hair type were also detected by THDT: virgin hair (wet/dry pair) p=0.01, bleaching, perming and thermal treatments (wet/dry respective pairs) p=0.001. The most notable reductions in tensile break strength of damaged hair were in line with the literature data, although differing in magnitude. For example, Robbins (2012) suggests that a 15% decrease in wet tensile strength is typical for a normal permanent wave treatment, while 32% decrease was measured in this study. One possible explanation is over-processing of the hair tress due to the use of a very strong perming product. The bleached hair exhibited a 41% tensile strength reduction in the dry state, which also suggests over-processing of hair. According to Robbins (2012), a change in the dry strength of bleached hair is most evident in excessively processed hair, after 60% of the cystine molecules have been oxidised. Reduction of the strength of thermally straightened hair fibres appeared to be proportional in both states. According to Milczarek et al (1992), melting of the cortex crystalline structures is expected to occur at high temperatures (>185 C), which might render the hair tensile strength reduction irreversible from wet to dry state. In summary, the tensile strength tests confirmed the method s sensitivity to the types of hair damage and the wet/dry states of hair. Tensile strength of damaged hair after conditioning treatments Statistical analysis of the tensile strength of post-treatment hair was carried out to determine the most efficacious active for each hair damage type. The test results of all protein active treatments were statistically compared to: (a) untreated damaged hair and (b) the cationic base emulsion (control) treatment. Results for Bleached Tresses (a) Compared to untreated bleached hair The results, presented in Figure 5, show that the treatment of tresses with the base emulsion did not significantly improve tensile strength. The HK treatment produced the highest overall 6

percentage increase and the lowest p-values for both wet and dry treatment, followed by HC. Figure 5: Break point force (N) and tensile strength percentage increase for bleached tresses after treatments, for wet and dry state, respectively; ()p=0.000, (**)p=0.001, (*)p=0.01 1.6 65% Force (N) 1.2 0.8 12% 13% 54% ** 17% 39% ** 24% * 43% 0.4 0 Wet Dry Wet Dry Wet Dry Wet Dry Base Conditioner Keratin Wheat Collagen (b) Compared to cationic base emulsion Both wet and dry states of the active treatment were combined and compared to the cationic base emulsion (control) treatment using 2-way ANOVA and THDT. No statistically significant differences were found between control and HWP/WS, control and HC, while a significant difference was found between the base emulsion and HK (p=0.003). In summary, the results for bleached hair infer that HK is overly the most efficacious treatments, followed by HC. Thus, the protein actives of animal origin appeared to be more effective. Results for Permed Tresses (a) Compared to untreated permed hair As shown in Figure 6, HK produced the most significant tensile strength increase in the wet state, followed by HC. The HWP/WS treatment resulted in significant tensile strength improvement in both dry and wet state. A decrease in tensile strength was observed for tresses treated with the base emulsion in both states, though statistically the difference was not significant. (b) Compared to cationic base emulsion Both wet and dry states of the active treatment were combined and compared to the cationic base emulsion treatment and analysed via 2-way ANOVA and THDT; Statistical differences from control were found for all treatments, i.e. HK p=0.001, HWP/WS p=0.000, HC p=0.02. 7

Figure 6. Break point force (N) and tensile strength percentage increase for permed tresses after treatments, for wet and dry state, respectively; ()p=0.000, (*)p=0.01 1.6 1.2 85% 7% 39% 12% * 38% 5% Force (N) 0.8 0.4 * 0-0.4-3% -24% Wet Dry Wet Dry Wet Dry Wet Dry Base Conditioner Keratin Wheat Collagen The most effective treatment overall for permed hair was HWP/WS, improving tensile strength in both wet and dry state. The results infer that the wet hair tensile strength of permed hair can be improved by all investigated actives and that there is no relationship between efficacy and molecular size or origin of the active. Results for thermally damaged tresses Figure 7: Break point force (N) and tensile strength percentage increase for thermally damaged tresses after treatments, for wet and dry state, respectively; ()p=0.000, (*)p=0.01 1.6 Force (N) 1.2 0.8 23% 3% 76% 18% 44% 29% 67% * 23% * 0.4 0 Wet Dry Wet Dry Wet Dry Wet Dry Base Conditioner Keratin Wheat Collagen 8

(a) Compared to untreated thermally damaged hair The HK produced the highest increase in the wet state and HWP/WS in the dry state. The base emulsion did not show any significant differences in either state. (b) Compared to cationic base emulsion Both wet and dry states of the active treatment were combined and compared to the cationic base emulsion treatment and analysed via 2-way ANOVA and THDT; significant differences were shown across all actives, with HC being the most effective (p=0.000). HWP/WS and HC were both deemed efficacious for treatment of thermally damaged tresses, due to their overall efficacy. Thus higher molecular weight protein actives appear to enhance the tensile strength of thermally damaged hair more effectively than lower molecular weight proteins. Results for bleached tresses exposed to heat during treatment Bleached tresses were also treated with actives at their optimal concentrations (HK- 0.5%, HWP/WS- 2% and HC- 2%) for 5 minutes at 104 C to determine if temperature would accelerate their diffusion into the cortex. The results are presented in Figures 8 and 9 for wet and dry measurements respectively. Figure 8: Comparison of the wet tensile break-strengths of bleached tresses treated for 5 minutes at room temperature and 104 C; (**) p=0.001, (*) p=0.01 For the wet state (Figure 8) HWP/WS and HC treatments showed a positive response with the increase in temperature. The THDT paired comparisons for each treatment at high and room temperatures were statistically different. Heat treatment resulted in a net negative effect on the tensile strength of tresses treated with HK, although this was not statistically significant. 9

Figure 9: Comparison of the dry tensile break-strengths of bleached tresses treated for 5 minutes at room temperature and 104 C; () p=0.000, (*) p=0.01 In the dry state, HK and HC treated tresses (Figure 9) at elevated temperature showed statistically higher tensile strength compared to their respective room temperature treatments. The results in both states suggest that investigated conditioning treatments were temperature dependent; however, further investigation of the effects of the base emulsion itself are needed in order to make a definitive conclusion on the role of actives under these conditions. 4. CONCLUSION The aim of this study was to evaluate the efficacy of hydrolysed proteins in the improvement of the tensile strength of damaged hair. It has been confirmed that the chemical and heat processing significantly reduced the force requited to break the hair fibres, when compared with virgin hair. The overall results show that almost all test treatments have shown statistically significant improvement compared to the untreated damaged hair and hair treated with the base emulsion formulation (control). These results infer that hydrolysed proteins are effective in improving the tensile strength of wet and dry hair fibres that have been chemically or thermally damaged. Bleached hair showed the most significant response to all treatments in both wet and dry states, particularly to hydrolysed keratin (HK), which contains high quantities of cysteine-rich residues and has lower molecular weight compared to the other two actives. The result infers that hydrolysed keratin has better affinity to hair compared to other two actives, specifically to bleached hair, due to its increased porous regions and lower cross-link density. Furthermore, the hydrolised keratin treatment response to the changing temperature suggests a higher mobility of the active and weaker bonding to the hair at high temperatures, caused by its lower molecular weight and higher water affinity. Hydrolysed keratin improved the wet tensile strengths of the permed and heat treated hair, thus emerging as the most efficacious treatment for wet hair. Hydrolysed wheat protein (and) Wheat Starch (HWP/WS) was most effective in improving the tensile strength of dry hair, specifically permed and heat treated. This effect could be partly attributed to the presence of wheat starch, which might strengthen the hair fibre surface upon drying. Hydrolised collagen (HC) has also provided tensile benefits to all three types of damaged hair, but has offered most consistent effect on thermally treated hair. 10

In can be concluded that a treatment comprising one type of hydrolysed protein is unlikely to deliver optimal tensile strengthening to both dry and wet hair, regardless of the nature of hair damage. Therefore, further work should include an assessment of the combination of small and high molecular weight protein-based actives. Theoretically, this combination should provide attachments to multiple locations on both cuticle and cortex, efficiently increasing hair strength. REFERENCES Buehler, M. J. and Yung, Y. C. (2010) How protein materials balance strength, robustness and adaptability Human Frontier Science Program Journal. Vol. 4, pp.26-40. Feughelman, M. (1997) Mechanical Properties and Structure of Alpha-keratin Fibers: Wool, Human Hair and Related fibres. Sydney: University of New South Wales Press. Laryea, J. (2001) ISP Hair Care Applications Test Method: Test Protocol for Subjective Evaluation of Conditioning Shampoo and Hair Conditioner/Rinse. International Speciality Products. McMullen, R. & Jachowicz, J. (1998) Thermal degradation of hair. I. Effect of curling irons. J. Cosmet. Sci., Vol. 49, pp.223-244. Milczarek, P., Zielinski, M. & Garcia, M. L. (1992) The mechanism and stability of thermal transitions in hair keratin Colloid and Polymer Science. Vol. 270, 1106-1113. Robbins, C. R., and Kelly, C. (1969). Amino acid analysis of cosmetically altered hair, Journal of Cosmetic Science. Vol 2, pp 555-564. Robbins, C. R. (2012) Chemical and Physical Behaviour of Human Hair 5th ed. Berlin Heidelberg: Springer-Velag.pp.540-543. Ruetsch, S.B. and Kamath, Y.K., (2005), Penetration of cationic conditioning compounds into hair fibers: A TOF-SIMS approach, Journal of Cosmetic Science. Vol 56, pp.323-360. Sinclair, R., and Flagler, M.J., and Jones, L., Rufaut, N., and Davis, M.G. (2012). The proteomic profile of hair damage, British Journal of Dermatology. Vol. 166 (Suppl.2), pp.27-32. Swift, J. A. (1997). Cosmetic Science Monographs, Fundamentals of human hair science, Weymouth: Micelle Press pp.11-15. Teglia, A. and Secchi, G. (1999) Proteins in Cosmetics In: Goddard, E. D., & Gruber J, V. (ed.) Principles of Polymer Science and Technology in Cosmetics and Personal Care New York: Marcel Dekker. pp. 433-453. Zhou, Y., Rigoletto, R.,Koemel, D., Zhang, G., Gillece, T.W., Foltris, D.J., Moore, X. QU, Sun, C. (2011). The effect of various cosmetic treatments on protecting hair from thermal damage by hot flat ironing, Journal of Cosmetic Science. Vol. 62, pp.265-282. 11