Contents lists available atScienceDirect
Journal of Photochemistry & Photobiology, B: Biology
journal homepage:www.elsevier.com/locate/jphotobiol
Cystine-thiamin-containing hair-growth formulation modulates the response to UV radiation in an in vitro model for growth-limiting conditions of human keratinocytes
Thomas Hengl
a, Jörn Herfert
a, Alexander Soliman
a, Kim Schlinzig
a, Ralph M. Trüeb
b, Harry F. Abts
a,⁎aMerz Pharmaceuticals GmbH, Alfred-Wegener-Straße 2, 60438 Frankfurt, Germany
bCenter for Dermatology and Hair Diseases, Bahnhofplatz 1 A, CH-8304 Wallisellen, and University of Zurich, Switzerland
A R T I C L E I N F O Keywords:
Proliferation Hair loss Hair growth Keratinocyte UV radiation Telogen effluvium
A B S T R A C T
Background:Ultraviolet radiation (UVR) is known to be harmful to normal human epidermal keratinocytes (NHEKs) of the epidermal skin layer, as well as to hair-follicle-associated keratinocytes. An oral formulation containingL-cystine, thiamin, calciumD-pantothenate, medicinal yeast, keratin and p-aminobenzoic acid (Panto [vi]gar®) has demonstrated clinical efficacy for the treatment of diffuse telogen effluvium; however, its mode of action at the cellular level, and in particular whether protective mechanisms are involved, has yet to be eluci- dated.
Objectives:To assess the capacity of ingredients of this oral formulation, both separately and in combination, to modulate the effects of UVR in growth-limited NHEKsin vitro.
Methods:NHEKs were incubated in keratinocyte basal medium, keratinocyte basal medium lacking cystine, thiamin, calciumD-pantothenate, folic acid and biotine (minimal growth medium [MGM]) or MGM plus test compound. Test compounds comprised the following four ingredients related to the oral formulation:L-cystine, thiamin, calciumD-pantothenate and folic acid (a proposed metabolite of p-aminobenzoic acid), and a combi- nation of these (Panto[vi]gar®-in vitrocorrelate; P-IC). The effect of different doses of these compounds on the metabolic activity and proliferation of NHEKs was tested, as well as their influence on the impact of UV light on NHEKs assessed by monitoring metabolic activity, cell number and apoptosis induction.
Results:Compared with basal medium, MGM reduced the proliferation of NHEKs in a time-dependent manner.
Reduced proliferation is a characteristic of the multifactorial and complex phenotype associated with diffuse hair loss.L-cystine (50 μM) increased metabolic activity and proliferation 3-foldversusMGM (p < 0.05). Thiamin also had a significant effect (p < 0.05) on proliferation and metabolic activity of NHEKs, but calciumD-pan- tothenate and folic acid did not when tested individually in thisin vitromodel. In the presence of P-IC, metabolic activity increased 4-fold and proliferation 3-fold compared with MGM alone (p < 0.05 for both).
Following UV irradiation, cells in MGM showed a 72% reduction in metabolic activity, while P-IC-treated cells showed only a 12–18% reduction. The observed prevention of the UV-induced reduction in metabolic activity was not simply due to filtering UVR by the P-IC components, as P-IC-mediated reduction of this effect persisted even when P-IC was washed out during UV irradiation.
Conclusion:This study demonstrated thatL-cystine and thiamin are essential for proliferation of epidermal keratinocytes and suggests a novel, UV-protective potential of formulations combiningL-cystine and thiamin in growth-limited inter-follicular NHEKsin vitro.
https://doi.org/10.1016/j.jphotobiol.2018.09.005
Received 14 March 2018; Received in revised form 25 July 2018; Accepted 8 September 2018
Abbreviations:MGM, Minimal growth media; NHEKs, Normal human epidermal keratinocytes; PABA, p-aminobenzoic acid; P-IC, Panto(vi)gar®-in vitrocorrelate;
UV-A, Ultraviolet A; UV-B, Ultraviolet B; UVR, Ultraviolet radiation.
⁎Corresponding author.
E-mail addresses:[email protected](T. Hengl),[email protected](J. Herfert),[email protected](A. Soliman), [email protected](K. Schlinzig),[email protected](R.M. Trüeb),[email protected](H.F. Abts).
Available online 17 September 2018
1011-1344/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).
T
1. Introduction
The hair and scalp, like the rest of the skin, are exposed to a variety of noxious environmental factors, including ultraviolet radiation (UVR).
While the consequences of sustained UVR on unprotected skin are well appreciated – mainly aging and photocarcinogenesis – the effects of UVR on the hair are less understood. However, clinical and morpho- logic observations, as well as theoretical considerations, suggest that UVR has some negative effects [1].
Camacho et al. originally reported a specific type of telogen ef- fluvium, characterized by frontovertical hair shedding and an increase of telogen hairs by trichogram, that occurs 3–4 months after sunburn of the scalp in those with hairstyles that left areas of scalp uncovered during prolonged periods of sun exposure [2]. The pathogenesis of this type of telogen effluvium is not clear; however, it has been shown that UV-B light negatively affects the viability of upper hair follicles [3], perhaps directly by the energy delivered to absorbing structures within cells (e.g. DNA) or indirectly via oxygen radicals and activation of cell- surface growth factor and cytokine receptors [4–6]. In addition, it has been proposed that the columns of cells in the hair shaft act as an ef- ficient ‘fiber-optic’-type system, transmitting UV light down into the hair follicle. Morphologically, the keratinocytes within the hair shaft are arranged in compressed linear columns that resemble the coaxial bundles of commercial fiber-optic strands. Thus, hair follicular mela- nocytes located in the region of the hair matrix may function as UV biosensors and respond to photic inputs [7]. In a study of organ-cul- tured human anagen hair folliclesin vitro, the effects of UV irradiation at the molecular level were reported to be a differentially modified hair growth cycle and promotion of cell death, among other events [8]. In summary, it can be concluded that depending on the quantity of UVR exposure, it is conceivable that photodamage may also occur in the hair follicle, and thus contribute to telogen effluvium.
Telogen effluvium is by far the most common cause of hair loss and results from increased shedding of hairs during the telogen phase of the hair cycle. An increase of > 20% in the percentage of follicles in telogen leads to a pathologic increase in the shedding of club hairs. While a number of attempts have been made to determine the underlying pa- thologic dynamics of telogen effluvium and its classification, Headington's classification of the functional types of telogen effluvium remains the most comprehensive [9]: ‘telogen effluvium may result either from synchronization phenomena of hair cycling, or from a de- crease of the anagen phase duration’. Follicles that would normally complete a longer cycle by remaining in anagen enter telogen prema- turely and in a synchronized manner.
Normal cyclic hair growth activity occurs in a random mosaic pat- tern, with each follicle possessing its own individual control mechanism over the evolution and triggering of the successive phases. However, systemic factors, as well as external factors linked to the environment, have an influence (e.g. hormones, cytokines and growth factors, toxins, and deficiencies of nutrients, vitamins and energy [calories]). It is a major characteristic of anagen that the hair shaft is not only growing but that most epithelial hair follicle compartments undergo prolifera- tion, with the hair-matrix keratinocytes showing the highest pro- liferative activity – this is the phase that is most vulnerable to noxious environmental influences.
As previously mentioned, the hair and the scalp are constantly challenged by solar UVR. Damaging effects of UVR on the extra- cutaneous part of the hair shaftin vivo[10] and at the cellular level for organ-cultured hair follicles [8] have previously been described. The accessibility of different skin depths by UVR is wavelength dependent, with UV-B being absorbed mostly in the epidermal layer while UV-A penetrates more deeply, also reaching the dermal layer [11]; thus, normal human epidermal keratinocytes (NHEKs) are the primary UVR target. In vitro, two-dimensional cultures of NHEKs have been estab- lished as an important tool for investigating general aspects of skin physiology. Proliferation and high metabolic activity of hair-matrix
keratinocytes is associated with functional, healthy anagen hair folli- cles. NHEKs and hair-follicle-associated keratinocytes can be function- ally differentiatedin vivo. Nevertheless, scatter plot analysis of DNA- microarray data revealed a good correlation of the gene expression profile of these two cell types, with only a limited number of genes differentially expressed between them [12]. Moreover, with regard to more basic functions like proliferation and general control of metabolic activity, keratinocytes from the hair follicle and epidermal skin layer can be expected to behave in a similar way.
Measuring the metabolic activity and proliferation of NHEKs is therefore considered to be a preliminary and simple (but robust) sur- rogate model and read-out for hair follicle functionality, and useful indirectly as a simplified indicator for potential enhanced hair growth.
In clinical practice, a combination ofL-cystine, thiamin, calciumD- pantothenate, yeast, p-aminobenzoic acid (PABA) and keratin has been successfully used to treat diffuse hair loss [13–15]. The rationale for the use of L-cystine is based on the biochemistry of cystine metabolism, clinical observations of disorders of cystine metabolism, and cystine deficiency. As a group, B complex vitamins (including thiamin [B1] and calciumD-pantothenate [B5]) are important for metabolic functions as they enable utilization of other nutrients, such as carbohydrates and amino acids. Yeast is a rich natural source of amino acids and B com- plex vitamins.
To mimic the reduced activity of keratinocytes associated with di- minished hair growth, we established anin vitrogrowth-limiting system using inter-follicular NHEKs for screening the potential UV-protective capacity of the hair-growth-promoting formulation. Based on expected systemic bioavailability, we established an in vitro correlate of the clinical hair growth formulation (Panto[vi]gar®-in vitrocorrelate; P-IC) consisting ofL-cystine, thiamin, folic acid (an assumed metabolite of PABA) and calciumD-pantothenate forin vitroanalysis.
2. Materials and Methods 2.1. Cell Culture and Treatment
NHEKs (from the ear of a healthy female, age 23 years; PromoCell Lot: 0102008.1) were propagated in ‘growth medium’ (serum-free keratinocyte growth medium 2; C-20011, PromoCell, Heidelberg, Germany) at 37 °C in the presence of 5% CO2in a humidified atmo- sphere. Cells were seeded into flat-bottom 48/96-well microplates (Nunc™) at a density of 1–3 × 104cells. After overnight incubation, cells were switched to ‘basal medium’ (keratinocyte basal medium 2; C- 20211, PromoCell, Heidelberg, Germany), ‘minimal growth medium’
(MGM; keratinocyte basal medium 2; C-20211, PromoCell, Heidelberg, Germany, lacking cystine, thiamin, calciumD-pantothenate, folic acid and biotine, and made to order by PromoCell), or MGM plus test compounds as indicated.
The effects of four components of the oral formulation (L-cystine, thiamin, calciumD-pantothenate and folic acid – as a presumed meta- bolite of PABA) on the metabolic activity and proliferation of NHEKs were tested in a dose-dependent manner as individual compounds and as a combination (P-IC). The four test ingredients were prepared as stock solutions in dimethyl sulfoxide, with the exception ofL-cystine, which was prepared in MGM. These stock solutions were further diluted in MGM to obtain the required test compound concentration. The highest concentrations tested (10 x P-IC) were 1 mM thiamin, 0.5 mML- cystine, and 0.01 mM calciumD-pantothenate and folic acid. All ex- periments were carried out with a maximal solvent concentration of 1%
(v/v). The final concentration of P-IC components in each assay was 100 μM thiamin, 50 μML-cystine, and 1 μM calciumD-pantothenate and folic acid unless otherwise indicated in the text or figure legend (e.g. P- IC diluted 10 times in MGM [P-IC/10] resulted in a concentration of 10 μM thiamin, 5 μML-cystine, and 0.1 μM calciumD-pantothenate and folic acid).
Unused stock solutions were stored at −20 °C. The negative control
wells consisted of the same media as the test wells, but with the ap- propriate vehicle added. Compound interference controls and back- ground controls were also included.
2.2. Metabolic Activity (Resazurin), Proliferation (CyQUANT®Direct/
BrdU) and Apoptosis (Caspase-Glo®3/7) Assays
Metabolic activity, proliferation and apoptosis were determined 24–72 h after the cells had been incubated with their respective test compounds. The amount of metabolic activity was measured using re- sazurin (Sigma-Aldrich) – a non-toxic, water-soluble, redox-sensitive dye. Following 24–72 h after seeding, cells were incubated with media containing 1:11 diluted resazurin (stock solution 0.1 mg/mL) for 90 min at 37 °C. Metabolically active cells converted the blue (non-fluorescent) resazurin to the pink (highly-fluorescent) resorufin, which was subse- quently measured at 560 nm excitation and 590 nm emission using a fluorescence plate reader (Synergy™ H4, BioTek).
The CyQUANT®assay (Invitrogen™) and the BrdU chemilumines- cence kit (Roche Life Science) were used to measure cell proliferation (480 nm excitation/535 nm emission) following the manufacturer's in- structions. Cell apoptosis was measured using the Caspase-Glo® 3/7 assay (Promega) as per the manufacturer's instructions.
2.3. UV Irradiation of NHEKs
For measurements of metabolic activity and apoptosis after UV ir- radiation, cells were cultured and treated as described inSection 2.1 andSection 2.2.
To determine the metabolic activity after UV irradiation, after 24 h of incubation under indicated conditions, cells were irradiated using a solar simulator (Atlas, Suntest®CPS, using wavelengths of 290–800 nm to represent the solar daylight spectrum). Control cells were kept in the solar simulator, but covered with UV-impermeable plates to avoid ir- radiation. The lids of all sample plates were removed and plates were covered with a UV-permeable foil. The distance from the irradiation source to test plates was 20 cm and each well contained 200 μL medium. UV dosing was checked and confirmed by use of a calibrated spectral radiometer (UV-Pad, OpSyTec, Ettlingen, Germany). As de- scribed inSection 2.2, metabolic activity was determined 24 h after UV irradiation by recording the fluorescence intensity of metabolized re- sazurin. Apoptosis was analyzed using the Caspase-Glo®3/7 assay. In order to identify the optimal time point for analysis after irradiation for further experiments, a UV-B dose of 300 mJ was used and apoptosis, as measured by caspase 3/7 activity, was found to be maximal at 14 h post-UV irradiation; therefore, this time point was selected for sub- sequent experiments. The UV-B dose of 200 mJ was selected for further experiments to induce significant UV-stress without inducing high amounts of necrotic cell death, as occurred with 300 mJ. UV-filtering effects of culture medium alone (e.g. by phenol red) were minor (see absorbance spectra) and considered to be negligible. Moreover since phenol-red-containing medium was used for all experiments (e.g. irra- diation with and without P-IC) the potential impact of media filtering was the same for all conditions. To further exclude any filtering effects of medium or compounds, experiments were performed where the culture medium, containing treatments where indicated, was changed to MGM only before irradiation and switched back, after irradiation, to the same treatment-containing medium as has been used prior to irra- diation (PI-C or 10× PI-C). After the medium had been thus replaced, cells were cultured for an additional 14 or 24 h prior to the apoptosis and metabolic assays, respectively.
2.4. Data Analysis of Metabolic Activity, Proliferation and Apoptosis Mean values of relative fluorescence and luminescence units (RFU and RLU, respectively) and standard deviation (SD) of replicates were calculated using data analysis software (Gen5™; BioTek). Statistical
significance was assessed by comparing mean ( ± SD) values using a one-way ANOVA (SigmaPlot™ V13, Systat Software Inc.) for in- dependent groups (Figs. 1B,2 and 3). For Figs. 4 and 5, data were analyzed using a two-way ANOVA (SigmaPlot™ V13, Systat Software Inc.). If results of the ANOVA were significant, a Holm-Sidakpost hoc test was conducted. Statistical significance for both the one-way and two-way ANOVA was set at p < 0.05. Data from a single, re- presentative experiment of at least 3 independent experiments per- formed are shown and are based on 4 replicates in each experiment except forFig. 1where there were 12 replicates and the proliferation assays which included 6 replicates.
3. Results
3.1. Establishment of ‘Minimal Growth’ Assay System
Conventional cell culture medium provides an optimal growth en- vironment resulting in maximal cell proliferation. To mimic the re- duced activity of keratinocytes believed to be associated with diffuse hair loss, we established a growth-limitingin vitrosystem using NHEKs cultivated in different growth-impacting media. A cell-number-depen- dent increase in metabolic activity with a linear relationship could be detected in keratinocytes cultivated either in MGM, basal medium, or growth medium (Fig. 1A). Time-dependent analysis of metabolic ac- tivity (Fig. 1B) demonstrated only a small increase in the metabolic activity (a surrogate indicator of cell number) of cells cultivated in MGM, whereas keratinocytes cultivated in basal and growth medium showed nearly a 2-fold increase after 48 h and up to a 4–5-fold increase after 72 h. Taken together, compared with NHEKs grown in basal or Fig. 1.Metabolic activity of NHEKs cultivated in three different media: MGM, basal medium and growth medium. (A) Cell-number-dependent metabolic ac- tivity of NHEKs cultivated for 24 h in MGM, basal medium and growth medium;
(B) Metabolic activity of NHEKs cultivated in MGM, basal medium and growth medium over time. Data were normalized to MGM 24-h value. A one-way ANOVA following Holm-Sidak Test versus MGM data was performed;
*p < 0.05 for the comparisons shown.
Fig. 2.Impact of ingredients of the oral formulation, both separately and as the P-IC combination, on (A) metabolic activity and (B) proliferation of NHEKs cultivated for 72 h in MGM, MGM +L-cystine (Cys), thiamin (Thi), calciumD-pantothenate (CaP) or folic acid (FA) and a combination of these constituents (MGM + P-IC) at different concentrations (undiluted, 1/10 dilution and 1/5 dilution). P-IC consists of 50 μML-cystine, 100 μM thiamin, 1 μM calciumD-pantothenate, 1 μM folic acid.
(C) Impact of P-IC andL-cystine on DNA synthesis measured via BrdU incorporation. A one-way ANOVA following Holm-Sidak Test with pairwise comparison was performed; *p < 0.05 for the comparisons shown.
growth media, the use of MGM reduced the proliferation of NHEKs significantly at 48 and 72 h (p < 0.05), representing anin vitromodel for NHEKs under growth-limiting conditions. The metabolic activity of the cellsper se did not appear to be affected during the investigated cultivation time since the same cell number results in similar metabolic activity values, independent of the culture conditions (Fig. 1A).
3.2. Impact of P-IC on Metabolic Activity and Proliferation
The effect of the test compounds on metabolic activity and pro- liferation was analyzed either directly (using MGM supplemented with thiamin, calcium D-pantothenate,L-cystine) or indirectly (using folic acid as a presumed metabolite of PABA) in the establishedin vitrotest system. Two components (L-cystine and thiamin) showed a clear posi- tive impact on metabolic activity and proliferation of NHEKs in a dose- dependent manner compared with MGM alone (Figs. 2A and B). Of the individual components tested,L-cystine was observed to have the lar- gest effect on metabolic activity (50 μML-cystine: 3-fold increase versus MGM) and proliferation (50 μML-cystine: 3-fold increase versus MGM);
however, thiamin also had a notable effect on the metabolic activity of NHEKs, while calciumD-pantothenate and folic acid were demonstrated to have no significant impact (Figs. 2A and B).
Further to the effects of the individual components, a significant effect on metabolic activity, proliferation and DNA synthesis was de- monstrated for the P-IC combination (Figs. 2A, B and C; p < 0.05) showing that the combination of these four components overcame the negative impact of the MGM on proliferation. In the presence of P-IC, there was a dose-dependent, 4-fold increase in metabolic activity and a 3-fold increase in proliferation compared with MGM, at maximum
concentration (Figs. 2A and B). Moreover, the effect of P-IC was su- perior to that ofL-cystine (50 μM) alone.
DNA synthesis was found to be increased 100-fold in P-IC-treated cells compared with MGM-cultivated cells (Fig. 2C). Addition ofL-cy- stine, especially at concentrations of 5 μM or greater (i.e. in 1/10 P-IC or P-IC), with other P-IC components appeared to facilitate the in- creased DNA synthesis capacity, with P-IC resulting in 40% more DNA synthesis than the individualL-cystine dose equivalent (50 μM).
Fig. 3.UVR under limited growth conditions. (A) Data showing the metabolic activity of NHEKs in MGM, measured 24 h post-UVR dose as indicated (UV-B in mJ); (B) Data showing the apoptosis of NHEKs over time following UVR com- pared with control, non-irradiated cells, measured by caspase 3/7 activity. A one-way ANOVA following Holm-Sidak Test with pairwise comparison was performed; *p < 0.05 for the comparisons shown.
Fig. 4.The effects of P-IC on UVR (200 mJ UV-B) of keratinocytes. (A and B) Microscopic analysis of the impact of UVR on cell density and morphology of NHEKs cultivated in MGM; (C and D) Microscopic analysis of the impact of UVR on cell density and morphology of NHEKs cultivated in MGM + P-IC; (E) Apoptosis of NHEKs post-UVR; (F) Metabolic activity of NHEKs post-UVR.
NHEKs were cultivated in MGM, MGM + P-IC and MGM + 10 x P-IC. Apoptosis of NHEKs was detected 14 h post-UVR. Data were normalized to control values.
Metabolic activity was analyzed 24 h post-UVR. Data were normalized to con- trol values. A two-way ANOVA following Holm-Sidak Test with pairwise comparison was performed; *p < 0.05 for the comparisons shown.
3.3. Effect of P-IC on the Response of Human Keratinocytes to UV Radiation
Pilot experiments were conducted to find suitable doses and time points for subsequent studies. In MGM, NHEKs showed a UVR dose- dependent decrease in metabolic activity 24 h after irradiation (Fig. 3A). Analysis of induction of apoptosis revealed a maximal in- duction of caspase 3/7 activity 14 h after irradiation (Fig. 3B). Since irradiation with 300 mJ resulted in almost maximal toxicity, for sub- sequent investigations of the impact of UVR on metabolic activity and apoptosis, a UVR dose with a UV-B portion of 200 mJ was chosen.
Microscopic analysis and caspase 3/7 activity revealed that P-IC and 10 x P-IC significantly (p < 0.05) reduced the amount of UV-induced apoptosis of NHEKs cultured in MGM alone (Figs. 4A–E).
The negative impact of UV irradiation on metabolic activity was attenuated by the addition of P-IC to MGM, as demonstrated by a
reduction of only 12–18% in metabolic activity observed for non-irra- diated, P-IC-treated NHEKs compared with the 72% reduction observed in UV-irradiated cells versus non-irradiated cells cultured in MGM alone (Fig. 4F).
When P-IC was compared with the UV-filtering reference com- pound, PABA (Fig. 5A), the prevention of the effects of UVR with P-IC was greater than that with PABA, as demonstrated by the greater me- tabolic activity and lower degree of apoptosis observed in irradiated cells treated with P-IC compared with those treated with PABA only (Figs. 5B and C). Treatment with P-IC resulted in only 30% of the level of apoptosis observed in the presence of PABA (Fig. 5C).
In order to distinguish between a simple filtering effect (requiring the presence of P-IC during irradiation) and a possible effect of P-IC on cell metabolism being responsible for these observations, the effect of treatment with P-IC on metabolic activity and apoptosis was in- vestigated further. InFigs. 5D and E, NHEKs were incubated with P-IC Fig. 5.Effects of P-IC on UVR of keratinocytes occur even when P-IC is not present during the irradiation. (A) UV absorption spectra of MGM, MGM + P-IC and MGM + UV filtering compound (100 μM PABA); (B and C) Metabolic activity and apoptosis of NHEKs after UVR (200 mJ UV-B). In the first set of experiments (B and C), NHEKs were cultivated in MGM, MGM + P-IC, MGM + 10 x P-IC, and 100 μM PABA continuously before, during and after UV irradiation. In the second set of experiments (D and E), NHEKs were cultivated in MGM, MGM + P-IC and MGM + 10 x P-IC before and after UV irradiation, but treatments were excluded (exc) during the actual period of UV irradiation when cells were incubated in MGM alone to wash out the P-IC. A two-way ANOVA following Holm-Sidak Test with pairwise comparison was performed; *p < 0.05 for the comparisons shown.
before and after UV irradiation; however, the treatments were washed out prior to UV irradiation and replaced with MGM alone so that P-IC was excluded for the actual period of UV irradiation. As shown in Figs. 5D and E, P-IC continued to prevent the UVR-induced reduction in metabolic activity and UVR-induced increase in apoptosis, even when washed out during UV irradiation.
4. Discussion
The aim of this study was to investigate the impact of compounds frequently provided by hair-growth formulations, separately and in combination, on cellular UV response using a growth-limited NHEKsin vitromodel.
First, anin vitrogrowth-limiting system lacking the compoundsL- cystine, thiamin, calcium D-pantothenate, folic acid and biotine was established to emulate maximal nutrient deprivation. Under these nu- trition-deprived conditions, normal human keratinocytes showed a significant reduction in proliferation, mimicking the reduced activity of NHEKs believed to be associated with diminished hair growth.
Furthermore, this system allowed us to investigate the effect of the combination of these compounds on the response to UVR. Our data demonstrated that P-IC delivers crucial components (thiamin, calcium
D-pantothenate,L-cystine and folic acid) which ensure normal metabolic activity, DNA synthesis and cell proliferation of epidermal keratino- cytes. Elimination of these compounds, by cultivation in MGM, resulted in a significant reduction of cellular activity within 72 h.
In general keratinocytes can be identified as the main cell type within the hair follicle and, thus, normal human epidermal-derived keratinocytes were selected for thein vitrosystem studied here though they resemble inter-follicular keratinocytes more than hair follicle cells.
While clear functional differences can be observed between hair-fol- licle-derived and epidermis-derived keratinocytes, gene-expression analysis has found only limited differences between kerationcytes from the epidermis and hair-follicle when cultivatedin vitro[12,16,17]. In addition it is believed that findings concerning general cell physiology such as the impact on metabolic activity and cell proliferation in epi- dermal keratinocytes will also be applicable to follicular keratinocytes.
Our data confirm and extend previous findings [18] on the re- levance ofL-cystine for metabolic activity and proliferation of human keratinocytes. However, to our knowledge this is the first study to show an impact of thiamin on the proliferation of keratinocytes.
Importantly, the UV irradiation analysis indicated that P-IC may be protective against the negative impact of solar UVR. In daily life, the hair and scalp are challenged constantly by solar UVR. UV-B causes direct damage to DNA, negatively affecting the viability of cells [3] and although the penetration of UV-B into the skin is less than that of UV-A [19], hairs may act fiber-optically to transmit UV-B to deeper skin structures. Here, we demonstrate a UVR (290–800 nm) dose-dependent decrease of cell number/metabolic activity in keratinocytes cultivated in MGM and an associated dose-dependent increase in apoptosis. The reduction in apoptosis and the prevention of decreased metabolic ac- tivity by P-IC was substantially greater than that of PABA, a simple UV- absorber, and occurred even if P-IC was not present during the actual period of UV irradiation. This is suggestive of a protective effect of P-IC against the negative effects of solar radiation on keratinocytes but also suggests that the effect of P-IC goes beyond a simple absorption of UV-B or direct quenching activity of the compounds. In addition the bene- ficial effects on cellular metabolism, enabling the cells to resist the damaging effects of UV-B, may be mediated via metabolic support, as demonstrated by the lack of requirement for P-IC to be present at the time of UV irradiation.
The ability of P-IC to reduce apoptosis raises the possibility that either the primary damage by UV (e.g. DNA damage) is reduced (our current hypothesis) or that the ability of the cell to repair this damage is improved. Further data, for example quantification of the extent of DNA-damage, would be needed to elucidate the mode of action of P-IC
in more detail.
Previous studies have established UV-protective benefits of other cystine derivatives [20]. Notably, thiol compounds can act as direct scavengers of radicals and UV-induced reactive oxygen species [21].
Moreover, oral administration ofL-cystine and vitamin B6 prevented hair loss in C57BL/6 mice exposed to cigarette smoke, another en- vironmental factor that is both genotoxic and generates oxidative stress [22,23]. The effect may be related to the glutathione-related detox- ification system and is of particular relevance to humans, since a po- pulation-based cross-sectional survey of men aged 40 years or older showed statistically significant positive associations between moderate or severe androgenetic alopecia and smoking status, current cigarette smoking of 20 cigarettes or more per day, and smoking intensity [24].
In addition, Bahta et al. cultured dermal hair papilla cells (DPCs) from both balding and non-balding human scalps and demonstrated that balding DPCs grew slowerin vitrothan non-balding DPCs. Loss of proliferative capacity of balding DPCs was associated with changes in cell morphology, expression of senescence-associated beta-galactosi- dase and markers of oxidative stress and DNA damage as well as de- creased expression of proliferating cell nuclear antigen and Bmi-1 [25].
This suggests that balding DPCs are particularly sensitive to environ- mental stressors such as cigarette smoke [26] and UVR [1]. These findings, along with the existing unmet need in the treatment of an- drogenetic alopecia beyond minoxidil and finasteride, suggest further pathogenic pathways contributing to hair loss may be relevant and represent opportunities for further therapeutic strategies.
The data shown here for P-IC provide evidence for a potential novel UVR-protective role for the oral formulation which containsL-cystine and thiamin, especially in NHEKs under growth-limiting conditions.
The prevention of environmental stress-mediated cell death would have a dual impact on hair growth by: (1) protecting hair-matrix keratino- cytes of existing hair follicles; and (2) protecting hair follicle stem cells in active hair follicles. Thus, not only might the current hair follicle be protected, but also delayed hair loss, as a consequence of environmental stress by impairment of the hair-follicle stem cells in the bulge region, may be prevented.
Although chronic telogen effluvium represents a benign medical condition, affected individuals experience significant psychoemotional stress, with a reduction in quality of life and presence of secondary morbidities [27] and effective treatments are needed. The quantity and quality of hair are closely related to the nutritional state of an in- dividual [28–30]. Since an important commercial interest lies in the nutritional value of vitamin and amino acid supplements, a central question is whether supplementing an already adequate diet with spe- cific amino acids and vitamins may further promote hair growth in individuals where a specific nutritional deficiency is not a critical contributory factor to hair loss.
However, it is possible that the large number of internal and ex- ternal factors (including UVR) that influence hair health, do so to such a degree that nutritional therapy can boost hair that is negatively affected by these factors e.g. UVR and supply, uptake and transport of proteins and other nutrients.
Follow-up studies using hair-follicle-derived keratinocytes would be helpful to confirm and extend these preliminary data. Since the hair- follicle-specific phenotype also depends on organotypic cell-cell inter- actions, in addition to 2D-cultivation, 3D-cultivation of hair-follicle cells would provide a valuable additionalin vitromodel. Another area for future study might be to explore thisin vivosince UV-B radiation potentially penetrates only the upper epidermis. It would also be in- teresting to determine the impact of indirect UV mediators, such as oxygen radicals on the possible protective role of Panto(vi)gar®.
The effect of P-IC on metabolic activity and proliferation was su- perior to that ofL-cystine alone. However, the present data do not allow determination of whether the effects ofL-cystine and thiamin together are additive or synergistic; this area requires further exploration.
In conclusion, P-IC delivers crucial compounds required for
physiologic activities, such as metabolism and proliferation of kerati- nocytes. Our data further suggest a novel role for P-IC in protection against UVR, which may provide insights on the mechanism of action for the protective effects hair-growth promoting formulations, like Panto(vi)gar®which has proven efficacy for the treatment of diffuse hair loss [13,14].
Disclosures
Thomas Hengl is an employee of Merz Pharmaceuticals GmbH.
Jörn Herfert has been working temporarily for Merz Pharmaceuticals GmbH (student internship).
Alexander Soliman has been working temporarily for Merz Pharmaceuticals GmbH (student internship).
Kim Schlinzig is an employee of Merz Pharmaceuticals GmbH.
Ralph Trüeb performs consultant activities for Apomedica Switzerland, Lexington International LLC and Merz Pharmaceuticals GmbH.
Harry Abts is an employee of Merz Pharmaceuticals GmbH.
Acknowledgments
Editorial assistance was provided by Ogilvy Healthworld, Oxford, UK and funded by Merz Pharmaceuticals GmbH.
References
[1] R.M. Trueb, Is androgenetic alopecia a photoaggravated dermatosis? Dermatology 207 (4) (2003) 343–348.
[2] F. Camacho, J.C. Moreno, M.J. Garcia-Hernandez, Telogen alopecia from UV rays, Arch. Dermatol. 132 (11) (1996) 1398–1399.
[3] S. Muller-Rover, et al., Overexpression of Bcl-2 protects from ultraviolet B-induced apoptosis but promotes hair follicle regression and chemotherapy-induced alopecia, Am. J. Pathol. 156 (4) (2000) 1395–1405.
[4] J. Cadet, E. Sage, T. Douki, Ultraviolet radiation-mediated damage to cellular DNA, Mutat. Res. 571 (1–2) (2005) 3–17.
[5] S. Pillai, C. Oresajo, J. Hayward, Ultraviolet radiation and skin aging: roles of re- active oxygen species, inflammation and protease activation, and strategies for prevention of inflammation-induced matrix degradation - a review, Int. J. Cosmet.
Sci. 27 (1) (2005) 17–34.
[6] L. Rittie, G.J. Fisher, UV-light-induced signal cascades and skin aging, Ageing Res.
Rev. 1 (4) (2002) 705–720.
[7] B. Iyengar, The hair follicle: a specialised UV receptor in the human skin? Biol.
Signals Recept. 7 (3) (1998) 188–194.
[8] Z. Lu, et al., Profiling the response of human hair follicles to ultraviolet radiation, J.
Invest. Dermatol. 129 (7) (2009) 1790–1804.
[9] J.T. Headington, Telogen and effluvium. New concepts and review, Arch. Dermatol.
129 (3) (1993) 356–363.
[10] A.C. Santos Nogueira, I. Joekes, Hair color changes and protein damage caused by ultraviolet radiation, J. Photochem. Photobiol. B 74 (2–3) (2004) 109–117.
[11] J. Krutmann, B.A. Gilchrest, Photoaging of Skin, in: J. Krutmann, B.A. Gilchrest (Eds.), Skin Aging, Springer, Germany, 2006, pp. 33–44.
[12] M. Nakano, et al., Establishment of a new three-dimensional human epidermal model reconstructed from plucked hair follicle-derived keratinocytes, Exp.
Dermatol. 25 (11) (2016) 903–906.
[13] H. Petri, P. Pierchalla, H. Tronnier, The efficacy of drug therapy in structural lesions of the hair and in diffuse effluvium–comparative double blind study, Schweiz Rundsch. Med. Prax. 79 (47) (1990) 1457–1462.
[14] J. Budde, et al., Systemic therapy of diffuse effluvium and hair structure damage, Hautarzt 44 (6) (1993) 380–384.
[15] N. Lengg, B. Heidecker, B. Seifert, R.M. Trueb, Dietary supplement increases anagen hair rate in women with telogen effluvium: results of a double-blind, placebo- controlled trial, Therapy 4 (1) (2007) 59–65.
[16] G.E. Rogers, Hair follicle differentiation and regulation, Int. J. Dev. Biol. 48 (2–3) (2004) 163–170.
[17] M.R. Schneider, R. Schmidt-Ullrich, R. Paus, The hair follicle as a dynamic mini- organ, Curr. Biol. 19 (3) (2009) R132–R142.
[18] D.H. Obrigkeit, et al., Xenobiotics in vitro: the influence ofL-cystine, pantothenat, and miliacin on metabolic and proliferative capacity of keratinocytes, Cutan. Ocul.
Toxicol. 25 (1) (2006) 13–22.
[19] A. Tewari, et al., UVA1 is skin deep: molecular and clinical implications, Photochem. Photobiol. Sci. 12 (1) (2013) 95–103.
[20] D.P. Steenvoorden, G.M. Beijersbergen van Henegouwen, Cysteine derivatives protect against UV-induced reactive intermediates in human keratinocytes: the role of glutathione synthesis, Photochem. Photobiol. 66 (5) (1997) 665–671.
[21] L. van den Broeke, G. Beyersbergen Van Henegouwen, Thiols as potential UV ra- diation protectors: an in vitro study, J. Photochem. Photobiol. B: Biol. 17 (3) (1993) 279–286.
[22] F. D'Agostini, et al., Induction of alopecia in mice exposed to cigarette smoke, Toxicol. Lett. 114 (1–3) (2000) 117–123.
[23] F. D'Agostini, et al., Chemoprevention of smoke-induced alopecia in mice by oral administration ofL-cystine and vitamin B6, J. Dermatol. Sci. 46 (3) (2007) 189–198.
[24] L.H. Su, T.H. Chen, Association of androgenetic alopecia with smoking and its prevalence among Asian men: a community-based survey, Arch. Dermatol. 143 (11) (2007) 1401–1406.
[25] A.W. Bahta, et al., Premature senescence of balding dermal papilla cells in vitro is associated with p16(INK4a) expression, J. Invest. Dermatol. 128 (5) (2008) 1088–1094.
[26] R.M. Trueb, Association between smoking and hair loss: another opportunity for health education against smoking? Dermatology 206 (3) (2003) 189–191.
[27] I.M. Hadshiew, et al., Burden of hair loss: stress and the underestimated psycho- social impact of telogen effluvium and androgenetic alopecia, J. Invest. Dermatol.
123 (3) (2004) 455–457.
[28] A.M. Finner, Nutrition and hair: deficiencies and supplements, Dermatol. Clin. 31 (1) (2013) 167–172.
[29] D.H. Rushton, Nutritional factors and hair loss, Clin. Exp. Dermatol. 27 (5) (2002) 396–404.
[30] C.L. Gummer, Diet and hair loss, Semin. Dermatol. 4 (1985) 35–39.