Enlarged rete pegs with excessive accumulation of melanosomes leading to darker aging spots revealed by histomorphological measurements of internal structures of the epidermis

Various histological studies of facial pigmented spot sites such as solar lentigo have been reported, but few studies have used quantitative indices by histomorphometric analysis of the internal structure of pigmented spot sites using non‐invasive methods. In the present study, to quantitatively elucidate morphological changes in the epidermis in male, darker‐pigmented spots and female, light‐pigmented spots, indices that characterize the internal structure of the epidermis in pigmented spot sites were measured using in vivo confocal laser scanning microscopy (CLSM).


Conclusion:
From the measurements of quantitative parameters, morphological changes in the epidermis were clearly related to the surface colour tone of pigmented spots. Darker pigmented spot sites tended to show longer rete pegs in the epidermis. Accumulation of melanosomes in epidermal basal cells could be considered to increase with the degree of elongation of rete pegs at pigmented spot sites and, thus, induce darker pigmented spots.

INTRODUCTION
Solar lentigo refers to pigmented spots on the skin surface, representing an important aesthetic problem encountered with aging. Melanin causes pigmented spots and is well known to play an important role in protecting the skin from damage caused by ultraviolet (UV) radiation. A typical example is the supranuclear melanin cap found in epidermal basal cells [1]. Melanin is synthesized by the action of tyrosinase in melanosomes, which are organelles within melanocytes. Synthesized melanin-containing melanosomes can be secreted from the dendritic tips of melanocytes for transfer to surrounding epidermal cells [2]. In epidermal basal cells, melanosomes are localized above the nucleus to protect the nuclear DNA from UVinduced damage [3].
The accumulation of melanosomes in these epidermal basal cells is usually transient and reversible, such as suntan. However, when proliferation of epidermal basal cells is abnormally and irreversibly accelerated, melanosomes are thought to accumulate excessively in these increased basal cells, resulting in pigmented spots. Because excessive proliferation of epidermal basal cells would logically seem likely to cause deformation of the structure of the rete pegs, determination of the structure of the epidermal basal layer at pigmented spots is considered important for elucidating the causative mechanisms underlying the darkening of pigmented spots.
On the other hand, senile pigment freckles have already been clinically reported to show the following three types of characteristic morphological changes: (1) hyperpigmentation type, with the main histological finding of hyperpigmentation of the basal layer in the epidermis; (2) lentigo type, with the most characteristic features being small club-shaped projections of epidermal cells into the dermis and many melanocytes and melanin pigments in the basal layer of epidermal club-like projections and (3) pigment incontinentia type, showing small amounts of pigment in the epidermis and many macrophages in the upper part of the dermis [4]. Hodgson reported that senile-pigmented macules undergo a gradual process of maturation, with the protrusion of epidermal cords seen only in completed senile-pigmented macules, and epidermal thickening being a late change [5]. Melanosome complexes within keratinocytes are large and numerous melanosomes that accumulate in solar lentigo [6]. However, few studies have quantified indices from histomorphometric analysis of the internal structure of pigmented spot sites using noninvasive observational analyses.
Since in vivo confocal laser scanning microscopy (CLSM) allows non-invasive observation of the internal structure of the skin, this modality has been widely used for studying structural changes in the epidermis [7][8][9]. In particular, melanin produces stronger backscattering, allowing melanocytes and pigmented keratinocytes to be clearly visualized [10,11].
In the present study, to elucidate morphological changes in the epidermis using quantified indices, internal structures of the epidermis at pigmented spot sites were measured by CLSM. Parameters indicative of the internal structure of the epidermis at pigmented spot sites were calculated using image analysis. Representative pigmented spot sites were selected after measurements of surface colour tone at pigmented spots on the cheek by age group.

Darkness of pigmented spots
Participants in this study were 112 Japanese volunteers (43 men, 69 women) with a mean age of 48.4 years (range, 32-59 years). All volunteers had normal, healthy skin without acne or dermatitis. This study was conducted in accordance with the principles of the Declaration of Helsinki. Facial images from left, right, and frontal views were captured using a VISIA® skin analysis imaging system (Canfield Scientific Inc.) 20 min after washing with cleansing foam in an air-conditioned room, under constant temperature (21 ± 1°C) and relative humidity (50 ± 5%).
All pigmented spots ≥6 mm (long diameter) on the cheek were analysed using image analysis software (ImageJ; Wayne Rasband, National Institutes of Health). RGB values of the pigmented area and four surrounding non-pigmented areas were measured. The L* value was calculated using XYZ values converted from RGB values obtained from facial images via linear RGB values [12][13][14]. pourrait envisager que l'accumulation de mélanosomes dans les cellules basales épidermiques augmente selon le degré d'allongement des crêtes épidermiques au niveau des sites de taches pigmentées, et entraîne ainsi des taches pigmentées plus foncées.
The lightness difference value (ΔL*) between pigmented spots and non-pigmented areas was calculated from the mean L* values of each area.
The ΔL* values were calculated for each pigmented spot and obtained as mean values and standard deviation for each pigmented spot site in males and females per age group. Student's t-test was applied for statistical evaluations.
A conspicuous and relatively large pigmented spot (long diameter, ≥12 mm) was defined as a 'noticeable pigmented spot', and the incidence of these spots was analysed by age group for males and females.

Classification of pigmented spots by facial images
Participants comprised 20 Japanese volunteers (9 men, 11 women) with a mean age of 48.6 years (range, 42-59 years). All volunteers had normal, healthy skin without acne or dermatitis. Facial images were captured using the VISIA® system 20 min after washing with cleansing foam in an airconditioned room, under constant temperature (21 ± 1°C) and relative humidity (50 ± 5%). Informed consent was obtained from each participant prior to image capture.
Typical pigmented spots on the cheek suitable for observation were analysed using ImageJ. RGB values of the pigmented area and four surrounding non-pigmented areas were measured. The L* value was calculated from RGB values.
The ΔL* value between pigmented spots and nonpigmented areas was calculated from the mean L* values of each area. Pigmented spots with a ΔL* value ≥11 were classified as dark-coloured pigmented spots, and spots with a ΔL* value ≤7 were classified as light-coloured pigmented spots. Pigmented spots with a ΔL* value <11 but >7, classified as moderately dark, were excluded from analysis in this study.

In vivo confocal laser scanning microscopy imaging
Thirteen participants (six men and seven women) with a mean age of 48.3 years (range, 42-59 years) were selected from the above 20 subjects based on the surface colour of pigmented spot sites. CLSM imaging of the epidermal basal layer was performed at a pigmented lesion located on the cheek after the appearance was photographed. Informed consent was obtained from each participant prior to CLSM imaging.
CLSM was performed using a VivaScope® 1500 noninvasive skin imaging system (Caliber Imaging and Diagnostics) equipped with a × 30 objective lens with a numerical aperture of 0.9. Ultrasonic couplant gel was used as an immersion fluid for the lens. Mineral oil was used between the skin and ring template to provide good optical contact.
Images were collected from these 13 subjects at pigmented spot sites using two methods: automatically adjusting the laser power (to a maximum of 16 mW) and using fixed power (5 mW). After attaching the adhesive ring to the pigmented spot area for measurement, a colour image was captured using the microscope. The CLSM lens was then fixed to the adhesive ring and a wide range of block images including pigmented spots and surrounding non-pigmented areas were observed. Based on the obtained microscope image and block image, horizontal images (500 μm × 500 μm) from the surface of the stratum corneum (SC) to the dermis (maximum depth, 250-300 μm [15]) were scanned at intervals of 1.52 μm at 5 points within the pigmented spot area and 5 points within non-pigmented areas.

Image analysis
Measurements of skin parameters were modified from the methods previously reported by Robartson and Rees [16]. Parameters of the epidermal structure measured by image analysis based on the CLSM image are shown in Figure 1. The thickness of the SC was measured using the derivative method from the change in brightness of the vertical cross-sectional image, as described below. Stacks of images were imported into ImageJ software. Using ImageJ, mean brightness of each image in a stack was calculated and a profile plot was generated. The first derivative of the data was calculated, and a graph of relative reflection intensity F I G U R E 1 Parameters of the internal structure of the skin. (brightness) versus depth (in micrometers) was plotted. This method was predicated on the fact that because the SC is the most optically dense area within the epidermis, the maximum and minimum derivatives of serial optical densities were considered to adequately define the start and end of the SC. Measurement of SC thickness was repeated at least fives times at different positions in the measurement area. Minimum and maximum thicknesses of the epidermis were measured, in a similar manner to that of the SC, as mean distance from the skin surface to the top or bottom of the dermal papillae, respectively.
The area and width of epidermal basal cells surrounding the dermal papilla containing melanosomes were measured after binarization of the CLSM image (horizontal section) of the intermediate depth of the dermal papillae. The total area of epidermal basal cells was obtained by summing the area of epidermal basal cells surrounding the dermal papilla containing melanosomes for each CLSM image. The mean area of epidermal basal cells was taken as the area of epidermal basal cells per dermal papilla, calculated as the mean epidermal cell area in five surrounding dermal papillae. The mean width of epidermal basal cells surrounding the dermal papilla measured at 5 points per dermal papilla was calculated. Five dermal papillae were selected and measured for each CLSM image, and the mean width was calculated. Only measurements of the area and width of epidermal basal cells surrounding a dermal papilla containing melanosomes used CLSM images obtained at a fixed laser power of 5 mW. All other measurements used CLSM images obtained with automatically adjusted laser power.
Melanosome content was evaluated from the histogram of three-dimensional images constructed from stack images of CLSM after binarization processes as the volume of epidermal cells containing melanosomes. In addition, the density of melanosomes was calculated as the ratio of the volume of epidermal cells containing melanosomes to the total volume from the top of the epidermal-dermal junction to the bottom of the dermal papillae. Area and shape descriptors of dermal papillae were evaluated using the image analysis tools in ImageJ from the CLSM image of the intermediate depth of the dermal papillae.
All parameters were measured using five CLSM images for each pigmented spot and obtained as mean values. Data were measured at 7 points each for dark and lightcoloured pigmented spot sites from males and females, and mean values were calculated. Student's t-test was applied for statistical evaluations.
All volunteers had normal, healthy skin without acne or dermatitis. This study was conducted in accordance with the principles of the Declaration of Helsinki.
Washing and acclimation were performed similarly to the method described above. Skin tone was measured using a CM-700d spectrophotometer (Konica Minolta). The effect of exposure to sunlight on skin tone was quantified using the value (Δ (I-C) L*) obtained by subtracting cheek (exposed area) tone data from inner upper arm (unexposed area) tone data.
Stratum corneum samples were collected with cellophane tape (Nichiban) from the largest solar lentigo, nonpigmented area on the cheek, and inner upper arm.
Soluble protein and insoluble protein contents in the SC extract were quantified using a BCA Protein Assay kit (Takara Bio Inc.) and with o-phthalaldehyde reagent, respectively, and the sum of soluble and insoluble protein contents was designated as the total protein content of the SC.
As a substrate for the measurement of activity, t-butyloxycarbonyl-L-valyl-L-leucyl-L-lysine 4-methylcoumaryl-7-amide (Peptide Institute Inc.) prepared with Tris-HCl buffer (pH 8.0) was added to the SC extract and the mixture was incubated at 37°C [17]. The fluorescence of substrate-derived 7-amino-4-methylcoumarin (MCA) standard solution was measured at 380 nm (excitation wavelength) and 460 nm (emission wavelength) using an EnSight Multimode Plate Reader (PerkinElmer Inc.). MCA production per total protein content in the SC (fmol/ min/μg) was applied for plasmin-like activity.

RESULTS
A comparison of the darkness of pigmented spots on the cheek between male and female subjects confirmed that the difference in mean L* values (ΔL*) between pigmented spot sites and non-pigmented areas was significantly larger in male subjects than in female subjects ( Figure 2a). Figure 2b shows representative photographs of male and female subjects with conspicuous pigmented spots. A comparison of the darkness of pigmented spots in female and male subjects by age group revealed a significant increase in mean ΔL* value for male subjects in their 40s and 50s compared with female subjects (Figure 3). Figure 4 shows the incidence of 'noticeable pigmented spots' (long diameter, ≥12 mm) by age group for male and female subjects. The incidence of 'noticeable pigmented spots' increased among male subjects in their 40s and among female subjects in their 50s.
Based on the above surface image analyses, 13 subjects with typical dark-and light-coloured pigmented spots were selected for observation of the internal structure of the epidermis at pigmented spot sites for histomorphometric analysis. Typical colour microscopic images and block images of each dark pigmented spots on male and female subjects taken using CLSM are shown in Figure 5. The internal structure of the skin differed between pigmented spot areas and surrounding non-pigmented areas. As previously reported, strong melanosome-derived reflexes were observed in the basal layer of the epidermis [18]. No melanosomes were observed to have dropped into the dermis in these images obtained from CLSM analysis. Figure 6 shows typical photographs of pigmented spots and horizontal sections of CLSM images inside pigmented spot sites. The shapes of dermal papillae observed at pigmented spot sites differed between the group of darkcoloured spots in males and other groups. Figure 7 shows horizontal and vertical cross-sections of CLSM images in each pigmented spot. The intricate structure of multiple dermal papillae was observed in typical dark-coloured pigmented spot sites from male subjects. In contrast, regular structures of the dermal papillae were observed in the group of light-coloured pigmented spot sites in males and the groups of pigmented spot sites in females.
Parameters indicating epidermal structure as measured by analysis of CLSM images are presented in Table 1. A significant increase in the maximum thickness of the epidermis from the surface of the SC to the tip of the rete pegs was confirmed in pigmented spot sites of any measurement groups compared with surrounding nonpigmented areas. A significant increase in melanosome content (reflecting the number of epidermal cells containing melanosomes) was detected in pigmented spot sites of all groups compared with non-pigmented areas. In particular, melanosome content in the group of dark-coloured pigmented spots in males was significantly increased to about eight times that of non-pigmented areas, and more than double that of the group of light-coloured pigmented spots in males. Figure 8 shows plots of these results. A significant increase in melanosome density was obtained for pigmented spot sites of all groups compared with nonpigmented areas, similar to the melanosome content. Melanosome content and melanosome density in pigmented spot sites tended to be lower in the female groups than in the male groups.
Shape descriptor indices of the dermal papillae, such as circularity and solidity, were significantly decreased in the group of dark-coloured pigmented spots in males compared with surrounding non-pigmented areas ( Table 1). Figure 9 shows three-dimensional stereoscopic images of rete pegs in characteristic dark-coloured pigmented spot sites in males and females. Detailed observation of stereoscopic images constructed by stacking horizontal images from CLSM of typical dark-coloured pigmented spots in male subjects showed that a few rete pegs were entwined in a twisting manner and extended into the dermis. These deformations were not observed in light-coloured pigmented spots in male subjects or any pigmented spots in female subjects. The total area of basal cells surrounding dermal papillae in horizontal sections was increased in the groups of dark-coloured pigmented spots in males and pigmented spots in females compared with nonpigmented areas. The degree of increase in the total area of basal cells was smaller than the degree of increase in melanosome content.
Contrasting with these results, no significant differences in the thickness of the SC were observed in any measurement groups. In addition, only the group of darkcoloured pigmented spots in males showed significant  That is, deformation of the structure of the epidermis at the pigmented spot site appeared confined to the lower spinous cell layer and basal cell layer. In particular, in dark-coloured pigmented spots in males, it is possible that marked proliferation accompanies deformation of the rete pegs.
Some light-coloured pigmented spots in females showed three characteristic patterns of internal structure with no clear observation of dermal papillae. Figure 10 shows the three typical horizontal and vertical cross-sections from CLSM images of these peculiar light-coloured pigmented spot sites in females. These patterns comprised reticulated structures, radial structures around the hair root, and planar structures. These vertical cross-sectional images confirmed that epidermal basal cells showing a reticulated or planar morphology were localized in a relatively shallow part of the epidermis. On the other hand, basal cells, which show a morphology that radiates from the hair root, existed at the same depth as the pigmented spots that could be observed in the dermal papillae.
Plasmin-like activity in exfoliated SC was compared between solar lentigo of the cheek, the non-pigmented area of the cheek, and the medial upper arm as a nonsun-exposed area. As a result, plasmin-like activity was significantly higher at the solar lentigo compared with the non-pigmented area (p < 0.01). Moreover, plasmin-like activity was significantly higher at the solar lentigo and nonpigmented area on the cheek compared with the inner upper arm (p < 0.01) (data not shown).
Cheek L* values were subtracted from the inner upper arm L* values to quantify the effect of UV exposure on skin tone (Δ (I-C) L*). Figure 11 showed the correlation analysis between Δ (I-C) L* values and plasmin-like activity in exfoliated SC obtained at the same sites. The correlation coefficient was r = 0.45 and statistical analysis of the correlation coefficient showed a significant correlation (p < 0.01).

DISCUSSION
Measurements of the darkness of pigmented spots in the cheek of male and female subjects by age group showed that male subjects over 40 years old had darker spots than female subjects, and large, conspicuous pigmented spots in the cheek increased earlier in male subjects than in female subjects. These results suggest that the face of male subjects may be more susceptible to generating darker pigmented spots than female subjects due to intrinsic and extrinsic factors such as UVB exposure, hormonal balance and inflammation. It should also be considered that malespecific lifestyle habits must contribute to the darker pigmented spots.
Morphological changes of dark-and light-pigmented spots were investigated using parameters quantified from image analysis of the internal structure of the epidermis as constructed using images from CLSM. A significant increase in the maximum thickness of the epidermis indicates that the rete pegs extend towards the dermis in the area of pigmented spot sites. Based on non-invasive histomorphological observation, a significant increase in melanosome content indicates that melanosomes accumulate in the proliferated epidermal F I G U R E 3 Darkness of pigmented spots on cheeks in male and female subjects by age group. The ΔL* values were measured at 15-39 pigmented spots in the 40s and 50s age groups for male and female subjects. Student's t-test was applied for statistical evaluations. No significant differences in ΔL* values were observed in the 30s age group when the number of pigmented spots was low. Age groups basal cells at pigmented spots. The significant increase in melanosome density is attributed to an increase in the proportion of basal cells containing melanosomes. The number of epidermal basal cells was significantly increased in pigmented spot sites, but no significant change in SC thickness was observed in any measurement groups. In addition, only the group of darkcoloured pigmented spots in males showed a significant change in the minimum thickness of the epidermis. Moreover, the mean area of epidermal basal cells in horizontal sections was confirmed to show no significant change compared with the non-pigmented area. From these observations, morphological changes to rete pegs were speculated to be promoted by the proliferation of epidermal basal cells taking priority over altering the differentiation of epidermal cells. F I G U R E 5 Colour microscopic images and in vivo confocal laser scanning microscopy (CLSM) block images. Colour microscopic images of dark-coloured pigmented spot sites in males (a) and females (b), and CLSM block images inside the red dashed lines of a (c) and b (d). The internal structure of the skin differs between pigmented spots and surrounding non-pigmented areas. Strong melanosome-derived reflexes are observed in pigmented spot sites. Significant deformation of the rete pegs, represented by significant increases in area and decreases in shape descriptor indices of dermal papillae, such as circularity and solidity (Table 1), could indicate excessive proliferation of epidermal basal cells in dark-coloured pigmented spot sites in the male subjects. Pigmentation could be expected to be darker as proliferation of basal cells is increasingly promoted. Since the number of basal cells increased significantly in the typical dark-coloured pigmented spots in the male subjects, basal cells were considered to excessively spread in the horizontal direction in the basal layer, not only in the vertical direction. As a result, basal layers surrounding multiple dermal papillae were expected to coalesce and take on an intricate shape in the epidermis. The fusion of extended epidermal rete pegs has been observed in the histopathology of seborrheic keratosis sites, and the image of dark spots in males in Figure 9 may represent a three-dimensional image of fused epidermal rete pegs [19]. Epidermal basal cells in the rete pegs that deformed and increased in volume could thus be speculated to play a role in the accumulation of melanosomes. From a clinical perspective, the surface swelling seen in seborrheic keratosis may be attributable to excessive proliferation of epidermal basal cells containing melanosomes.
In the female subjects of the same age group, the abnormal deformation of rete pegs as seen in typical darkcoloured pigmented spot sites in the male subjects was not observed, and the degree of deformation of dark-coloured pigmented spot sites was similar to that in light-coloured pigmented spot sites in the male subjects. Furthermore, the degree of deformation of light-coloured pigmented spot sites in the female subjects was confirmed to be less than that of dark-coloured pigmented spot sites. Based on these facts, proliferation of epidermal basal cells at pigmented spot sites is considered to be related to the accumulation of melanosomes.
Inflammation caused by UV radiation is thought to be one of the factors promoting the proliferation of epidermal basal cells. The present study confirmed that the expression level of proinflammatory protease plasmin was higher at the cheek than at the inner upper arm, and plasmin-like activity in the SC at the cheek increased significantly as skin darkened.
Plasmin is converted from plasminogen by plasminogen activator (PA). Plasminogen is localized to the peripheral cytoplasmic region of epidermal basal cells, and plasmin produced by the action of keratinocyte PA plays a role in epidermal differentiation [20]. A recent study showed that plasmin plays an important role in promoting   the release of Kit ligands by activating matrix metalloproteinases [21]. Many epithelial tissues appear to express the transmembrane tyrosine kinase receptor KIT only when they are transformed but have low malignancy, which might reflect a situation in which proliferation is stimulated, and differentiation is suppressed [22,23], so high KIT expression in the basal layer of solar lentigo lesions is considered to be associated with the epidermal hyperplasia of rete pegs [6]. These reports hint at a possible relationship between the proliferation of epidermal basal cells and the activation of plasmin in pigmented spot sites. Epidermal stem cells reside in an underlying basal layer, where they either self-renew within this compartment by symmetric division or differentiate with upward delamination by asymmetric division to contribute to the permeability barrier of the SC [24,25]. In other words, the epidermis is thought to maintain homeostasis by optimizing the balance of these two types of basal cell division [24,26]. These facts indicate that elongation of the rete pegs observed at pigmented spot sites may have been induced by the proliferation of stem cells via symmetric division.
Some light-coloured pigmented spot sites in the female subjects showed no clear dermal papillae but exhibited special structures visible as geometric patterns. A characteristic basal layer structure was seen, rather than the classic basal layer structure in which the rete pegs and dermal papillae are staggered and regularly aligned. These structures are also thought to be the result of anomalous basal cell division but could not be explained in detail.
In general, the male face is expected to be more susceptible to UV damage, due to the less frequent use of makeup and lower consciousness of UV protection [27]. A previous report indicated that male skin was easily affected by UV and oxidation stress compared with female skin [28]. Furthermore, the male face is thought to be easily inflamed by daily shaving. In fact, our awareness studies also found that male subjects were less conscious of skincare and less likely to use sunscreen on a daily basis. In addition, many male subjects responded that they were not concerned with pigmented spots on the face, even if such pigmented spots were actually present. However, the favorability women expressed regarding male appearance was significantly reduced with increases in the size, number and density of pigmented spots in male subjects (data not shown).
These facts may indicate that males are more in need of UV protection and anti-inflammatory agents for the prevention of developing darker pigmented spots.

CONCLUSION
Parameters measured from CLSM images indicating the internal structure of the epidermis suggest that melanosomes are accumulated in extended rete pegs in pigmented spot sites on the cheeks, resulting in darkening of the skin

F I G U R E 9
Three-dimensional stereoscopic images of rete pegs in dark-coloured pigmented spots from males and females constructed from stack images of horizontal sections after the binarization processes. Multiple rete pegs are intertwined to form a large and intricate complex structure in dark-coloured pigmented spots from male subjects. The rete pegs in dark-coloured pigmented spots from female subjects are elongated smoothly and independently.
surface. The number of melanosomes that accumulate in epidermal basal cells increased with the elongation of rete pegs, and the surface colour tone of pigmented spots also darkened. The cause of this elongation of rete pegs could be considered to involve proliferation of epidermal basal cells induced by inflammatory reactions involving intrinsic and extrinsic factors such as UVB exposure, and inflammatory substances including plasmin. Suppression of not only the synthesis of melanin in melanocytes with skin-lightening products but also the excessive proliferation of epidermal basal cells including melanosomes with daily UV protection by sunscreen products and antiinflammatory products after shaving could, thus, be expected to prove useful in preventing and improving darker pigmented spots.

ACKNO WLE DGE MENTS
We wish to express our deep appreciation to Professor and Chairman Daisuke Tsuruta and Associate Professor Toshiyuki Ozawa, Department of Dermatology, Osaka Metropolitan University Graduate School of Medicine, for their guidance on the diagnosis of pigmented lesions and for their valuable advice and fruitful discussions throughout this study.

CONFLICT OF INTEREST STATEMENT
The authors state no conflict of interest.
F I G U R E 1 0 Typical horizontal and vertical sections of in vivo confocal laser scanning microscopy images of light-coloured pigmented spots from female subjects. Basal cells containing melanosomes form a reticulated structure at a relatively shallow depth (a), a radial structure around the hair root (b) or a planar structure (c). Red lines on images indicate the positions of each crosssection.