The Patterns of Melanosome Distribution in Keratinocytes of Human Skin as One Determining Factor of Skin Colour

H.-Y. Thong, S.-H. Jee, C.-C. Sun, R.E. Boissy


The British Journal of Dermatology. 2003;149(3) 

In This Article

Materials and Methods

Panellist recruitment and biopsy removal for this study were performed at the Departments of Dermatology at National Taiwan University Hospital, Taipei, Taiwan and University of Cincinnati Hospital, Cincinnati, OH, U.S.A. All procedures were subjected to internal ethical approval prior to commencement of the study, and all panellists were required to give informed consent before participating.

Healthy volunteers with skin of Fitzpatrick phototype IV-V were recruited from the Chinese residents in Taipei, Taiwan (n = 15). Specifically, biopsies from three boys (10-13 years) and three men and three women each of young adult (22-31 years) and older (47-73 years) age groups were obtained. Healthy volunteers of skin phototype II (n = 3) and VI (n = 3) were recruited from American residents in Cincinnati, OH, U.S.A. Specifically, biopsies were from three Caucasians (men aged 22, 25 and 49 years, respectively) and three African/American individuals (two men aged 18 years and a woman aged 52 years).

Three or four 4-mm punch biopsies were subsequently obtained by a qualified physician from the volar aspect of the forearm (4 cm distal to the elbow flex joint). The designated area was first cleansed with a sterile alcohol gauze pad. Local anaesthesia with 2% xylocaine, with or without adrenaline, was administered to the biopsy site and a 2-3 mm thick plug of skin was obtained from the designated site using a sterile biopsy punch and sterile scissors to remove the tissue. The biopsy site was then sutured or dressed as deemed necessary by the physician.

The skin obtained was fixed with half-strength Karnovsky's fixative overnight at 4°C as previously described.[10]

For ultrastructural study of the skin, the tissue was then cut into quarters and processed for routine electron microscopy. All specimens were washed three times in 0.2 mol L-1 sodium cacodylate buffer and treated with 1% osmium tetroxide containing 1.5% potassium ferrocyanide for 30 min. The tissues were then washed, stained en bloc with 0.5% uranyl acetate for 30 min, dehydrated and embedded in Eponate 12. The specimens were then sectioned on an RMC MT 6000-XL ultramicrotome. Ultrathin sections were stained with aqueous solutions of uranyl acetate (2%) and lead citrate (0.3%) for 15 min each, and viewed and photographed in a JEOL JEM-1230 transmission electron microscope. All tissue-processing supplies were purchased from Ted Pella (Tustin, CA, U.S.A.); all tissue sectioning and electron microscopic analysis were performed in Cincinnati, OH, U.S.A.

The distribution pattern, morphology and size of melanosomes in keratinocytes were analysed from electron micrographs using computer-assisted image analysis. For melanosome distribution, 20 micrographs (at x 15 000 magnification) were obtained for each individual in the various sex and age groups described above. These micrographs were subsequently analysed for number of melanosomes per basal keratinocyte within all keratinocytes in each field, and grouped as being individually distributed or in membrane-bound clusters of two to three, four to six, or greater than six.

For melanosome size, melanosomes within keratinocytes were assessed for size (i.e. area) from blinded and coded electron micrographs randomly selected from each individual within each age and sex group, using a Zeiss LSM Image Browser version 2.30.011. Firstly, from each coded group of micrographs, the area of transferred melanosomes (µm2 x 10-2) was determined for 200 random (indiscriminately measured individual and clustered) unclassified melanosomes within all keratinocytes in each field. Secondly, the areas (µm2 x 10-2) for 100 individually distributed melanosomes and the areas (µm2 x 10-2) for 100 clustered melanosomes were determined within all keratinocytes in each field.

Statistical analysis was performed to determine the variance between age and sex groups and between the racial groups. The parameters of interest tested were the population mean and variance. A 100 (1- )% confidence interval for the mean was constructed. The relevant test statistic was Z. To confirm an expected difference between two investigated groups (individual means compared and clustered means compared for each group), the P-values for the Z statistic of the one-tailed test should be less than = 0.05. One-way ANOVA was used to compare the differences in the mean values of the melanosome size among various skin types. All calculations were performed using Microsoft Excel 2000 (Microsoft Corp., Seattle, WA, U.S.A.) and SigmaStat (SPSS Inc., Chicago, IL, U.S.A.).