Morphological and transcriptional evaluation of multiple facial cutaneous hyperpigmented spots

Abstract

Background: Morphological characteristics of major facial hyperpigmented spots have been well documented. However, detailed alterations of respective transcriptional profile for each spot and in-depth comparisons across multiple spot types have not been reported.

Objectives: To comprehensively assess and compare multiple facial hyperpigmented spot types at the morphological and molecular levels by utilising transcriptional expression profiling with correlation to quantified histological features.

Methods: Multiple types of facial spot biopsies were collected from Chinese women and compared to additional biopsies taken from adjacent healthy skin. The types of spots included Solar Lentigos with both elongated dermal-epidermal junction (DEJ) (SL[E]) and flat DEJ (SL[F]), Seborrhoeic Keratosis (SK), Melasma, Freckles, Post-inflammatory hyperpigmentation of resolving acne (PIH[A]) and other stimuli (PIH[O]). Combined histomorphometry, immunohistology, and transcriptome analysis for suprabasal-epidermis, basal-epidermis, and dermal compartments dissected by Laser Capture Microdissection (LCM) were conducted and compared across different spot types.

Results: Each spot type was confirmed to have the unique histological pathology already documented elsewhere. Most of the spot types except Melasma and PIH (A) revealed similar melanocyte density to adjacent skin. All spots exhibited increased melanin synthesis, melanosome transportation, as well as enhanced melanocyte dendricity, however, each spot revealed a distinct transcriptome regulation pattern in pigmentation pathways. Upregulation of pigmentation genes was also observed in the dermis of SL(F), SL(E), SK and PIH(O), associated with significant modulation of DEJ related genes in basal-epidermis and/or dermal compartments, suggesting potential melanocyte infiltration into the dermis due to impaired DEJ quality. Beyond upregulated pigmentation, for most spots, gene expression in the suprabasal-epidermis regulating keratinisation was significantly upregulated in conjunction with thickened stratum corneum. Furthermore, downregulation of tight junction related genes represented by claudin-1 was observed in majority of spot types, suggesting compromised barrier function could be a similarity across spots. Additionally, Cyclin-Dependent Kinase Inhibitor 2A (CDKN2A) was upregulated in all types of spots, indicating involvement of cell senescence as a common theme.

Conclusion: This comprehensive and comparative study based on the histological and transcriptional analysis of three skin compartments provided unique insights into specific causations as well as differences and similarities across multiple hyperpigmented spot types.

FIGURE 1

Representative images of haematoxylin & eosin (Columns a and b) and Fontana Masson (Columns c and d) stained skin biopsies of both non‐spot (Columns a and c) and seven different spot types (Columns b and d). Scale bar = 200 microns

FIGURE 2

Skin feature analysis on (a) epidermal thickness (b) rete ridge undulation (c) stratum corneum thickness from H&E images and (d) melanocyte density from MITF images. Mean ± SEM. **p < 0.05, *p < 0.1 versus non‐spot (NS) by student t‐test. () indicates significantly different to lower direction

FIGURE 3

Representative images of Pmel17 immunofluorescent staining of skin biopsies from both non‐spot (Column a) and 7 different spot types (Column b). Scale bar = 100 microns

FIGURE 4

Overall gene expression patterns of seven spot types in each layer; suprabasal epidermis, basal epidermis and dermal. Cluster analysis (groupings above the heat maps) indicate the similarity of expression patterns of each spot within each skin layer. Green indicates downregulation, red indicates upregulation with colour intensity signifying magnitude of fold change. Numbers in parenthesis under spot names are significantly modulated probe set numbers versus adjacent normal skin (p < 0.05). Total probe set numbers of used Affymetrix GeneTitan U219 array plate is 49 293

FIGURE 5

GAGE analysis on melanin synthesis (a and b) and dermal‐epidermal junction (c and d) related gene sets in basal epidermis (a and c) and dermal compartments (b and d). Red dash lines represent GAGE p = 0.05 thresholds. Log10 based GAGE p‐values are plotted with positive values representing upregulation and negative value representing downregulation of a gene set activity

FIGURE 6

GAGE analysis in melanogenesis and microtubule activities for all spot types at basal epidermis layer based on GO terms. The numbers in the table are Log10(p) of GAGE p‐values; positive value indicate upregulation; negative values indicate downregulation. The colour legend indicates colour intensity with p‐value range

FIGURE 7

Key biological processes associated with epidermal development for all spot types at suprabasal epidermis and basal epidermis layers based on GO terms. The numbers in the table are Log10(p) of GAGE p‐values; positive value indicate upregulation; negative values indicate downregulation. The colour legend indicates colour intensity with p‐value range

FIGURE 8

Regulation of key tight junction genes for suprabasal epidermis layer in all spot types. The numbers in the table are log2 fold changes versus adjacent normal tissue. The colour legend indicates colour intensity with p‐value range. Red indicates upregulation; blue indicates downregulation

FIGURE 9

(a) Log2 fold changes of CDKN2A (a cellular biomarker of ageing and cell senescence) gene expression in both suprabasal epidermis and basal epidermis layers versus adjacent normal tissue for all spot types (b) Example images of CDKN2A (p16) protein immunostaining for PIH(a) spot tissue and non‐spot tissue. Scale bar = 100 microns