Investigation into the significant role of dermal-epidermal interactions in skin ageing utilising a bioengineered skin construct

Abstract

Increased prevalence of skin ageing is a growing concern due to an ageing global population and has both sociological and psychological implications. The use of more clinically predictive in vitro methods for dermatological research is becoming commonplace due to initiatives and the cost of clinical testing. In this study, we utilise a well-defined and characterised bioengineered skin construct as a tool to investigate the cellular and molecular dynamics involved in skin ageing from a dermal perspective. Through incorporation of ageing fibroblasts into the dermal compartment we demonstrate the significant impact of dermal-epidermal crosstalk on the overlying epidermal epithelium. We characterise the paracrine nature of dermal-epidermal communication and the impact this has during skin ageing. Soluble factors, such as inflammatory cytokines released as a consequence of senescence associated secretory phenotype (SASP) from ageing fibroblasts, are known to play a pivotal role in skin ageing. Here, we demonstrate their effect on epidermal morphology and thickness, but not keratinocyte differentiation or tissue structure. Through a novel in vitro strategy utilising bioengineered tissue constructs, this study offers a unique reductionist approach to study epidermal and dermal compartments in isolation and tandem.

Keywords: ageing; bioengineering; dermal‐epidermal crosstalk; extracellular matrix; human skin equivalent; inflammation.

Figure 1

Ageing dermal compartments exhibit decreased extracellular matrix deposition. Neonatal or ageing fibroblasts were seeded in Alvetex® Scaffold and cultured for 30 days before being harvested for analysis (a). H&E staining of dermal compartments generated with neonatal (bi) and ageing fibroblasts (bii) demonstrate fibroblast population of the scaffold. Representative immunofluorescence staining of collagen I (biii) and collagen III (biv) is reduced in ageing dermal compartments. Collagens are stained green and nuclei are stained with DAPI. Scale bars: B = 50 μm.

Figure 2

Coculture of an ageing dermal compartment with neonatal keratinocytes results in epidermal thinning independent of keratinocyte proliferation. Dermal compartments were generated using both neonatal and ageing fibroblast populations, upon which, neonatal keratinocytes were seeded and allowed to stratify at the air‐liquid interface for 14 days before analysis (a). Representative H&E stained images display significant epidermal thinning upon an ageing dermis (bi, ii). Immunofluorescence staining of proliferation, stemness and senescence‐associated biomarkers (green): p63 (biii, iv), Ki67 (Bv‐vi), HMGB1 (vii‐viii) and Lamin b1 (bix,x). Nuclei are stained in blue. Quantification of epidermal thickness (c) and number of viable epidermal layers (d) are significantly reduced in the presence of an ageing dermal compartment. Whereas, the percentage of ki67, p63 and HMGB1 remains unchanged regardless of dermal foundation (e). Data represent mean ± SEM, n = 9 and a two‐tailed, unpaired t‐test with Welch's correction was performed where: ****p < 0.0001 and p > 0.05 is nonsignificant. Scale bars: Bi, ii = 50 μm, Biii‐x = 25 μm.

Figure 3

Coculture of ageing fibroblasts with neonatal keratinocytes significantly impacted keratinocyte morphology but not differentiation or intracellular junctions. Representative immunofluorescence images of dermal‐associated extracellular matrix (ECM) proteins (a): collagen I (ai, ii), collagen III (aiii, iv) and integrin α6 (av, vi), stained in green. Immunofluorescence images of epidermal intracellular junctional biomarkers (b): claudin‐1 (bi, ii), periplakin (biii, iv) and E‐cadherin (bv, vi), stained in green. Representative immunofluorescence analysis of biomarkers associated with epidermal differentiation (c) including: keratin‐10 (green) and keratin‐14 (red, ci, ii), involucrin (ciii, iv) and filaggrin (cv, vi) stained in green. In all immunofluorescence images nuclei are stained with DAPI in blue. Biometrics quantification of aspects of basal keratinocyte morphology including area (d), width (e), height (f) and height:width ratio (g) are significantly impacted in the presence of an ageing dermis. Data represent mean ± SEM, n = 9 and a two‐tailed, unpaired t‐test with Welch's correction was performed where ****p < 0.0001. Scale bars: 25 μm.

Figure 4

Ageing fibroblast conditioned medium significantly impacts epidermal morphology in a monoculture epidermal construct, independent of keratinocyte proliferation. Populations of neonatal or ageing fibroblasts were expanded in 2D culture and samples of their conditioned medium obtained for analysis or application to tissue constructs. Epidermal‐only human skin equivalents (EO‐HSEs) containing neonatal keratinocytes were cultured for 7 days at the air‐liquid interface to form a mature epidermis before fibroblast conditioned medium was added to the constructs for a further 7 days before analysis (a). (a) Representative H&E staining of EO‐HSEs cultured in neonatal or fibroblast conditioned medium (bi, ii). Representative immunofluorescence staining of biomarkers (green): p63 (biii, iv), ki67 (bv, vi) and HMGB1 (bvii, viii). Nuclei are stained in blue by DAPI. Biometric analysis of epidermal thickness (c) and number of viable epidermal layers (d) reveal significant differences between the two conditioned media types. Whereas, quantification of percentage positive epidermal cells of ki67, p63 and HMGB1 (e) reveals no significant difference between conditions. Data represent mean values ± SEM, n = 9 and a two‐tailed, unpaired t‐test with Welch's correction was performed where **p < 0.01, ****p < 0.0001 and p > 0.05 is nonsignificant. Scale bars: bi, ii = 50 μm, biii‐viii = 25 μm.

Figure 5

Ageing fibroblast conditioned medium significantly impacts basal keratinocyte morphology but not differentiation and stratification of neonatal keratinocytes in a monoculture epidermal construct. Representative immunofluorescence images of epidermal‐only HSEs (EO‐HSEs) cultured in medium conditioned by either neonatal or ageing fibroblasts and stained for keratinocyte differentiation related biomarkers: keratin‐10 (green) and keratin‐14 (red, ai, ii), and loricrin (aiii, iv). Nuclei are stained by DAPI in blue. Quantification of aspects of basal keratinocyte morphology reveal significant differences, including: area (b), height (c), width (d) and height:width ratio (e). Data represent mean values ± SEM, n = 9 and a two‐tailed, unpaired t‐test with Welch's correction was performed where **p < 0.01, ***p < 0.001, ****p < 0.0001. Scale bars: 25 μm.

Figure 6

Ageing fibroblasts exhibit a more pro‐inflammatory secretome as opposed to neonatal fibroblasts. A significant increase in the concentration of pro‐inflammatory cytokine release was observed from medium conditioned by ageing fibroblasts. Concentrations of cytokines such as: TNF‐α (a), GM‐CSF (b), IL‐6 (c), IL‐8 (d), IL‐10 (e), IL‐13 (f) and MCP‐1 (g) were all significantly increased in conditioned medium collected from ageing fibroblast populations. Data represent mean values ± SEM, n = 6 and a two‐tailed, unpaired t‐test with Welch's correction was performed where **p < 0.01, ***p < 0.001, ****p < 0.0001.