The glyoxalase enzymes are differentially localized in epidermis and regulated during ageing and photoageing

Abstract

Skin ageing is a complex process resulting from both chronoageing, also called intrinsic ageing, and photoageing or extrinsic ageing. Age-related skin modifications are mainly attributed to molecular damages including alteration of proteins that can be modified by many oxidative processes such as glycation 1. In addition to reaction with glucose, advanced glycation endproducts (AGE) can also be produced when the dicarbonyl compounds glyoxal (GO) and methylglyoxal (MG) react with proteins. Detoxification of GO and MG compounds is mainly achieved by the glyoxalase (GLO) system, composed of two intra-cellular enzymes glyoxalase 1 (GLO1) and 2 (GLO2) that function in tandem 2. As GLO1 expression changes during ageing along the life in human and mice, as well as in animal models 3-7 and the GLO system seems to be less efficient with age 5, 6, the role of GLO1 and GLO2 has been studied in the ageing process. Overexpression of the system induces a prolonged lifespan, partly explained by a lower content in glycated mitochondrial proteins. Conversely, knock-down of GLO1 leads to hypersensitivity to MG and a shorter lifespan 6, 7. Even though accumulation of glycated modified protein was often associated with the skin ageing process, the role of GLO proteins in skin homoeostasis is still largely unknown. In this study, we addressed for the first time the localization and expression of GLO enzymes in young versus aged or photoexposed human skin samples. We showed that Nε-carboxymethyl-lysine (CML)-modified protein levels are controlled by the GLO system in keratinocytes, emphasizing its role in protein maintenance in skin epidermis and in protecting keratinocyte progenitors during chronological ageing and photoageing. The experiments were performed on biopsies from human skin or normal human adult epidermal keratinocytes (NHEK) isolated from abdominal skin surgery. We studied the effect of chronological and photoageing on GLO system by analysing the immunoreactivity and morphological distribution of GLO1, GLO2 and CML-modified proteins in young and old photoprotected/photoexposed human skin samples. Cultured human keratinocyte cells were used to investigate the regulation of GLO1 activity and GLO expression, by immunoblotting and qPCR experiments, upon ageing, replicative senescence and in vitro differentiation. Detailed methods are described in the supplementary data. Cryosections of young and aged human epidermis were immunostained with antihuman GLO1 and GLO2 antibodies, respectively (Fig. 1a, c). Glyoxalases are mainly detected in the epidermis with GLO1 mostly localized in the basal layer of epidermis, whereas GLO2 exhibited a prominent staining in upper layers keratinocytes. GLO staining was found to vary depending on the age of the donor, and surprisingly, it was shown stronger in old donors when compared to the young ones (Fig. 1b, d). Glyoxalases are involved in detoxification of GO and MG, which are responsible for the formation of AGE-modified intra-cellular proteins with CML, a GO-mediated AGE, described as the most prevalent AGE found in old skin. To determine how these enzymes are involved in protein protection during chronoageing, CML-modified proteins were detected by immunofluorescence in skin biopsies. As expected, CML labelling was more pronounced in the old dermis compared to the young ones but conversely, when we focused on the epidermis basal layer, it exhibited a significant lower intensity in old donors (Fig. 1e, f). This result indicates that during chronoageing lower AGE formation can be associated with GLO1 overexpression, suggesting a role for this enzyme in protecting the skin progenitor cells from the epidermis basal layer against ageing-dependent protein modifications. To understand the regulation of GLO enzymes during ageing and analyse the effect of chronoageing and replicative senescence, we used NHEK from 26 and 65 years old donors. Cells were submitted to replicative senescence up to passage 10, and expression of p16 was used as a marker of senescence (Fig. 1g). Senescence, as well as chronoageing, reduces GLO1 activity but only in keratinocytes from young donors (Fig. 1h). Nevertheless, this weaker activity is not accompanied by a modification of GLO1 protein level (Fig. 1g). It is well known that UV exposure amplifies the skin ageing process through ROS production, DNA alteration and protein oxidation. In young subjects, no obvious difference was found in GLO1 enzyme expression as well as in CML-bound protein levels in sun-exposed (forearm) versus sun-protected (upper inner arm) skin biopsies (Fig. S1). Conversely, in old donors, a huge accumulation of glycated proteins was detected upon photoexposure, mainly in the dermis compartment, but also in the epidermis (Fig. 2a, b). GLO2 enzyme showed a drastic lower expression in sun-exposed compared to sun-protected epidermis (Figs S1C, D, 2c, d), indicating that UV can alter glyoxalase system-mediated protection of keratinocytes, altho