Biological age reveals personal skin aging variations - An innovative skin aging index based on non-invasive internal skin RNA expression
Podium 48
Presented by: Shodai Tanaka
Introduction
Methods and Results
To examine whether SSL-RNA profiles associate with skin aging, we extracted 368 genes correlated with chronological age significantly among women aged 20 to 59 years. Gene ontology analysis of these genes showed consistent changes with previously reported aging-related biological functions, including inflammation, cell death, cellular senescence, epidermal differentiation, ATP synthesis, and others. These results suggested that SSL-RNA profiles included the information regarding biological process related to skin aging, thus enabling non-invasive estimation of biological age. Then, we attempted to construct the machine learning model for prediction of chronological age based on the expression levels of 368 age-correlated genes, resulting in having a generalizing capability because of a high correlation between predicted and actual chronological age. Therefore, predicted age based on the SSL-RNA profile was defined as “biological age”. Among subjects with the same chronological age, some showed higher and others showed lower biological age as compared to chronological age. We verified whether biological age estimated by our model accurately reflects the degree of skin aging. Skin aging parameters, such as elasticity and roughness, were compared between the higher and lower biological age groups. As a result, higher group showed skin with significantly greater levels of aging, with more skin roughness and less skin elasticity for instance, suggesting that biological age estimated from SSL-RNA is a valid index of skin aging.
Discussion and Conclusion
Various biological functions maintain skin homeostasis and help to retain a healthy skin condition, thus a homeostatic imbalance gradually leads to skin aging. Chronological age is not the only contributor to that, as a complex group of variables cause various degrees and phenotypes related to skin aging. In cosmetic research settings, it is difficult to perform a detailed examination of the skin aging mechanism using comprehensive information regarding biomolecules contained in skin due to the high level of invasiveness. In contrast, SSL-RNA provides a non-invasive method for acquiring internal skin information as a comprehensive gene expression profile. We examined changes in SSL-RNA profiles associated with aging and estimated biological age by predicting chronological age based on those findings. Estimations of biological age showed close associations with skin aging parameters, suggesting that biological age reflects the degree of skin aging. In other words, an understanding of biological age may help to unravel various skin aging factors in individuals and reveal the degree of internal skin aging, which could not be assessed before the appearance of aging phenotypes. We consider that biological age is an innovative index useful for examination of skin aging that provides more functional understanding of that factor in individuals in a timely manner as well as a personalized cosmetic future plan.
Skin aging is an inevitable process, though the degree of progression and phenotype, such as shown by spots, wrinkles, and sagging, vary among individuals. While chronological age is one of the most common general indexes of skin aging, it is not always an accurate indicator of degree or phenotype. Recently, the concept of biological age, estimated based on amount of decline in physical functions, has been used indicate the degree of aging and drawn global attention. It is also known that information regarding biomolecule changes associated with aging, such as RNA and proteins, can be a helpful tool for estimating biological age. In our previous study, the presence of human RNA in sebum, termed skin surface lipids-RNA (SSL-RNA), was examined. Since SSL-RNA samples are obtained by only wiping off sebum using an oil blotting film, specimens for examinations of human comprehensive gene expression (SSL-RNA profile) can be acquired in a non-invasive manner. Based on our previous research, it has been suggested that SSL-RNA profiles reflect various skin and body conditions, such as atopic dermatitis, circadian rhythm, and menstrual cycle. In this study, in order to understand the personal skin aging non-invasively, we aim to examine whether SSL-RNA profiles also reflect skin aging and estimate biological age using SSL-RNA profiles and machine learning.
Methods and Results
To examine whether SSL-RNA profiles associate with skin aging, we extracted 368 genes correlated with chronological age significantly among women aged 20 to 59 years. Gene ontology analysis of these genes showed consistent changes with previously reported aging-related biological functions, including inflammation, cell death, cellular senescence, epidermal differentiation, ATP synthesis, and others. These results suggested that SSL-RNA profiles included the information regarding biological process related to skin aging, thus enabling non-invasive estimation of biological age. Then, we attempted to construct the machine learning model for prediction of chronological age based on the expression levels of 368 age-correlated genes, resulting in having a generalizing capability because of a high correlation between predicted and actual chronological age. Therefore, predicted age based on the SSL-RNA profile was defined as “biological age”. Among subjects with the same chronological age, some showed higher and others showed lower biological age as compared to chronological age. We verified whether biological age estimated by our model accurately reflects the degree of skin aging. Skin aging parameters, such as elasticity and roughness, were compared between the higher and lower biological age groups. As a result, higher group showed skin with significantly greater levels of aging, with more skin roughness and less skin elasticity for instance, suggesting that biological age estimated from SSL-RNA is a valid index of skin aging.
Discussion and Conclusion
Various biological functions maintain skin homeostasis and help to retain a healthy skin condition, thus a homeostatic imbalance gradually leads to skin aging. Chronological age is not the only contributor to that, as a complex group of variables cause various degrees and phenotypes related to skin aging. In cosmetic research settings, it is difficult to perform a detailed examination of the skin aging mechanism using comprehensive information regarding biomolecules contained in skin due to the high level of invasiveness. In contrast, SSL-RNA provides a non-invasive method for acquiring internal skin information as a comprehensive gene expression profile. We examined changes in SSL-RNA profiles associated with aging and estimated biological age by predicting chronological age based on those findings. Estimations of biological age showed close associations with skin aging parameters, suggesting that biological age reflects the degree of skin aging. In other words, an understanding of biological age may help to unravel various skin aging factors in individuals and reveal the degree of internal skin aging, which could not be assessed before the appearance of aging phenotypes. We consider that biological age is an innovative index useful for examination of skin aging that provides more functional understanding of that factor in individuals in a timely manner as well as a personalized cosmetic future plan.