Introduction
The Yellow Emperor Internal Classic is the earliest medical classic in China, written from the pre-Qin to the Western Han Dynasty. As the foundation of traditional Chinese medicine, this book reflects the wisdom of ancestors on health issues with rich elaborations on life formation, diseases, health preservation, etc., and guides our life through relevant theories and experiences gained from long-term practice (Liu et al., 2024). According to long-term medical practice, the Yellow Emperor Internal Classic takes “seven” as the life rhythm of women, and summarizes the characteristics of the circadian rhythm of women’s growth and aging. In “Suwen Ancient Innocence Treatise”, it’s written that “At three sevens, “Qi” in kidney is average, so the true teeth grow; at four sevens, the muscles and bones are strong, the hair is extremely long, and the body is strong; at five sevens, the Yangming pulse gets weak, the face is scorched, and the body starts declining” (Chen & Li, 2021). It means that at the age of 29-35, all organs of the body, mental state, psychological condition, etc. will be affected by aging. Sun et al. (2018) studied women’s follicular fluid through metabolomics and found significant differences in hormones and phospholipids in women at different age, revealing the connotation of the “Seven Seven Theory” in the Yellow Emperor Internal Classic. There is no doubt that the skin of the human body and its appendages are also affected by the “Seven Seven Theory”. In the traditional Chinese culture, it is believed that at the age of 35, the Yangming pulse began to decay. Since Yangming pulse correlates with the face and forehead, so wrinkles starts appearing on face and skin color gets gradually dark. Existing studies suggest that with age, thin epidermis and dermis layers in skin structure (Choi, 2019), decrease of Langerhans cells, and decrease of antigen-specific immunity change the immune composition of the skin (Zhang et al., 2019; West & Bennett, 2018). Thus, the skin barrier function is impaired, and the incidence of skin cancer and skin infections increases. However, the changes in different skin types with age are not clear, and the underlying markers of significant changes in the skin aging process remains to be identified.
Lipids are an important component of the skin barrier. In recent years, more and more studies have shown the effects of aging on the lipid composition of the skin (Choi, 2019; Chamers & Vukmanovic-Stejic, 2020), as well as on skin physiological parameters (Machková et al., 2018). Skin lipidomics is a part and extension of metabolomics (Raunegger et al., 2021), which can be used to comprehensively and systematically analyze the types and abundance of skin surface lipids (Smirnov et al., 2019). Our research group has established a complete set of UPLC-QTOF-MS-based sample collection, lipid extraction, data acquisition and analysis methods, and revealed the importance of lipids in skin status through the combination of lipidome with physiological indicators (Zhou et al., 2020a; Zhou et al., 2020b; Zhou et al., 2021).
To analyze the changes and trends of important lipid components in the aging process of facial skin in women with sensitive skin, in this study, we divided the women’s age stages according to the Yellow Emperor Internal Classic, and recruited Chinese female volunteers with sensitive skin aged 22-28 and 29-35 years old. Ultra-performance liquid chromatography quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS) was used to systematically analyze the changes of lipid composition. Omics technology was used to analyze the clustering of facial skin lipid groups, the content and characteristics of ceramides and fatty acids. The correlation between skin aging markers and skin physiological parameters including transepidermal water loss of facial skin, skin moisture content, skin surface pH, skin elasticity, skin color, wrinkles was explored. The changes of sensitive skin lipid components in the aging process and the influence on skin physiological parameters were discussed. In summary, our study provides precise skin care theory guidance for development of cosmetics for Chinese women with sensitive skin at age stages of 22-28 and 29-35 years old.
Materials and Methods
1. Study subjects and inclusion criteria
According to the age stages of women in the Yellow Emperor Internal Classic, a total of 65 females with sensitive skin in China were recruited, including 32 subjects aged 22-28 (S1 group, average age: 23.84±1.39) and 33 subjects aged 29-35 (S2 group, average age: 31.48±1.94). Exclusion criteria were as follows: major illnesses and being treated with medication, such as heart disease, liver disease, mental illness, etc.; serious disorders of one’s own endocrine system, such as diabetes mellitus; autoimmune diseases; patients who were uncooperative or did not complete required information. All volunteers were fully informed of the purpose of the experiment and signed an informed consent form. This is a non-invasive study that follows the principles of the Declaration of Helsinki.
2. Skin physiological parameters and lipid collection
Volunteers first clean their faces with water and then sit still for 30 minutes in a constant temperature (24℃) and relative humidity (50%). Considering the influence on facial lipids during the collection of physiological parameters, physiological parameters were collected after the lipid collection.
Sebutape® (purchased from CuDerm Corporation, Dallas, TX, USA) test strips were attached to the right cheek of subjects, and removed after three minutes. Subsequently, the test strips were put in sample tubes, and stored at -80℃ in the refrigerator.
Volunteer facial physiological indices were performed, including skin stratum corneum hydration (Keratometer CM 825); TEWL value (Tewameter TM 300); skin sebum (Sebumeter SM 815); skin pH (Skin pH meter 905); skin glossiness (Glossymeter GL200); skin melanin and red pigment (Mexameter MX 18); skin wrinkles, smoothness and roughness (Visioscan VC 98); skin elasticity (R2) (Cutometer dual MPA580); skin folds, smoothness and roughness (Visioscan VC 98).
3. Lipid extraction and sample preparation
The samples were removed from the -80℃ freezer. The lipids were extracted using the modified Bligh and Dyer method, followed by desiccation under a nitrogen blowing instrument. Finally, the lipids were reconstituted with methanol-isopropanol 1:1 v/v mixture (Zhou et al., 2020a; Zhou et al., 2020b; Zhou et al., 2021).
4. UPLC analysis conditions
According to the laboratory’s previous UPLC-QTOF-MS methodology (Wang et al., 2020), the following setup was used: column, Waters UPLC CSH C18 (1.7 µm, 2.1 mm×100 mm); Phase A (acetonitrile:water=4:6; 0.1% formic acid; 10 mmol/L ammonium formate) and Phase B (acetonitrile:isopropanol=9:1; 0.1% formic acid; 10 mmol/L ammonium formate); flow rate, 0.3 mL/min; injection volume: 2.0 µL; column temperature: 50℃.
Mass spectrometry was performed in positive ion mode using electrospray ionization (ESI) with a mass spectrum scanning range of 50-1200 m/z and a carrier gas of nitrogen. Leucine enkephalins (m/z 554.2771) was used as an external marker. Data acquisition was performed using MassLynx 4.1 (Waters Corporation, Milford, USA) data management software.
5. Data analysis
The raw data was processed using Progenesis QI 2.0 software and then transferred to Ezinfo 3.0 software and SIMCA 14.1 software for principal component analysis (PCA) and orthogonal projections of latent structural discriminant analysis (OPLS-DA). Important lipids were filtered with VIP>1, p-value<0.05, fold change>2. GraphPad Prism 8.0.2 was used to determine the significance of differences in biological parameters between two groups, *** p<0.001; ** p<0.01; * p<0.05. Finally, the human lipid database was used to search for these compounds and obtain the composition and information of the characterized lipids, and then the different lipids were analyzed by correlation analysis.
Results
1. Effects of aging on the physiological parameters of female sensitive skin
The skin barrier-related physiological parameters including moisture content, transepidermal water loss, oil content, and pH in S1 and S2 group are shown in Figure 1. The moisture content (p=0.506) and TEWL values (p=0.284) in S2 group were higher than those in S1 group. The oil content (p=0.156) was lower in S1 group than in S2 group. The pH value (p=0.804) did not change significantly. However, no significant differences on the above four physiological parameters were detected between S1 and S2 group.
The L* value, a* value, and b* value represent the black and white brightness, redness, and yellowness of skin color, respectively. The larger the L* value, the closer it is to white; the larger the a* value, the more red the skin tone; the higher the b * value, the more yellow the skin tone. The ITA value indicates the individual type angle, and a larger ITA value indicates a lighter skin tone.
Among the skin color-related physiological parameters, as shown in Figure 1, there was no significant difference in L* value (p=0.536) and a* value (p=0.207) between two groups. The b* value in S2 group was significantly higher compared to that of S1 group (p=0.013), while the ITA value showed opposite change (p=0.006). Thus, the skin tone of S2 group was significantly darker than that of S1 group.
Among the physiological parameters related to skin aging, as shown in Figure 1, the skin elasticity of the S2 group was significantly reduced compared with the S1 group (p=0.002). There was no significant difference in fine lines at the corners of the eyes between the two groups (p=0.613).
2. Facial lipid clustering in S1 and S2 groups
To evaluate the repeatability of the UPLC-QTOF-MS method quality control (QC) samples were analyzed. As shown in Figure 2A, QC samples clustered well, indicating stable machine performance and accurate experiments.
Following non-targeted lipidomics analysis, the OPLS-DA model was used for multivariate data analysis. The obtained model score plot is shown in Figure 2B. The difference between two groups is not significant (cumulative variance contribution rate is 31.46%, less than 50%), indicating minor difference in lipids between two groups.
After data collation of S1 and S2 groups, it was found that 605 lipids of the 917 lipids exhibited high concentration variation (the coefficient of variation CV mostly exceeded >100%. After filtering out these 605 lipids, 312 lipids were re-mapped (Figure 2C) and two groups separated well (cumulative variance contribution of 51.91%).
3. Relative abundance of facial lipids in S1 and S2 groups
The total lipid contents of the two groups was compared, and no significant difference in the relative abundance of total lipids between the two groups was found (p=0.649) (Figure 3A).
A total of 917 lipids were classified into lipid main classes, including 253 fatty acyls [FA], 113 glycerides [GL], 142 glycerophospholipids [GP], 72 polyketones [PK], 41 pregnancy esters [PR], 6 glycolipids [SL], 225 sphingolipids [SP], and 65 sterol lipids [ST]. As shown in Figure 3B, among the eight lipid classes, FA (p=0.162), GL (p=0.104), SP (p=0.940), and ST (p=0.437) in S2 group are higher than in S1 group, while GP (p=0.289), PK (p=0.526), of PR (p=0.047) and SL (p=0.530) in S2 group were lower than in S1 group. Only the relative abundance of PR was significantly different among two groups (p<0.05).
Twenty-six lipids were detected in PR class, including 24 [PR01], 1 [PR03], and 1 [PR04]. There were no significant differences in the three subclasses (p>0.05) (Figure 4).
4. Identification of differential lipid species between S1 and S2 groups
Based on the criteria of CV>100, p<0.05, Fold Change>2, VIP>1, 5 differential lipids (VIP lipids) were identified, including 1 FA, 3 GP, and 1 SP (Table 1).
5. Correlation between five VIP lipids and physiological parameters
As shown in Figure 5, 16,16-dimethyl-PGA2, PG(20:0/22:1 (11Z)), PA(20:0/22:2(13Z,16Z)) were significantly and negatively correlated with R2 and ITA values (p<0.05), PC(12:0/26:0) was significantly and negatively correlated with ITA values (p<0.05). The above four lipids were positively correlated with b * value and displayed higher relative levels in S2 group. Cer(d18:0/24:0) with higher relative content in S1 group was positively correlated with skin elasticity R2 and ITA values. Therefore, the changes in skin elasticity and color of female sensitive skin between age of 22-28 and 29-35 may be closely related to these five lipids.
Among them, the content of Cer(d18:0/24:0) was significantly different between the two groups (Figure 6).
Discussion
With age, the physiological state of the skin gradually changes. Based on data from thousands, multicenter, multi-ethnic, and multi-regional studies, prior research (Flament et al., 2021a; Flament et al., 2021b; Frederic et al., 2020; Flament et al., 2018; Flament et al., 2020) showed that increased wrinkles are the main feature of skin aging, which seems to be associated with changes in skin structure (Kim et al., 2020). In our study, female volunteers aged 22-28 years versus 29-35 years with sensitive skin did not show significant facial wrinkle changes according to facial physiological parameter measurement (Figure 3B). The significant decrease in facial elasticity (Figure 3A) may be due to the decrease of skin thickness, the loss of collagen and elastin, and the skin internal structure changes with age, but it has not caused the formation of external wrinkles. These data indicate that women with sensitive skin aged 29-35 years undergo “first aging”. Previous studies of facial aging in women have shown that skin radiance decreases with age (de Rigal et al., 2010). In our study, female volunteers with sensitive skin aged 29-35 years displayed significantly higher b * values than volunteers with sensitive skin aged 22-28 years. The value of ITA is significantly reduced in S2 group than S1 group. Among them, the b * value is related to oxidized proteins (carbonylated proteins, glycosylated proteins) (Ogura et al., 2011), carotenoids, etc. (Whitehead et al., 2012). The increase in b * value and the decrease in ITA value also showed the initial aging characteristics of gradually dull skin in women with sensitive skin aged 29-35 years.
It is important to note that our results differ from previous studies (Kim et al., 2019) in terms of skin barrier physiological parameters such as skin moisture content, transepidermal water dispersion, oil content, and pH. Kim et al. (2019) compared the physiological parameters of the skin barrier in healthy Chinese women at different ages, and concluded that the skin TEWL value and oil content decreased significantly with age (p<0.05). Similar results were obtained by Luebberding et al. (2013) and Zhao et al. (2021). However, in our study, the increasing age of women with sensitive skin did not result in significant changes in physiological parameters related to the skin barrier. We speculate that this may be due to the fact that sensitive skin has been accompanied by impaired skin barrier, reduced water content, and elevated TEWL values (McCormick et al., 2023; Anqi et al., 2022), and the effect of sensitive skin on the skin is stronger for a certain period of time than the change of skin barrier function by age. This needs to be proven by follow-up studies.
Lipid clustering showed that the facial lipids of the S1 and S2 groups had a good differentiation. In the analysis of lipid main classes, only pregnenolone esters exhibited significant differences. Pregnenolone lipids and their phosphorylated derivatives play an important role in the transport of oligosaccharides between cell membranes, and their derivatives have been shown to possess anti-inflammatory activity (Cai et al., 2021). Pregnenolone is also a hormone involved in the production of steroid hormones such as progesterone, mineralocorticoids, glucocorticoids, androgens, and estrogens (Rogers et al., 2012). The total facial pregnenol lipids in S2 group were significantly lower than those in S1 group, suggesting increasing facial inflammation and hormonal changes with age.
Combining the analysis of lipid classes and subclasses, we did not find lipid classes with significant differences. However, lipid subclasses with significant differences were found in lipid classes without significant differences, such as FA03 in S2 group than that in S1 group (p=0.044), when all lipid subclasses were combined. This means that the facial lipids of the subjects in S1 and S2 groups were not distinguished in main class, but there were significant differences in the subclass of the same main class (such as fatty acyls, sphingolipids, etc.). This is also confirmed by the study of VIP lipids. Among the screened VIP lipids, Cer(d18:0/24:0) content was higher in S1 group than in S2 group (Figure 6), which might be due to various factors such as the differentiation of skin keratinocytes and the sphingomyelin content. Previous study found that the expression level of keratinocytes in healthy subjects is often higher than that of psoriasis patients (Yokose et al., 2020). In a screening study of facial lipid markers in women with sensitive skin, Cer(d18:0/24:0) was more likely to occur in women with non-sensitive skin (Nakatsuji et al., 2016). This phenomenon possibly relates to the fact that formation of an epidermal lipid barrier contributes to its anti-inflammatory properties, suggesting that inflammation occurs in skin with age, which further increases sensitivity.
When an inflammatory response occurs in the skin, ceramide lipid (sphingolipids) contents reduced (Gruber et al., 2020). Ceramides are mainly produced from glucosylceramide catalyzed by the lysosomal enzyme glucocerebrosidase-1 (GCase). GCase is mainly regulated by exosome proliferator-activated receptor signaling pathways (PPARs), which promote the secretion of lipid components of the skin, including the formation and secretion of epidermal lipids and lamellar bodies, and enhances the activity of lipid-related metabolic enzymes (Hamilton et al., 2018). Studies have confirmed that activated PPARs inhibit the release of inflammatory factors and the differentiation of inflammatory cells and reduce UV-induced skin inflammation (Aioi, 2020). The reduction of facial ceramides in volunteers in the S2 group indicates that the inflammatory response occurs at all times in the process of skin aging, which is also consistent with the study of inflammatory aging of the skin (Jarrold et al., 2022; Chen et al., 2022).
Combined with the correlation analysis of VIP lipids and physiological parameters (b * value, ITA value, R2), PG(20:0/22:1(11Z)), PA(20:0/22:2(13Z,16Z)), PC(12:0/26:0) are phosphatidylglycerol, glycerophosphatidic acid, phosphatidylcholine, respectively, all of which belong to the glycerophospholipid compounds. A previous study investigating serum glycerophospholipid profile and aging-related changes in mice (Kim et al., 2014), showed that changes in glycerophospholipids are associated with aging and phosphatidylcholine content was higher in the older group, which is consistent with our findings. In addition, 16,16-dimethyl-PGA2 is a lipid-derived signaling molecule produced by arachidonic acid via cyclooxygenase, which inhibits collagen production and induces fibroblast expression of matrix metalloproteinase 1 (MMP1) in vitro (Shim, 2019). PGA2-induced inhibition of collagen expression and MMP1 promotion are aging mechanisms. Therefore, its expression was higher in S2 group, and there was a significantly and negatively correlation with R2 and ITA values.
Conclusion
Compared with the skin of volunteer aged 22-28 years, the skin of volunteer aged 29-35 years is characterized early aging in skin physiological parameters and facial lipids, and this aging is accompanied by inflammatory response. Our research provides new clues for preventing and treating sensitive skin aging. However, there are certain limitations in this study. For example, volunteers with non-sensitive skin were not included, and the sample size was relatively small. Therefore, more comprehensive and in-depth research is required to elaborate the mechanism underlying aging in sensitive skin.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation 
E-Mail 
Print 
