Genetic insights into the gut microbiota and risk of facial skin aging: A Mendelian randomization study

Abstract

Background

A growing number of experimental studies have shown an association between the gut microbiota (GM) and facial skin aging. However, the causal relationship between GM and facial skin aging remains unclear to date.

Methods

We conducted a two‐sample Mendelian randomization (MR) analysis to investigate the potential causal relationship between GM and facial skin aging. MR analysis was mainly performed using the inverse‐variance weighting (IVW) method, complemented by the weighted median (MW) method, MR‐Egger regression, and weighted mode, and sensitivity analysis was used to test the reliability of MR analysis results.

Results

Eleven GM taxa associated with facial skin aging were identified by IVW method analysis, Family Victivallaceae (p = 0.010), Genus Eubacterium coprostanoligenes group (p = 0.038), and Genus Parasutterella (p = 0.011) were negatively associated with facial skin aging, while Phylum Verrucomicrobia (p = 0.034), Family Lactobacillaceae (p = 0.017) and its subgroups Genus Lactobacillus (p = 0.038), Genus Parabacteroides (p = 0.040), Genus Eggerthella (p = 0.049), Genus Family XIII UCG001 (p = 0.036), Genus Phascolarctobacterium (p = 0.027), and Genus Ruminococcaceae UCG005 (p = 0.012) were positively associated with facial skin aging. At Class and Order levels, we did not find a causal relationship between GM and facial skin aging. Results of sensitivity analyses did not show evidence of pleiotropy and heterogeneity.

Conclusion

Our findings confirm the causal relationship between GM and facial skin aging, providing a new perspective on delaying facial aging.

Abbreviations

GM

gut microbiota

GWAS

Genome‐Wide Association Study

IVs

instrumental variances

IVW

inverse‐variance weighted

LD

linkage disequilibrium

MR

Mendelian randomization

SE

standard error

SNP

single nucleotide polymorphism

WM

weighted median

1. INTRODUCTION

As the largest organ of immune protection, the skin provides a physical barrier and immune defense for the body, and with increasing age, the immune system function declines and the skin barrier undergoes a typical immunological decline, with a gradual decrease in skin antimicrobial and immune response effectiveness and tumor cell defense, and impaired skin barrier integrity, known as skin immune senescence. ¹ Meanwhile, the appearance and texture of aging skin changes, which may affect an individual's self‐esteem and self‐confidence, inducing and exacerbating psychological problems of low self‐esteem and depression, which can negatively affect mental health.

Trillions of microorganisms exist in the gut, and the composition and activities of microorganisms constitute most of the processes of human health and disease. ² Therefore, the gut microbiota(GM) is also known as the “forgotten organ”. ³ In recent years, the important role of gut flora microorganisms in human aging has attracted widespread attention, ⁴ , ⁵ , ⁶ as a key factor in maintaining homeostasis in the host body, pathophysiological changes in the gut flora can cause immune senescence and inflammation during the aging process of the human body, thus accelerating aging‐related pathologies, ⁷ which have an impact on the quality of life of the elderly population. Facial skin aging, which is an important visual feature of organisms undergoing aging, includes structural and functional changes such as laxity, wrinkles, hyperpigmentation, reduced barrier effect capacity, and delayed wound healing. There is accumulating evidence that skin conditions are associated with gut microbes, and skin and gut cells originate from a uniform germ layer and share similar signal transduction and innervation pathways. ⁸ Gut microbial imbalance increases host vulnerability and disrupts mucosal immune tolerance, ⁹ which in turn induces a variety of skin disorders such as atopic dermatitis, acne, psoriasis, and skin aging, ¹⁰ , ¹¹ an association described as the gut‐skin axis. ¹² Gut microbes are key regulators of the gut‐skin axis, and the influence exerted by gut flora on the skin may stem from the gut's role in regulating systemic immunity. ¹³

The beneficial effects of GM on aging and skin health have been demonstrated in several rodent studies, with one study finding that transplantation of microbiota from long‐lived populations into mice transferred beneficial bacteria and reduced levels of lipofuscin and β‐galactosidase, which are biomarkers of aging. ¹⁴ Meanwhile, research by Gao et al. observed a significant decrease in the diversity and abundance of GM, ¹⁵ an up‐regulation of transepidermal water loss, and a down‐regulation of the levels of skin barrier‐related proteins, such as SPT1, ceramide, and Loricrin, Keratin 10, and Desmoglein1, in aging mice, and the abnormal alteration of the structure and composition of the skin barrier is considered to be a hallmark of aging of the skin. ¹⁶ When probiotic supplementation was administered, gut flora disorders, impaired ceramide synthesis, and skin barrier function impairment improved. This suggests that intervening in the gut flora may be strongly associated with delayed skin aging. Regulating the GM composition may be a feasible new idea for slowing down facial aging. However, there is not yet strong evidence in existing studies to confirm a causal relationship between facial skin aging and more gut microbes, and further exploration of the association between facial skin aging and GM will provide new therapeutic strategies for studying delayed facial aging.

Mendelian randomization (MR) reveals the relationship between various phenotypes and diseases by examining exposure‐related variables to assess causal relationships that may be associated with outcomes. To date, large‐scale Genome‐Wide Association Study(GWAS) of the GM has been applied to a variety of skin diseases, for example, psoriasis, ¹⁷ lichen planus, ¹⁸ and urticaria. ¹⁹ MR analysis can help researchers to determine the causal relationship between GM and these skin diseases, and to explore the potential impact of GM on skin diseases. This inspired us, to further understand the association between GM and facial aging, in this study we assessed the relationship between GM and facial skin aging through a comprehensive two‐sample MR analysis, looking for signature gut flora that cause or ameliorate facial skin aging.

2. MATERIALS AND METHODS

2.1. Study design

In this study, we used two‐sample MR analysis to reveal causal relationship between GM taxa and facial skin aging. Three assumptions should be based on causal inference in MR studies ²⁰ : (1) Genetic variation is strongly correlated with exposure (GM); (2) Genetic variation and exposure outcomes are independent of potential confounders; and (3) Genetic variation affects outcomes only through exposure. The fundamental assumptions and research design of MR are shown in Figure 1.

FIGURE 1.

Schematic diagram of MR analysis proc.

2.2. Exposure and outcome data sources

We obtained genome‐wide genotype and fecal microbiome data from a large‐scale multi‐ethnic GWAS meta‐analysis of European individuals, ²¹ totaling 211 GM taxa. GM were classified by phylum, class, order, family and genus, and a total of 196 GM were included in the study (nine phyla, 16 classes, 20 orders, 32 families, 119 genera) after excluding 15 unknown taxa.

The data source for the facial skin aging outcome was derived from the UK Biobank GWAS dataset, ²² which included 423,992 adult participants. Sample inclusion/exclusion criteria were as follows:(1) Only European‐origin participants were included, (2) Each participant was required to complete a baseline assessment of questionnaires, body measurements, and donor samples, and (3) Those who did not complete or were unwilling to answer the questions were excluded from the analysis. Genome‐wide analysis to control for potential confounders using the linear mixed model approach implemented in BOLT‐LMM.

2.3. Ethical statement

The analyzed data used in this study were all published GWAS abstract data, and each study included institutional review board approval and written informed consent from participants. Therefore the analysis did not require ethical approval or patient consent.

2.4. Instrumental variables

Significant single nucleotide polymorphisms (SNPs) were selected for further study as instrumental variances (IVs) of the GM, and for the selection of IVs, we followed the following requirements: (1) Filtering for significant SNPs: we set a threshold of p < 5 × 10⁻⁸ to filter out SNPs with significant associations, (2) Ensuring SNP independence: we set the linkage disequilibrium (LD) coefficients as the criterion parameter (threshold set to r ² = 0.001 and window size to 10,000 kb), and (3) Testing the association hypothesis: we used the F‐statistic (F = beta2/se2) to assess the strength of the IV to determine if there was a weak instrumental bias in the results of the MR analysis, ²³ and we considered an F‐statistic greater than 10 to indicate the presence of a strong association (Table S1).

2.5. Statistical analysis

In this study, inverse‐variance weighting (IVW), weighted median(MW), MR‐Egger regression and weighted mode were used to assess the causal effect of gut microbial composition and risk of facial skin aging. We used the IVW method as the primary method of analysis, a method that provides accurate estimates if the assumption that all included SNPs can be used as valid IVs is met. ²⁴ Based on the InSIDE assumption, the MR‐Egger regression provides a consistent estimate of the causal effect; however, the method is less accurate. ²⁵ , ²⁶ WM can combine data from multiple genetic variants into a single causal estimate, providing consistent estimates with higher accuracy than MR‐Egger regression even if 50% of IVs are invalid. ²⁷

In addition, to rule out potential limitations of MR analysis, for the associations of p‐values < 0.05 in the IVW analysis of the individual GM and facial skin aging, we used sensitivity analysis to validate the robustness and reliability of the results by assessing the heterogeneity and pleiotropy of the data. We used IVW and MR‐Egger tests to assess heterogeneity, which was quantified by the Cochran's Q statistic, with heterogeneity considered significant if the p‐value was less than 0.05. ²⁸ The intercept term of the MR‐Egger regression was used to determine the presence of pleiotropy, which if present in the IVs, that is, a p‐value of less than 0.05, would lead to the assumption of exclusion and independence of the data not being valid and the results may not be robust, and the MR‐PRESSO was also used to detect and correct for horizontal pleiotropy outliers in the MR test. ²⁹ Additionally, we performed leave‐one‐out analysis to determine whether causal estimates were driven by any single SNP.

This was an exploratory study, so multiple testing correction was not performed. All statistical analysis were performed using the “Mendelian Randomization” and “Two sample MR” package in R package. Statistical significance was defined as p < 0.05.

3. RESULTS

3.1. Causal effects of gut microbiota on facial skin aging

Preliminary results of genetically predicted correlations between GM and the risk to facial skin aging are presented in Table S2. Out of the 196 GM taxa examined, a total of 12 taxa were found to have a causal association with facial skin aging. These included one phylum, two families, and nine genera. However, no causal association was found at class and order levels. Detailed data can be found in Figure 2.

FIGURE 2.

Forest plots show significant MR analysis results in the discovery samples. CI, confidence interval; OR, odds ratio; SNPN, the number of SNP.

Based on the results of the MR analysis, the IVW results initially identified 12 GM taxa that may have a potential causal relationship with facial skin aging, of which three taxa were negatively correlated with facial skin aging, suggesting a potential protective effect against facial skin aging. At the family level, higher levels of genetically predicted Victivallaceae were associated with a lower risk of facial skin aging (OR = 0.972, 95% CI = 0.950–0.993; p = 0.010); at the genus level, two genera were identified as potential protective factors: Eubacterium coprostanoligenes group (OR = 0.951, 95% CI = 0.908–0.997; p = 0.038) and Parasutterella (OR = 0.957, 95% CI = 0.924–0.990; p = 0.011); and causal assessments from WM and weighted model analyses supported consistent correlations. The estimates of MR‐Egger suggested that Parasutterella was negatively associated with facial skin aging (OR = 0.948, 95% CI = 0.857–1.048; p = 0.314). However, for the other two taxa, the MR‐Egger results were inconsistent with the results of other MR analysis methods (Victivallaceae (OR = 1.017, 95% CI: 0.920–1.123, p = 0.752), Eubacterium coprostanoligenes group (OR = 1.008. 95% CI: 0.834–1.218, p = 0.937). Because the wide implementation interval of this method leads to less precise results, we mainly used it to assess horizontal pleiotropy. ³⁰ The detailed statistical results are shown in Figures 2 and 3.

FIGURE 3.

Scatterplot of three taxa of GM negatively associated with facial skin aging:(A) Family Victivallaceae causally associated with facial skin aging. (B) Genus Eubacterium coprostanoligenes group causally associated with facial skin aging. (C)Genus Parasutterella causally associated with facial skin aging. Lines sloping from left to lower right represent negative associations of GM with facial skin aging, indicating a protective causal effect and horizontal and vertical lines show 95% confidence intervals for each association.

In addition, the results of IVW analyses for the remaining nine taxa suggested significant differences (p < 0.05), and the study suggests that these nine GM taxa increase the risk of facial skin aging. At the phylum level, genetic prediction of Verrucomicrobia (OR = 1.047, 95% CI = 1.003–1.093; p = 0.034) was associated with an increased risk of facial skin aging; at the family level, a positive causal effect was observed only for Lactobacillaceae (OR = 1.045, 95% CI = 1.008–1.083; p = 0.017); and at the genus level, the preliminary analyses suggested a potential hazardous role for seven genera: Lactobacillus (OR = 1.037, 95% CI = 1.002–1.073; p = 0.038), Parabacteroides (OR = 1.067, 95% CI = 1.003–1.135; p = 0.040), Eggerthella (OR = 1.029, 95% CI = 1.000–1.059; p = 0.049), Family XIII UCG001 (OR = 1.053, 95% CI = 1.003–1.105; p = 0.036), Ruminococcaceae UCG005 (OR = 1.053, 95% CI = 1.011–1.095; p = 0.012), Phascolarctobacterium (OR = 1.051, 95% CI = 1.006–1.098; p = 0.027), Butyricimonas (OR = 1.066, 95% CI = 1.013–1.122; p = 0.014). MR‐Egger, WM, and weighted mode analysis supported this result (Figures 2 and 4).

FIGURE 4.

Scatterplot of the nine enteric flora taxa positively associated with facial skin aging: (A) Genus Butyricimonas causally associated with facial skin aging. (B–I) other eight enteric flora taxa causally associated with facial skin aging. Lines moving diagonally upwards from left to right show that taxa are positively correlated with facial skin aging, representing a pathogenic causal relationship. Horizontal and vertical lines show 95% confidence intervals for each association.

3.2. Sensitivity analysis

To detect the existence of potential bias in the MR analysis, we conducted a sensitivity analysis test, and the results of the heterogeneity test and pleiotropy test are summarized in Table 1. The results of the MR‐Egger intercept test and the MR‐PRESSO global test showed that there was no horizontal pleiotropy (p > 0.05) for the three taxa that were negatively correlated with facial skin aging, and this conclusion was likewise supported by the leave‐one‐out method (Figure 5). The heterogeneity test did not show evidence of significant heterogeneity among the three taxa (p > 0.05). The results suggest that our MR analysis is reliable and robust and that it is likely that the above three taxa of gut bacteria are protective against facial skin aging.

TABLE 1.

Sensitivity analysis results.

GM taxa	Pleiotropy test				Heterogeneity test
	MR‐Egger			MR‐PRESSO global test	MR‐Egger		IVW
	Intercept	SE	p		Q	p	Q	p
Family Victivallaceae	−0.007	0.007	0.381	0.768	6.791	0.745	7.63	0.746
Genus Eubacterium coprostanoligenes group	−0.004	0.006	0.55	0.938	5.153	0.924	5.534	0.938
Genus Parasutterella	0.001	0.004	0.848	0.603	11.417	0.494	11.455	0.573
Phylum Verrucomicrobia	0.003	0.005	0.639	0.326	12.934	0.227	13.237	0.278
Family Lactobacillaceae	0.006	0.006	0.355	0.320	8.178	0.317	9.325	0.316
Genus Lactobacillus	0.002	0.006	0.728	0.320	10.247	0.248	10.414	0.318
Genus Parabacteroides	−0.011	0.009	0.319	0.864	0.386	0.984	1.676	0.892
Genus Eggerthella	0.001	0.008	0.869	0.518	9.535	0.389	9.566	0.479
Genus Family XIII UCG001	−0.01	0.006	0.143	0.631	2.078	0.912	4.92	0.67
Genus Ruminococcaceae UCG005	0.004	0.005	0.403	0.666	10.007	0.615	10.758	0.631
Genus Phascolarctobacterium	−0.004	0.009	0.664	0.897	3.396	0.846	3.602	0.891
Genus Butyricimonas	−0.002	0.008	0.775	0.028	23.605	0.015	23.789	0.022

Abbreviations: GM, gut microbiota; IVW, inverse‐variance weighted; SE, standard error.

FIGURE 5.

Leave‐one‐out sensitivity results: (A–C) Leave‐one‐out analysis of the causal effect of three protective factors on facial skin aging. (D–l) Leave‐one‐out analysis of the causal effect of nine risk factors on facial skin aging.

Sensitivity analysis for the nine risk factors showed heterogeneity in the Cochran's Q test for Butyricimonas (Table 1), and although the results of the MR‐Egger regression intercept analysis indicated no directed horizontal pleiotropy (Egger intercept = −0.002; p = 0.775), but MR‐PRESSO analysis suggested horizontal pleiotropy (p = 0.028) (Table 1). Therefore, Butyricimonas was considered not causally related to facial skin aging.

The heterogeneity test for all eight taxa except Butyricimonas showed no evidence of heterogeneity (p > 0.05), in addition none of the MR‐Egger regression intercepts deviated from the zero value, and there was no evidence suggesting potential horizontal pleiotropy (p > 0.05 for all intercepts), and none of the global tests of the MR‐PRESSO achieved statistical significance (p > 0.05), a finding also supported by leave‐one‐out sensitivity (Table 1 and Figure 5). All results suggest that the risk of facial skin aging from gut flora is likely to come from the above eight gut microbial taxa except Butyricimonas.

4. DISSCUSSION

This study is the first in our knowledge to use MR to investigate the causal relationship between GM and facial skin aging. Using the most widely cited current GWAS data on GM ³¹ , ³² and facial skin aging data, ³³ , ³⁴ we explored the potential causal relationship between 196 GM taxa and facial skin aging by two‐sample MR analysis. Meanwhile, we used sensitivity analysis and leave‐one‐out analysis to exclude the influence of confounding factors to ensure the robustness of the findings. Through MR analysis, we identified causal relationships between 11 GM taxa and facial skin aging, including one phylum, two families, and eight genera, and at class and order levels, no association was found between GM and facial skin aging. The results showed that family Victivallaceae, genus Eubacterium coprostanoligenes group, and genus Parasutterella were negatively associated with facial skin aging and may be potential protective factors, while phylum Verrucomicrobia, family Lactobacillaceae, genus Lactobacillus, genus Parabacteroides, genus Eggerthella, genus Family XIII UCG001, genus Ruminococcaceae UCG005, and genus Phascolarctobacterium, on the other hand, maybe risk factors for facial skin aging. These results are consistent with the results of the sensitivity analysis.

In our study, we identified three GM taxa that were negatively associated with facial skin aging. Previous studies have shown that the genus Parasutterella is a core microbial component of healthy feces in the human gastrointestinal tract ³⁵ and can influence host physiological function by modulating the abundance or function of commensal bacteria in the gut. ³⁶ However, poor knowledge exists about the functions of two microbiota, family Victivallaceae and genus Eubacterium coprostanoligenes group, which were reported for the first time in this study as protective factors against facial skin aging. Despite the potential protective effects observed in this study, more in‐depth studies are needed to explain their correlation with facial skin aging. On the other hand, we identified eight microbiota that may be associated with the risk of facial skin aging. However, the results of the present study differ from the findings of some established studies, which, it should be noted, include genus Lactobacillus. Lactobacillus belongs to the family Lactobacillaceae, and the presence of both Lactobacillaceae and Lactobacillus in the present study suggests a strong positive correlation with facial skin aging, implying that overgrowth of Lactobacillaceae may be a facial skin aging risk factors. However, we note that some established studies have found that Lactobacillus counts in the skin microbiota of older women are, on average, lower than those of younger women. Furthermore, clinical trials have verified that EV exosomes secreted by Lactobacillus plantarum inhibit wrinkle formation, increase skin density, and reduce hyperpigmentation, ³⁷ suggesting that this flora has a protective effect against skin aging. This is contrary to our findings, and to understand this contradiction, we reviewed current research on Lactobacillus. Although Lactobacillus is the dominant flora in some environments, an increase in Lactobacillus in the intestinal flora has been observed in studies of Graves' disease, ³⁸ autoimmune liver disease, ³⁹ and alopecia. ⁴⁰ In addition, a study from Hezaveh et al. found that Lactobacillus metabolizes dietary tryptophan to indole and increases AhR activity in tumor‐associated macrophages, ⁴¹ thereby suppressing tumor immunity. These findings suggest that Lactobacillus may have a role in promoting inflammation and causing damage to intestinal metabolic function. This contradictory result emphasizes the complex role of Lactobacillus, which may have different effects in different environments and disease states. Future studies need to further explore the relationship between Lactobacillus and facial skin aging and elucidate its specific mechanism of action. The only known species of phylum Verrucomicrobia, Akkermansia muciniphila, is virtually undetectable in aged mice, and it has been shown that diseases associated with natural aging can be ameliorated by gut microbial remodeling and supplementation with A.muciniphila. ⁴² In addition, another study found that gut health was protected in aging rats by dietary supplementation with ark clams and that the intestinal flora of genus Parabacteroides was significantly increased. ⁴³ However, in our study, we found that both the phylum Verrucomicrobia and genus Parabacteroides showed a positive correlation with facial skin aging. Although most of the flora found to be associated with facial skin aging in our study have not been reported and even somewhat contradicted with existing research data, further investigative studies are needed to validate these findings.

We also identified several GM associated with an increased risk of facial skin aging, including genus Family XIII UCG001, genus Ruminococcaceae UCG005, genus Eggerthella and genus Phascolarctobacterium. However, there have been no studies on the association between these flora and skin aging. Surprisingly, we found a potential link between some of the GM associated with the risk of facial skin aging and cognitive disorders such as Alzheimer's disease, which provides new insights into the mechanisms of the “gut‐brain‐skin axis.” The “gut‐brain‐skin axis” is an interconnected, complex system in which emotional states such as depression, anxiety, and worry can alter gastrointestinal function and microbiota composition, ultimately leading to local or systemic inflammation, including skin inflammation. ⁴⁴ For example, Eggerthella and Ruminococcaceae UCG005 have been reported to be associated with major depressive disorder, ⁴⁵ , ⁴⁶ while the abundance of Phascolarctobacterium gradually increases in patients with Alzheimer's disease. ⁴⁷ In addition, it has been found that the intestines of patients with advanced melanoma are enriched in Phascolarctobacterium, ⁴⁸ whose increased abundance can be considered as an inflammatory disease activity and predictor. ⁴⁹ Facial skin aging is a complex process that is influenced by multiple internal and external factors. The GM plays an important role in maintaining skin health, and the flora associated with facial skin aging may be linked to mechanisms of the gut‐brain‐skin axis. Notably, our findings imply that there may be a potential link between cognitive impairment and facial skin aging, nevertheless, the specific mechanisms underlying the deleterious effects of these gut flora require further investigation.

At the same time, there are some limitations of the present study. First, our study was based on GWAS data; however, human skin aging is caused by a combination of factors, and it is difficult for GWAS to explain all the genetic features of complex phenotypes. Although MR methods can eliminate the interference of confounding factors to a certain extent, they still have limitations. Second, our study is based only on European populations. European populations differ significantly from other populations in terms of genetics and environment. Therefore, conclusions drawn from MR studies based on European populations may not encompass genetic variation on a global scale and cannot be directly generalized to other populations. In addition, our study only addressed specific GM and facial skin aging, and most of the flora we found to be associated with facial skin aging have not yet been reported, and these findings still need to be validated by further investigations. Future studies should aim to gain a deeper understanding of the associations between these flora and facial skin aging and to assess the relationship between genetic traits and facial skin aging more comprehensively by integrating a wider range of population samples and providing evidence for more generalized studies to advance our understanding of the complex relationship between GM and skin aging.

5. CONCLUSIONS AND PERSPECTIVES

In summary, we assessed the causal relationship between GM and facial skin aging by two‐sample MR analysis. The results showed that three GM had positive protective effects and eight GM were risk factors for facial skin aging. This study provides new insights into the future of gut flora‐mediated and ameliorated facial skin aging, with a personalized treatment plan based on regulating the GM to achieve delayed facial skin aging.

CONFLICT OF INTEREST STATEMENT

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Supporting information

Supporting Information

Chen M, Che Y, Liu M, et al. Genetic insights into the gut microbiota and risk of facial skin aging: A Mendelian randomization study. Skin Res Technol. 2024;30:e13636. 10.1111/srt.13636

DATA AVAILABILITY STATEMENT

MiBioGen repository: https://mibiogen.gcc.rug.nl/; Facial skin aging Web source. https://doi.org/10.5523/bris.21crwsnj4xwjm2g4qi8chathha.