Introduction

The skin is considered a barrier organ against the entry of foreign physical, chemical, and biological insults, thereby maintaining the internal homeostasis of the human body. In the past decades, Human Microbiome Project (HMP) has expanded our perception of the skin as not only a piece of placid “soil” but a vast “ecosystem” that harbors a myriad of microbial inhabitants (Human Microbiome Project Consortium 2012). It has been believed that the colonization of diverse microbes resulted from millions of years of mutual adaptation and functional integration (Lousada et al. 2021), and thus the human body forms a complex, synergistic entity, termed a holobiont or meta-organism (Bosch and McFall-Ngai 2011; Rosenberg et al. 2007). The environmental and nutrient conditions define the unique microhabitats for skin microbes (Flowers and Grice 2020), and in turn, these microbes can influence their survival environment (host skin) by stabilizing, mutually beneficial host–microbe interactions (Postler and Ghosh 2017). In various disease conditions, the host–microbe interactions became imbalanced, termed “dysbiosis”, presenting various shifts in microbiome from “healthy” to “diseased” states (Thomas and Jobin 2020).

Profiling the structure of skin microbial community is the first step to overview the ecosystem and to address host–microbe interactions. However, this system was proven to be highly individualized and extremely complex. Many factors were identified influencing the composition of the system, including race, gender, age, lifestyle (e.g., occupation, hygiene, skin product and medication usage, and diet) and environment (e.g., climate, geographical location, pollution, UV, and other radiation) (Wei et al. 2022; Grice and Segre 2011; Harris-Tryon and Grice 2022). Nevertheless, from the perspectives of classical ecology, most of these factors may only indirectly influence, but not drive the establishment and maintenance of the system. The primary selection pressures that form the driving forces for the ecosystem, include resource availability (presence of nutrients), environmental conditions (temperature, geographical access) and biological factors (predators and pathogens) (Williams 1996). In this review, we will sum-up related studies centered on these essential selection pressures, including the presence of different types of nutrients and favored micro-environment for dominant skin commensals, the occupation of the ecological niches through self-adaptation or microbe–microbe interactions, and eventually we will discuss how skin microbes, by their structures or bioactive molecules, reshape host skin phenotypes (Fig. 1).

Fig. 1
figure 1

Skin microbiome, metabolome and skin phenome, from the perspective of skin as an ecosystem. From left to right: (Blue box) Diverse substances, derived from the host (stratum corneum, skin appendages, and plasma), environment (xenobiotics) and microbial metabolism, cover the skin surface, forming the micro-environment for skin microbiota; (Green box) occupation of ecological niches by self-adaptation and microbe–microbe interactions, promoting commensals or inhibiting pathogens; (Yellow box) the skin microbes, by their own structures or bioactive molecules, reshape the host skin phenotypes

Full size image

Micro-environment of the Human Skin Ecosystem

The host skin offers nutrients and shelters for microbial survival, competition, and cooperation (Roth and James 1988). Nutrient substances may directly affect microbial colonization, growth and metabolism either through nourishing (Brüggemann et al. 2004) or persecuting (Ferrer et al. 2017); on the other hand, these substances may also finetune the local microenvironment, such as pH or moisture state, and thus exert indirect impact on microbial survival. The microbial energy substances are mainly from the host skin and the outside environment. The host skin-derived nutrients consist of lipids embedded in the “brick and mortar” structure (Chen 2018), piles of dead enucleated corneocytes in the stratum corneum (SC) (Abhishek and Palamadai Krishnan 2016), and the secretions from skin appendages [hair follicles (HFs) and glands]. The environment-derived nutrients include personal skincare products, medication, and other environmental xenobiotics. Here, we summarized the metabolites detected on the skin by various metabolome studies (Table1).

Table 1 Human skin metabolites: their primary source and functions
Full size table

It is known that individuals, even the same individual at different life stages, vary markedly in regards to the delicate structure or secretion function of the skin and appendages, which produce metabolites consistently and thus play an essential role in shaping diverse microenvironments with distinct pH, salt, moisture, sebum content, and extent of anaerobiosis (Grice and Segre 2011; Capone et al. 2011; Grice et al. 2009). Factors that influence systemic metabolisms, such as diet and gut microbiota, and hormone levels, can also significantly impact the skin’s local microhabitats (Prescott et al. 2017). Furthermore, one’s exposome, such as environmental pollution, UV levels, occupation environment, drug or skincare habits, is highly individualized (Khmaladze et al. 2020). These together form highly complex physical and chemical landscapes on the skin surface, likely to be the real biological explanation that underlies the substantial inter-individual variability in the skin microbiota. Indeed, our previous study showed two robust “cutotypes” of microbial networks on Chinese facial skin, C-cutotype and M-cutotype, possessed distinct patterns of skin properties (Li et al. 2021). The dominant two species, C. acnes and Moraxella osloensis, exhibited vastly varied nutrient-demand: whereas C. acnes was high nutrient demanding, M. osloensis was a non-fastidious bacterium that was able to grow in a mineral medium supplemented with a single organic carbon source (Juni 1974; Juni and Bøvre 2015). This species was shown to be incapable of utilizing any carbohydrates or possessing any saccharolytic activity, but strictly depend on other carbon sources such as acetic or lactic acid (Baumann et al. 1968; Juni 1974; Juni and Bøvre 2015; Moss et al. 1988).

Occupation of Ecological Niches by Self-adaptation and Microbe–Microbe Interactions

The skin surface formed diverse microhabitats, and many studies favored to divide them into four types (sebaceous, moist, dry, and foot) according to the physical properties of anatomical locations (Oh et al. 2014). Although such water/oil-based classification was not delicate enough, some prominent features for the growth and colonization of the microbiota were well identified. Other metabolites and physical properties were also identified in modulating microbial communities. Furthermore, microbe–microbe interactions are essential for shaping the skin ecosystem. In general, microbes deploy strategies to adapt to the living environment and compete for ecological niches via the following: (1) Self-adaptation to the specific environment conditions: skin microbiota changes their characteristic like metabolism pathways to adapt to the skin microenvironment. For example, Staphylococcus synthesized tensioactive agent to withstand the low pH and high salt content of sweat (Hentati et al. 2021; Scharschmidt and Fischbach 2013); (2) Competition for ecological niches through microbe–microbe interactions, for example, coagulase-negative Staphylococcus (CoNS) species can either directly kill or limit the virulence of Staphylococcus aureus through the secretion of different regulators (Flowers and Grice 2020). Here we will sum-up the findings of this part (Table 2).

Table 2 Features of dominant skin commensals for the occupation of ecological niches
Full size table

Compared to the skin surface, HFs provide a more moisture and acidic environment with ultraviolet light protection, facilitating the colonization of multiple bacteria, fungi, and viruses. The most abundant bacteria in the HFs were P. acnes spp. (Lousada et al. 2021). M. restricta and M. globosa are the dominant fungi (Lousada et al. 2021). Meanwhile, the HF virome comprises dependoviruses, Propionibacterium phage P100D and 101A, papillomaviruses and adeno-associated viruses (Hall et al. 2018). In addition, the mite (Demodex folliculorum) groups are often found in the distal infundibulum, usually with their dorsal body oriented against the hair shaft (Elston and Elston 2014).

From Microbes to Host Skin: How Microbes Reshape the Skin Phenome

Skin microbiota leverage “nutrients” from the host skin and environment and produce a series of bioactive molecules with vital functions (Chen et al. 2018). For example, skin microbiota can convert host proteins into amino acids by their protease (Holland et al. 1979; Byrd et al. 2018), ferment carbohydrates into lactic acids (Ong et al. 2020) or decompose sebum lipids such as triglycerides into free fatty acids (FFAs) (Traisaeng et al. 2019; Belkaid and Segre 2014). In addition, skin microbiota produces AMPs, phenol-soluble modulins (PSMs), and antibiotics (Belkaid and Segre 2014; Gallo and Hooper 2012). These metabolism products may further act on the host or other microbes, exert biological effects and reshape the skin phenome.

The most well-studied functions of skin commensals include the following: (1) pathogen colonization resistance by ecological niche blocking for the invasion of opportunistic or pathogenic microbiota, (2) immune education during early phases, and (3) regulation of immunity and inflammation. Given many comprehensive reviews already on these functions, we will take a particular focus on other functions that were usually missed, including the maintenance of skin physiology, such as pH and SC hydration, UV protection, odor production, and wound healing, which were also important functions in skin homeostasis.

Regulation of Immunity and Inflammation

The microbiota is a rich source of short-chain fatty acids (SCFAs) (Traisaeng et al. 2019). For example, C. acnes fermented carbohydrates into propionic acid (Traisaeng et al. 2019); S. epidermidis was able to ferment glycerol to butyric acid and acetic acid in vitro (Traisaeng et al. 2019; Keshari et al. 2019). SCFAs can regulate several immune cell functions, including the production of cytokines (TNF-α, IL-2, IL-6, and IL-10) (Traisaeng et al. 2019), activate resident skin regulatory T (Treg) cells, mitigate inflammatory skin reactions and thus contribute to the preservation of skin homeostasis in mice and human (Schwarz et al. 2017). Butyric acid significantly attenuated lipopolysaccharide (LPS)-induced nuclear factor-κB (NF-κB) activation and nitric oxide production in murine macrophage cell line (Chakravortty et al. 2000), reduced interferon-gamma (IFNγ)-induced proinflammatory IL-6 and TNF-α production in a macrophage cell line (Park et al. 2007) and mediated short-chain fatty acid receptor 2 (FFAR2) to modulate the production of proinflammatory cytokines induced by ultraviolet B (UVB) in mice (Keshari et al. 2019). Furthermore, the ability of immune cells to migrate to the foci of infection can be regulated by SCFAs (Vinolo et al. 2011). Given the potential anti-inflammatory of SCFAs, they are applied on psoriatic skin in vitro. This study found that decreased expression of G-protein-coupled receptors (GPR) GPR43 and GPR109a in psoriatic skin can be restored and expression of inflammatory factors can be inhibited by topical application of sodium butyrate (Krejner et al. 2018). However, SCFAs are not always anti-inflammatory. C. acnes-derived SCFAs inhibit histone deacetylase (HDAC) activity in skin keratinocytes (KCs) and stimulate inflammation through Toll-like receptor (TLR) signaling (Sanford et al. 2016). SCFAs from C. acnes conferred a robust proinflammatory effect in human sebocytes (Sanford et al. 2019). Expression of a major component of the Corynebacterium accolens cell wall, mycolic acid, promotes inflammation in an IL-23-dependent manner under a high-fat diet condition in mice (Ridaura et al. 2018).

The essential amino acid tryptophan (Trp) can be metabolized by human skin microbiota into 5-hydroxytryptophan (5-HTP), indole-3-aldehyde (IAId) and other metabolites (Yu et al. 2019). IAId was able to suppress thymic stromal lymphopoietin (TSLP) and thereby inhibited calcipotriol (MC903)-induced AD-like dermatitis in mice (Yu et al. 2019). IAId can also activate aryl hydrocarbon receptor (AhR), producing indoleamine 2,3-dioxygenase (IDO) and IL-10 in Langerhans cells (LCs), and thus negatively regulate skin inflammation (Liu et al. 2020).

S. epidermidis and other Gram-positive bacteria release adhesion molecules upon bacteriolysis, such as lipoteichoic acid (LTA) (Ginsburg 2002). LTA from Staphylococcal species suppressed inflammation during tissue injury through a Toll-like receptor 2 (TLR2)-dependent mechanism to prevent excessive damage (Lai et al. 2009). Staphylococcal LTA may also have applications in the treatment of inflammatory disease. For example, in an acne model of C. acnes-induced skin inflammation, staphylococcal LTA application abrogated inflammatory effects via induction of a microRNA, miR-143, destabilizes the TLR2 mRNA and decreases protein production (Xia et al. 2016).

In addition, many commensal species contain virulence strains. One major virulence factor of the microorganism is a secretory lipase that acts on triglycerides to release FFAs (Holland et al. 2010). C. acnes exist both in health and patients, but C. acnes from acne patients harbored unique genomic elements encoding virulence factors, including camp5, gehA, sialidases, neuraminidases, endoglicoceraminidases, lipases, proteases and hemolysins that were rarely present in C. acnes genomes from healthy controls (Brüggemann 2005; Burkhart et al. 1999). Several commensals are opportunistic pathogens that encode virulence factors such as toxins, exoenzymes, and adhesins (Brown et al. 2012). Skin microbiota may directly or indirectly mediate inflammatory responses by releasing various virulence factors under unhealthy conditions. Malassezia spp. can be the causative agents in disease. Many Malassezia spp. secrete extracellular vesicles that signal KCs to secrete proinflammatory cytokines (Vallhov et al. 2020; Watanabe et al. 2001; Zhang et al. 2019). Malassezia spp. metabolize sebum to different fatty acids such as phosphatidylcholine (PC) and phosphatidylserine (PS), which then act as irritants, causing flaking and irritation under dandruff, a frequent scalp issue and seborrheic dermatitis conditions (Celis Ramírez et al. 2020; DeAngelis et al. 2005; Han et al. 2019; Johansson et al. 2018).

Pathogen Colonization Resistance

Commensals compete for niches through microbe–microbe interactions, as mentioned above (Table 2). Direct induction of AMPs or cytokine expression in KCs is one of the main strategies used by skin commensals, such as Propionibacterium and S. epidermidis, in defending against pathogen invasion and shaping the skin microbiota community (Midorikawa et al. 2003; Wanke et al. 2011). In addition, commensals function as endogenous cofactors of the skin immune system to promote skin local immune response. Skin harbor considerable commensal-specific T-cell, e.g., Staphylococcus epidermidis-specific IL-17A+ CD8+ T cells (Naik et al. 2015). The activation of these cells can promote AMP production by keratinocytes, thereby promoting heterologous protection against pathogens infections (Braff et al. 2005). Staphylococcus epidermidis can also induce KC to express IL-1α, thus promoting skin αβ T cells to produce IL-17A and IFNγ in mice (Naik et al. 2012). IL-17A induces chemokines that recruit neutrophils and AMP production, thus protecting the host from pathogen infection. In adults, cutaneous mucosal-associated invariant T cells (MAIT cells) are a dominant population of IL-17A-producing lymphocytes (Constantinides et al. 2019). MAIT cells are absent in germ-free (GF) mice, and their development are controlled by microbial metabolites such as vitamin B2 (Treiner et al. 2003; Koay et al. 2016; Legoux et al. 2019). MAIT cells can respond to skin commensals or commensal-derived metabolites in an IL-1-, IL-18-, and antigen-dependent manner (Constantinides et al. 2019), thus enhancing inhibition of pathogen invasion.

Immune Education

The commensals play an essential role in regulating the development, proliferation, maturation and activation of immune cells of innate immunity. A previous study found that GF mice contain mast cells (MCs) that are largely undifferentiated and express abnormally low amounts of stem cell factor (SCF). Commensal bacteria induce KC-produced SCF, promote skin MCs mature. The migration of MCs in the skin is fully dependent on high levels of SCF, as produced by KCs (Wang et al. 2017b). In addition, γδT cells, which play an essential role in recognizing lipids, one of the microbial metabolites (Belkaid and Tamoutounour 2016), significantly reduced IL-17 secretion capacity in GF mice (Naik et al. 2012). Varying from the immune responses to invasive pathogens, adaptive immune responses respond to commensals under noninflammatory conditions, which help build immune homeostasis (Naik et al. 2015).

The skin contains one of the highest frequencies of FOXP3+ Treg cells within the body in mice (Suffia et al. 2006). In the skin of both mice and humans, Tregs reside in the dermis, and a large fraction of these cells can be found in close proximity to HFs, which serve as a natural habitat for skin-resident microorganisms (Ali et al. 2017; Sanchez Rodriguez et al. 2014). Tregs are essential in establishing and regulating immune tolerance to commensal microbes during a defined period of neonatal life in mice (Scharschmidt et al. 2015). S. epidermidis colonization on the skin surface two weeks after birth induces Treg cells’ tolerance to S. epidermidis in adult mice (Scharschmidt et al. 2015). Furthermore, it promotes the accumulation and migration of Treg cells into the skin (Scharschmidt et al. 2017). Further study found that Treg cell migration in Neonatal Skin is influenced by hair follicle development and microbes colonized in the hair follicle. In turn, colonization of microbes in HFs during the early stage is resisted and regulated by Treg cells (Scharschmidt et al. 2017). These results suggest a dynamic balance between microbe and host immune system.

Maintain pH and SC Hydration

Skin microbiota metabolizes dead corneocytes, sweat and sebum components, and other wastes (Pistone et al. 2021) and converts them into amino acids, such as glutamate and aspartate, proteins and various FFAs (Pistone et al. 2021; Timm et al. 2020). They also secrete lactic acid (Ong et al. 2020), a series of SCFAs (Christensen and Brüggemann 2014) and other organic acids (Garrote et al. 2000; Wang et al. 2017a; Bengoa et al. 2019). These acidic metabolites can regulate skin surface pH and SC hydration level (Watabe et al. 2013; McGrath 2008; Caspers et al. 2001; Cui et al. 2016; Pappas 2009).

The skin surface pH is slightly acidic, ranging from 4.5 to 5.5 in human (Braun-Falco and Korting 1986). The pH of the SC is crucial for many vital epidermal functions, including permeability barrier homeostasis, desquamation of corneocytes, initiation of inflammation, processing of secreted lamellar body (LB) polar lipids and antimicrobial defense (Lee and Lee 2014). In addition, variation in pH also affects the SC thickness and pigmentation (Sandby-Møller et al. 2003). These results indicate that many skin traits may intertwine, such as pH, trans-epidermal water loss (TEWL), skin thickness, SC hydration and pigmentation, and thereby may be modulated by skin microbiota and their metabolites.

Our previous study also revealed that cutotypes of microbial networks on Chinese facial skin possess distinct skin traits: C-cutotype skin is more hydrated and more oily, and the levels of skin surface sebum and its microbial metabolite porphyrin are increased; In contrast, M-cutotype skin is dryer and often occurs in the elder (Li et al. 2021). A study on the skin microbiome of Koreans found that Lawsonella had a negative correlation with skin moisture and brown spots; Staphylococcus and Corynebacterium both had negative correlations with the number of UV spots and positive correlations with TEWL; Staphylococcus aureus had a negative correlation with skin moisture parameters (Kim et al. 2021a). Moreover, two studies found a linkage between the skin microbiome and skin metabolites (Howard et al. 2022; Roux et al. 2022). A recent study demonstrated that S. epidermidis can significantly increase skin ceramide levels and thereby prevent water loss of damaged skin dependent on its sphingomyelinase in mice (Zheng et al. 2022).

Skin aging is a dynamic process with a series of changes in the skin phenome (Farage et al. 2008; Pochi et al. 1979; Cotterill et al. 1972; Howard et al. 2022) and skin metabolism, e.g., altered levels of natural moisturizing factors (NMFs), AMPs, vitamins and coenzyme Q10, and many other metabolites (Howard et al. 2022; MacLaughlin and Holick 1985; Kuehne et al. 2017). These changes may underlie the alterations in the microbiome. For example, age-related decrease in sebocyte area is positively correlated with Cutibacterium and negatively correlated with Streptococcus, Acinetobacter, Corynebacterium and Methylobacterium‒Methylorubrum abundance (Howard et al. 2022). Furthermore, anti-aging skincare products were reported able to persist on the skin for weeks and provide long-term contributions to the chemical environment (Bouslimani et al. 2019), thus shaping the specific skin microbial communities (Bouslimani et al. 2015). For example, lipid components of moisturizers could provide nutrients and promote the growth of lipophilic bacteria such as Staphylococcus and Propionibacterium (Bouslimani et al. 2015; Unno et al. 2017; Holland et al. 2010). More details regarding cosmetics can be found in Table 1.

UV Protection

Some skin commensals can protect skin from UV damage by secreting different metabolites (Souak et al. 2021). For example, S. epidermidis can produce 6-HAP to suppress UV-induced tumor in mice (Nakatsuji et al. 2018). Skin microflora produces cis-urocanic acid from l-histidine, affects UV-induced immune suppression and suppresses melanoma growth (Hug et al. 1999; Laihia et al. 2010). Some Streptomyces-derived compounds, such as amides exhibited UV-absorbing, antioxidant, and anti-inflammatory properties (Sánchez-Suárez et al. 2020). Propionic acid produced by Cutibacterium acnes fermentation ameliorates UVB-induced melanin synthesis (Kao et al. 2021). Cyanobacteria develop a diversity of defense mechanisms, including the biosynthesis of UV-absorbing/screening compounds, such as mycosporine-like amino acids (MAAs), and enzymes, including superoxide dismutases (SOD), which counteract oxidative stress (Souak et al. 2021).

Ultraviolet radiation (UV-R) is well known to inhibit the cellular growth of Malassezia furfur (Wikler et al. 1990). On the other hand, Malassezia furfur can produce pityriacitrin, a UV-filtering compound believed to be protective (Machowinski et al. 2006). It is hypothesized that this fungus developed the UV-filter compound to reduce UV damage and compete for survival over other commensals (Machowinski et al. 2006). However, they did not find any adverse effect of pityriacitrin on commensals such as S. aureus, S. epidermidis, or Candida albicans (Machowinski et al. 2006).

Odor Production

The metabolic activities of some skin microbes produce special odors. For example, human body odor is believed to result from bacterial growth and decomposition of secretions from specialized glands in the axillary region (Lam et al. 2018; Decréau et al. 2003; Natsch et al. 2003). Microbes are present in specific scent glands or tissue in mammals and modulate specific odors (Ezenwa et al. 2012). Skin microbes metabolize host sweat and produce volatile metabolites, enhancing the attractiveness of human sweat for the malaria mosquito (Brouwer 1960; Takken and Kline 1989). A recent study specified acetophenone, a volatile from the skin microbiota, promoted mosquito attractiveness in flavivirus-infected hosts (Zhang et al. 2022).

Skin commensal Moraxella osloensis (Li et al. 2021), a species highly tolerant to desiccation and UV irradiation, existed in various living environments, particularly in the laundry. This species has the potential to generate 4-methyl-3-hexenoic acid (4M3H), which is often described as a “wet-and-dirty-dustcloth-like malodor” or an “acidic or sweaty odor” (Kubota et al. 2012). In addition to bacteria, fungi are important sources of many volatile organic compounds (VOCs), including alcohols, aldehydes, esters, FAs, and terpenes (Belinato et al. 2019). In malignant fungating wounds (MFWs), metabolites such as dimethyl trisulfide (DMTS), four fatty acid volatiles (acetic acid, isobutyric acid, butyric acid, and isovaleric acid) and putrescine are linked with components of malignant fungating wound odor (Vardhan et al. 2019).

Wound Healing

Wound healing is a complex but highly regulated process critical for skin barrier function (Han and Ceilley 2017). The presence and abundance of microbes in skin wounds depend on wound type (chronic/acute wound) (Johnson et al. 2018) and shifts over time (Loesche et al. 2017). Studies demonstrated that skin microbiota was also involved in wound healing in multifaceted ways. S. epidermidis promotes rapid KC progression via upregulation of TLR and downstream modulation of TNF-α in skin CD8+ T cells (Linehan et al. 2018; Naik et al. 2015). A study with a wound-induced hair follicle neogenesis (WIHN) mouse model revealed that skin microbiota promoted skin regeneration via IL-1β and KC-dependent IL-1R-MyD88 signaling (Wang et al. 2021). Metabolites from microbiota promote wound healing, e.g., lipoteichoic acid from S. epidermidis can decrease inflammation via TLR2 signaling (Lai et al. 2009). On the other hand, some potential pathogens do not promote cutaneous wound healing. For example, S. aureus (Kirker et al. 2009; den Reijer et al. 2016), Acinetobacter. baumanni and A. junii (de Breij et al. 2012) form biofilms on the SC and have a detrimental effect on human dermal fibroblast migration and ultimately result in cellular apoptosis (Kirker et al. 2012). Microbial stability was believed to be essential for skin health; however, temporal stability in the chronic wound is associated with poor healing as instability in the microbiome reflects effective control of wound bacteria, which prevents any community structure from stabilizing (Loesche et al. 2017).

Conclusion

The present review centers on the current knowledge on skin microbiome from a perspective of skin as an ecosystem and tries to explore the fundamental driving force for the establishment and the balance of the highly personalized microbial feature. We believe that microenvironments that define the physical (e.g., pH, oxygen) and chemical (carbon sources and metabolites) conditions drive the microbiome composition. In turn, these microbes may reshape this environment via microbe–microbe or microbe–host interactions. Skin surface metabolome may be a critical approach to address causative correlations between the skin microbiome and skin phenome; therefore, future skin microbiome research should leverage those multi-omics to reveal these strong correlations and then validate them with the principle of Koch’s postulates. Furthermore, considering the higher complexity of the system due to the host genome and exposome, the longitudinal time-series study should be taken more into consideration for the control of these variables and for addressing the direction of those networks. Based on solid causative correlations, we can develop accurate interventions targeting specific skin microbe(s) and eventually reshape the skin conditions.

Of note, recent studies revealed that microbiota at strain level varies in the local microenvironment (Conwill et al. 2022), suggesting studies on higher resolution should be emphasized, which means deeper sequencing until strain level and more refined sampling sites up to single pore level. However, the greatest challenge for these designs is biomass, including metabolites and metagenomic biomass. This strongly relies on the technology development and iterative update of detection instruments to improve the sensitivity.

The significance of the human skin microbiome is increasingly appreciated. The approach from metagenomic sequencing (profiling) was gradually shifted to isolation/culturomics and function validation (mechanisms). However, some significant issues still exist, such as the lack of ideal ex-vivo skin models (e.g., reconstructed human epidermis (RHEs) and skin explants) that can reliably simulate the complexity of the host–microbe interactions (Harris-Tryon and Grice 2022; Larson et al. 2021). Some recent studies performed the function experiments with three-dimensional (3D) human skin equivalent. For example, a study using 3D skin tissue cultures revealed that a model microbiome or a mixed community of skin microbiome representatives led to pronounced changes in epidermal thickness, epidermal cell proliferation, and filaggrin production (Loomis et al. 2021). Another study investigated the interaction between the skin microbiota and environmental pollutant benzo[a]pyrene (B[a]P), with a microbially competent 3D skin model and demonstrated that commensal metabolism of xenobiotics can influence host toxicity (Lemoine et al. 2021). However, the limitations of these ex-vivo skin models are apparent, i.e., the lack of the histological/physiological/immunological complexity of RHEs, the paucity of inter-donor variability of skin explants, as well as short lifespan and the relatively high costs (Larson et al. 2021). Nevertheless, this is a matter of time to address these issues and push forward the skin microbiota targeted new intervention based on solid experimental evidence.