1. Genetic Risk Factors and Epigenetic Regulators
A family history with inherited traits is the strongest risk factor for AD, with gene mapping studies accounting for 20% of the estimated heritability
[1]. The term ‘heritability’ refers to the proportion of variation within a clinical feature that is attributable to genetic factors. Single null mutations in the gene encoding for the essential epidermal structural protein filaggrin (FLG) are currently the strongest genetic risk factor for AD, afflicting 10% of individuals of European ancestry, although 50% of those do not develop the disease
[2]. In addition to the European mutations (R501X, 2282del4, R2447X, S3247X), copy number variations (intragenic repetitive FLG gene sequences) and the amount of filaggrin monomer expressed are additional risk factors for AD. Other candidate genes such as filaggrin-2 (FLG2) and those involved in the Th2 immune response (chromosomes 16p12.1–p11.2 (IL-4 receptor) and 5q31.1 (IL-13 receptor)) are thought to pose a significant risk as well
[2]. FLG and FLG2, along with loricrin (LOR) and involucrin (IVL), are located on chromosome 1q21.3
[3][4][5][6]. Together with the S100A7 gene series, they constitute the bulk of the epidermal differentiation complex (EDC)
[7]. Expression of the complex is internally modulated by cytokines (IL-4,13,17A,22) derived from T-helper cells (Th1,2,17,22). On the other hand, external modulation can occur via microorganisms (
Malassezia spp. and
Staphylococcus epidermidis) and numerous phytochemicals via the activation of the aryl hydrocarbon receptor (AHR). Prolonged AHR activation can result in accelerated keratinocytic differentiation, loss of sebocytes, and keratinous cyst formation
[8]. Another modulator of keratinocyte differentiation is intracellular reactive oxygen species (ROS). ROS levels are kept in check by the oxidative stress-prone AHR and the antioxidative nuclear factor E2-related factor 2 (NRF2) systems.
Keratinocytic differentiation is key to stratum corneum formation and epidermal
[9][10] stratification. Their differentiation and subsequent development are regulated by the EDC genetic complex, signaling pathways, transcription factors, and epigenetic modifications. Epigenetic modifications can vary with exposure to acute and chronic environmental stressors leading to changes in genetic expression throughout all stages of keratinocytic development. These modifications include changes to the methylation status of DNA (methylation and hydroxymethylation), covalent additions or reductions to histone components (methyl-or acetyl groups), expression of noncoding RNAs (ncRNAs), nucleosome positioning, and ATP-dependent chromatin remodeling. Such changes are conducted via DNA methyltransferase (DNMT)1; histone deacetylases (HDAC)1/2; polycomb proteins (PcGs) Bmi1 and Ezh1/2; ATP remodelers Brg1 and Mi-2β of the chromodomain helicase DNA-binding (CHD) family; and histone demethylases (HDACs) Jmjd3 and Setd8, which together with several other transcription factors (in particular, p63) stimulate the proliferation and maturation of progenitor cells.
Of the many proteins, enzymes, and micro ribonucleic acids (miRNAs) involved in modulating genetic expression, the de/methylases and miRNA are the most prevalent. In a recently published
[11] epigenome-wide association adult study, significant differences in methylation at 19 CpG sites and a partial correlation in altered gene transcript levels between the epidermis of healthy controls and epidermal lesions in AD patients were reported. Additional epigenetic studies have shown that high exposure to smoke can lead to hypomethylation of the thymic stromal lymphopoietin (TSLP) 5′CpG island in pregnant women and is positively correlated with AD
[12]. Furthermore, tobacco smoke is thought to modify the Forkhead box-p3 (FOXP3) locus in umbilical cord blood and is associated with low Treg counts at birth. In addition to FOXP3, low Treg counts correlate with increased miR-223 expression. Other mRNAs that play a critical role in AD include (i) miR-124, which is downregulated in the lesional skin of patients with atopic eczema
[13]; (ii) miR-143, a tumor suppressor that targets the IL-13Rα1 receptor reducing IL13 levels and inflammation in epidermal keratinocytes
[14]; (iii) miR-146, reported to inhibit the expression of several proinflammatory factors, including IFN-γ and AD-associated genes CCL5, CCL8, and ubiquitin D (UBD)
[15]; (iv) miR-155, which is positively correlated with AD severity and essential for the differentiation of Treg and T helper 7 (Th17) cells, and was recently shown to target protein kinase inhibitor α (PKIα) leading to its downregulation along with TSLP in a murine model
[16]. TSLP promoter hypomethylation has been observed in keratinocytes isolated from AD patients resulting in the overexpression of TSLP in skin lesions
[17]
The various epigenetic factors contributing to the polarized expression of EDC genes, interleukins, and cytokines can result in numerous functional deficiencies (decreased water retention, altered lipid formation, pH imbalance, ceramides, and sphingosine) involving proteins filaggrin, loricrin (LOR), involucrin (IVL), and desmosomes K1, K10, (envoplakin and periplakin) impairing corneocyte integrity
[18][19]. Treatments have sought to address these deficiencies by either targeting inflammatory cytokines (monoclonal antibodies) or utilizing antioxidative AHR ligands and nuclear factor E2-related factor 2 (NRF2) activators (tapinarof, coal and soybean tar, and glyteer) to dampen reactive oxygen species (ROS)
[20].
Finally, the epigenetic role of Staphylococcal enterotoxin B (SEB) is worth mentioning. In addition to the pore-forming toxins of
Helicobacter pylori, Listeria, and
Salmonella that repress a subset of immune genes (CXCL2, MKP2, and IFIT3)
[21], SEB is adept at altering the methylation pattern of two gene regions (IKBKB and STAT-5B) both of which play a crucial role in T- cell maturation/activation
[22].
1.1. Factors That Influence Staphylococcus Aureus Colonization of AD Skin
The reduced colonization resistance of AD skin, coupled with barrier deficiencies and a polarized Th2 immune response, renders the skin susceptible to SA biofilm formation and infiltration. The first stage in colonization is the covalent attachment of SA (sortase rich) to the skin, involving the recruitment of fibronectin, corneocytes, loricrin, and cytokeratin 10
[23]. Other factors that enhance colonization
[24] and the potential for severe AD include IL-13 and IL-4, which down-regulate the expression of antimicrobial peptides (AMPs) such as human beta-defensin (HBD)-3 and cathelicidin (LL-37)
[25][26]. Conversely, AMP expression is upregulated in the presence of coagulase-negative staphylococci (CoNS),
S. hominis and
S. epidermidis, which secrete phenol-soluble modulins (PSMs) γ and δ further amplifying AMP potency, thereby restricting in vitro growth of group A
Streptococcus and
S. aureus [27]. In addition to AMPs, other factors that aid colonization resistance include numerous acids (urocanic and pyrrolidone carboxylic) and lipids (including free fatty acid, ceramides, and sphingosine), filaggrin, and filaggrin degradation products (FDP)
[28][29].
1.2. S. aureus Infection, Impairment of the Epidermal Barrier, and Potential Remodeling
When challenged, SA can secrete a battery of soluble receptor-mediated (α-toxin) and non-receptor-mediated (PSMs, δ-hemolysin) toxins, as well as enterotoxins A(SEA), SEB, and C(SEC), and a 22-kD superantigen, synonymous with toxic shock syndrome (TSS)
[28][29][30][31]. In addition, the potency of said toxins is enhanced when delivered via extracellular vesicles (EV), which have been shown to play multiple roles in the development of AD and the activation of inflammatory macrophages. For example, mice subjected to repeated intranasal EV exposure induced a Th1 and Th17 cell response via Toll-like receptor 2 (TL2), resulting in neutrophilic pulmonary inflammation
[32][33][34][35][36]. Many stressors (biochemical and physical) can induce SA and methicillin-resistant (MRSA) EV production, including iron depletion, sublethal concentrations of vancomycin or ampicillin, and temperature. The content of bacterial EVs has been shown to vary within a given temperature range (34–40 °C), with lower temperature SA EVs more cytotoxic to THP-1 cells (macrophages) than those secreted at 40 °C. Conversely, those secreted at 40 °C have been shown to contain virulence factors, primarily proteins and lipids, exhibiting amplified toxin-mediated erythrocyte lysis compared to their lower temperature counterparts
[37].
Apart from toxins, other virulent factors include protein A coagulase, proteases, aureolysin, Staphylokinase, and collagen adhesins, which are collectively involved in repurposing the hosts’ proteins and the seeding of biofilms. In addition to SA, EVs are produced by other colonizing microbes (Malassezia species) and by the hosts’ repair mechanisms (keratinocytes), suggesting that EVs can modulate infection and AD inflammatory processes.
Impairment of the epidermal barrier occurs via various mechanisms involving numerous pore-forming toxins (Leukocidins ED, SF-PVL, AB, MF, HlgCB, HlgAB, and α-hemolysin). The most notable of these is α-hemolysin
[38], which forms a heptameric β-barrel pore in the membranes of keratinocytes, initiating an ion imbalance. Cellular distribution is further enhanced with the binding of Protein A(PA) to tumor necrosis factor (TNF) receptors on the surface of keratinocytes triggering a sustained Th2 polarized inflammatory response via Langerhans cells. In the ensuing milieu, several other SA endogenous (kallikrein (KLK) 6, KLK13, and KLK14) proteases with a penchant for filaggrin and lipoproteins for the host’s fatty acids (linoleic acid (C18:2)
[39] further amplify the Th2 inflammatory response by inducing thymic stromal lymphopoietin expression in a TLR2/TLR6-dependent manner in primary human keratinocytes
[40]. Finally, SA becomes internalized by the keratinocytes inducing IL-1α production via Toll-like receptor 9 (TLR9).
Other virulent factors include δ-Toxin and PSMα2 and PSMα3, the latter acting synergistically with IgE in the absence of bound antigen
[41]. Last but not least is the role of superantigens (SAgs) and their impact on the corneocytes of the stratum corneum resulting in IgE antibody elevations following antigen presentation to Th2 cells
[42]. SAgs can also trigger mast cell degranulation and induce Th2 cells to release IL-31. Children colonized with bacteria expressing SAgs have a higher disease severity than those with strains
[43].
2. The Initial Impact of SA Toxins and AD Recurrence
At the time of record, the subject also exhibited (not shown) first systemic AD presentations in multiple cutaneous areas (neck, trunk, arms, and ankles). All AD presentations had dissipated within 12 h of administering antibiotics, and skin restoration was observed after 4 weeks.
Without testing, it is debatable whether patient X had subclinical Th2 and Th22 cellular elevations in those areas’ prior infection. However, it seems likely that multiple pools of antigen-primed T and effector memory cells were generated during infection, which could be triggered later
[43].
As discussed previously, only 5% with a genetic predisposition (European ancestry) for AD develop chronic disease. Unfortunately, patient X soon embarked on a long course of mild AD interspersed with periods of acute exacerbations (flares). The first of these resulted from a minor knee skin abrasion (no sign of infection), triggering a hypersensitive response in the folds of both arms coupled with intense itching < 12 h. Initial presentation appeared mixed (bacterial and fungi), with the administration of generic antibiotic creams (week 2) only serving to exacerbate. During the next two weeks, contact allergens were minimized by routinely (every morning) disinfecting bed linen (spraying) and bathing skin folds prior to sleep.
A similar case study involving a nine-year-old boy with chronic AD was reported by Kanchongkittiphon et al.
[44], in which the total serum IgE level was markedly elevated, and the possibility of allergen (mites) immunotherapy was discussed. Regarding patient X, bleach baths were considered, but a targeted bathing approach concentrating on the folds, in combination with chloroxylenol spraying of bed linens (patient absent), was adopted, with time to resolution akin to generic treatments.
Further exacerbations were recorded at the end of 2017 (travel stress), followed by severe flaring after another fall in early 2018, suggesting post-traumatic stress involvement. The impact of psychological stress on AD severity, frequency of exacerbations, and reduced healing potential is well documented
[45]. In such circumstances, the administration of selective serotonin reuptake inhibitors (SSRIs) sertraline and fluoxetine may have been considered, given the drug’s reported ability to reduce levels of anxiety, proinflammatory cytokines and IgE in AD patients
[46][47][48][49][50]. In addition, SSRI paroxetine and fluvoxamine administered to AD patients were shown to reduce pruritus, suggesting central and peripheral nervous system involvement and the neurotransmitter 5-hydroxytryptamine (5-HT). However, anxiety is a complex disorder, and side effects can present if the underlying phenotype has not been identified
[47][51]; consequently, recent investigations are currently restricted to topical SSRI usage, such as fluoxetine, which was shown to improve wound healing in mice by accelerating keratinocyte migration and rebalancing the local immune milieu
[51]. The interplay between immune cells, neurotransmitters, and hormones and the effect on the levels of cortisol, adrenocorticotropic, and corticotropin-releasing hormones released by mature and immature keratinocytes and fibroblasts remains under investigation
[52]. As a side note, for the first 24 months post-trauma (in the absence of an AD flare), the healing time for an abrasion remained significantly prolonged (10–14 days), suggesting IL-4 dysregulation. A model that probably best emulates that period might be the Interleukin-4 Transgenic Mouse developed by Zhao et al.
[53] regarding aberrant wound healing.
Lengthy and frequent AD recurrences can lead to observable thickening in the flexural folds with papule-nodule transitions, plaques with excoriation, lichenification, and significant remodeling. Changes in epidermal thickness resulting from chronic AD are usually accompanied by a higher density of nerve fibers
[54], Th2, and epidermal dendritic cells, increasing regional susceptibility to additional environmental factors such as heat, pressure, and sweat, the latter often associated with soft wound infections. Soft wound treatments range from cold atmospheric pressure plasma
[55], silver and zinc nanoparticles, platelet therapies, chitin bandages, and tens of others.
During 2018–2021, patient X relapsed multiple times, putting them at significant risk from the so-called atopic march that can lead to asthma and allergic rhinitis in later life. However, nodulations rooted in the dermis and epidermis did show significant shrinkage following topical povidone administration in early 2022. Moreover, additional wounds or abrasions rapidly treated with povidone (PV) later in the same year did not evoke flaring and healed quicker, which may have been due to limited IgE production leading to reductions in reagin antibody titers or how the treatment (psychological) was administered
[56]. Interestingly, a recent pilot phase II study investigating the combinatorial antiseptic and anti-inflammatory action of liposomal polyvinyl-pyrrolidone (PVP)-iodine gel reduced the global clinical severity score (GCSS) for all dermatoses
[57][58]. Moreover, PV preventions have been shown to inhibit
C. albicans, and SA biofilm formations often encountered with chronic AD
[59]. Side effects included itching and burning (9 and 14%). Povidone has been shown to counter various SA and MRSA armaments (α-hemolysin, phospholipase C, lipase, elastase, and β-glucuronidase)
[60] and remains the gold standard for biocides in hospital settings
[61][62].
However, the misuse of biocides can lead to a surge in MRSA cases (cross-resistance), as reported during the COVID-19 pandemic
[63]; thus, alternatives treatments need to be developed to limit the emergence of multi-drug resistant microbes (fungi, bacteria, and viruses)
[64]. A recent review by Bieber et al. documents a number of approved inhibitor therapies (IL-4Rα “dupilumab”, anti-IL-13 “tralokinumab”, and JAK1/2) and those currently in clinical trials
[65], which might have been a better option (dupilumab) for patient X.
3. Restoration of Microbial Homeostasis
The skin microbiome consists of regionally specific microbial communities hosting over 1000 species from 19 phyla. These communities’ microbial composition depends on on-site skin physiology, and the microenvironments (dry, moist, foot, sebaceous) encountered there. For example, where water loss is relatively high (dry) and follicle density (inner forearm) low, fungal species (
Malassezia) predominate; in moist regions (folds), bacteria species (
Staphylococcus and
Corynebacterium) hold sway and where follicle density and skin surface area are highest (Sebaceous) (head, neck, and chest) lipophilic Propionibacterium species dominate. Finally, various fungal (
Aspergillus, Cryptococcus,
Rhodotorula,
Malassezia, and
Epicoccum) species can be found in the feet
[66].
Collectively, these microbially diverse communities constitute the bulk of the skin microbiome. The density and diversity of these communities vary with age, sex, and race of the host. Interactions (communications) with the host occur via epithelial and immune cells and are mutually beneficial with the community’s providing protection from pathogenic or translocated species and the host offering “food and board.”
However, these interactions can be perturbed by environmental conditions (skin infection or trauma), providing opportunities for other species, such as SA, that are adept at prioritizing overgrowth and biofilm formations
[66][67], leading to prolonged barrier and immunological dysfunction. Of the many SA strains residing on the skin, the clonal complex 1 (CC1) takes precedence during severe AD, whereas CC30 is prominent on the skin of asymptomatic AD individuals.
Numerous therapeutic approaches, including endolysins, bacteriocins (Nisin Z, Lacticin 3147), small molecule inhibitors (2-aminoimidazole), bacteriophages, and nanoparticles or combinations thereof, can be used to prevent SA biofilm formation
[60][68]. However, in the case of patient X, concerns extended beyond the skin, as the longevity and frequency of the disease suggested the initial involvement of the gut (microbiome) and, later on, the brain (stress) via the gut–brain–skin axis
[69]. Animal model (murine) research regarding factors that affect the axis has revealed the immunoregulatory importance of dietary AHR ligands and hydrophilic bile acids. For example, mice fed an indole-3-Carbinol(I3C)-supplemented diet showed better monocyte-to-dendritic skin differentiation than controls
[70], which is consistent with previous reports highlighting the shift from macrophages via the modulation of transcription factors Irf4 and Blimp-1. Moreover, I3C diet supplementation reversed antibiotic-treated mice with impaired differentiation, suggesting a role for microbial-derived AHR ligands
[70]. The effects of dietary tryptophan-free deficiencies on an experimental autoimmune encephalomyelitis (EAE) model resulted in higher disease scores, increased expression of pro-neuroinflammatory molecules (Ccl2, Nos2, and Tnfa), and prolonged
[71] recovery in mice.
The catabolism of dietary tryptophan represents a major source of indoles for humans. A recent clinical study comparing elevated fasting serum and normal IgE levels in infant AD cohorts showed the former had increased serum indole-3-acetic acid (tryptophan metabolite) and tryptophan levels
[72], suggesting impaired tryptophan metabolism. Moreover, the cohort with normal IgE levels also showed a significant decrease in primary conjugated bile acids (taurocholic, taurochenodeoxycholic acid, and glycochenodeoxycholic). In contrast, the high IgE AD cohort showed cholic acid and chenodeoxycholic acid elevations, suggesting reduced conjugation with a potential risk of
E. coli intestinal overgrowth
[73].
Of the many indole derivatives currently under investigation
[74][75][76][77], the work by Yu et al. demonstrated the potential of indole-3-aldehyde treatments for lesional and non-lesional AD patients by attenuating skin inflammation and TSLP keratinocyte expression in an AD-like dermatitis murine model (MC903)
[78]. More recently, Fang et al. further emphasized the importance of microbially (
Limosilactobacillus reuteri) produced indole derivatives (indole lactic acid and indole propionic acid) in attenuating AD by demonstrating its ability to suppress IgE, TSLP, IL-4 and 5 levels in mice
[79]. Regarding the anti-inflammatory potential of hydrophilic bile acids (tauroursodeoxycholic acid (TUDCA)), reports have shown that they reduce the unfolded protein response (UPR) seen in keratinocytes
[80][81], hypersensitivity in asthmatic patients, as well as reverse the amyloid beta-oxidation of mitochondrial membranes in Alzheimer’s murine models
[82][83]. Whether TUDCA alone or blended bile acids treatments can re-establish skin microbial homeostasis directly or indirectly via changes in the gut microbiota (probiotic interventions or fecal matter transfer) requires further investigation. In addition to indole derivatives, a study by Hwang et al. investigated whether n-3 polyunsaturated fatty acids (PUFA) in red blood cells (RBC) had a protective effect on the atopy of 380 preschoolers. Preschoolers with atopy had lower total RBC n-3 PUFA levels and a greater n-6/n-3 PUFA ratio than the controls, suggesting lower n-3 PUFA may play a role in child atropy
[84].
Microbial analysis of AD fecal matter by Reddel
[85] and co-workers demonstrated that microbiomes of 19 AD children were characterized by an increase of
Faecalibacterium,
Sutterella,
Oscillospira, Bacteroides, and Parabacteroides and a significant reduction in short-chain fatty acid (SCFA)-producing bacteria (
Bifidobacterium,
Blautia,
Coprococcus,
Eubacterium, and
Propionibacterium) compared with 18 healthy individuals. More recently
[86], Kim et al. showed that fecal matter transfer (FMT) from a healthy murine donor ameliorated AD in mice; resulted in reduced IgE levels and the numbers of mast cells, eosinophils, and basophils; and the restoration of the Th1/Th2 balance. An initial clinical study (
n = 9)
[87] assessing the clinical efficacy of capsule FMT as a viable treatment in adults with moderate-to-severe AD over eight months demonstrated a significant reduction in the average SCORAD score from baseline pre-FMT as well as a decrease in corticosteroid usage, disease severity, and partial assimilation of the donor’s microbiota.
This entry is adapted from the peer-reviewed paper 10.3390/antibiotics12020222