Introduction
The human skin, the body’s largest organ, plays a crucial role as a protective barrier against external insults. Resident microorganisms on the skin contribute to its protective function, and any alterations in their composition have been associated with various skin conditions, including acne vulgaris.1
Acne vulgaris is a persistent skin condition characterized by specific manifestations like comedones, papules, pustules, and nodules. Its development is linked to factors such as increased sebum production, abnormal keratinization of follicles, bacterial colonization, and inflammation. Cutibacterium acnes (C. acnes), a common bacterial species in sebaceous gland-rich skin areas, is suggested as a potential acne-causing pathogen.1 Despite its ability to produce proinflammatory free fatty acids from sebum, studies have presented conflicting results, with C. acnes not consistently elevated in acne-afflicted skin compared to normal skin.2,3
The emergence of acne vulgaris is also associated with other skin microorganisms. The term “dysbiosis,” referring to an imbalance in the relative abundance of the entire skin microbiome, may contribute to acne vulgaris by affecting innate and adaptive immune responses within pilosebaceous follicles. The specific roles of individual microorganisms in the pathogenesis of acne vulgaris remain undefined.1,3 While Staphylococcus aureus (S. aureus) and Staphylococcus epidermidis (S. epidermidis) have been well-studied due to their proximity to the skin, many other minor skin microorganisms still lack comprehensive examination.
The microbiome comprises bacteria, viruses, fungi, their genes, metabolites, and surrounding environment, whereas microbiota refers to microorganisms within a specific niche. Microbial cells outnumber human cells by approximately tenfold, and skin and gut microbes are essential for immune, hormonal, and metabolic homeostasis.3 The bacterial composition varies between individuals and across body sites, influenced by environmental factors such as soaps, cosmetics, antibiotics, occupation, temperature, humidity, and UV exposure.3
Consequently, gaining a better understanding of the microbiome profile in acne vulgaris compared to normal skin is crucial. This study highlights the role of skin microbiome dysbiosis in the pathogenesis of acne vulgaris and provides a foundation for future microbiome-based therapeutic strategies. Further research is required to elucidate microbial mechanisms and to develop targeted interventions for the management of acne vulgaris.
Materials and Methods
Subject Recruitment
Thirty-two participants with acne vulgaris and one-hundred-eight controls were recruited consecutively during November 2022 to March 2023. Acne vulgaris diagnosis was made by dermatologists after thorough physical examination of each participant’s face. Only participants with no visible acne lesions and acne scars were included as controls.
The exclusion criteria were as follows: pregnancy, presence of other skin diseases on the face, presence of psoriasis or atopic eczema lesions in any part of the body, acne conglobata, primary or secondary immunodeficiency diseases, consumption of oral antibiotics, immunosuppressants, immunomodulators, hormones, or retinoids within 1 month before enrolment, history of using any topical acne prescription within 2 weeks before enrolment. Written informed consent following guidance from Declaration of Helsinki was taken from each participant. This study was approved by the ethics committee of Dr.
Cipto Mangunkusumo National Central General Hospital, Faculty of Medicine, Universitas Indonesia. The ClinicalTrials.gov registration number of this study is NCT05838534. Informed consent was obtained, and ethical approval was granted by the Health Research Ethics Committee of FKUI-RSCM (No.
KET-482/UN2.F1/ETIK/PPM.00.02/2022, dated May 23, 2022).
Sample Collection
Skin microbiome samples were obtained from the left cheek using sterile swabs pre-moistened with DNA/RNA Shield solution. Each swab was applied with uniform pressure in vertical, horizontal, and diagonal strokes across a 4 × 4 cm area for approximately 50 seconds. The swab tip was then aseptically detached and placed into a collection tube containing DNA/RNA Shield.
Samples were transported in an insulated container with dry ice (~4°C) and subsequently stored at –80°C within 24 hours of collection.
Prior to sampling, all participants were instructed to cleanse their face exclusively with a standardized mild facial cleanser provided by the investigators for at least one week. On the day of collection, the application of any topical products or cosmetics was prohibited for a minimum of 12 hours.
DNA Extraction and Sequencing
DNA extraction followed the DNeasy PowerSoil Kit™ protocol (Qiagen, CA, USA). After extraction, DNA purity and concentration were assessed, with each sample required to exhibit a DNA purity value between 1.6 and 2 and a quantity equal to or below 0.05 ng/mL. The amplification of the V3 and V4 hypervariable regions of the 16SrRNA gene utilized the forward primer 5′-CCTAYGGGRBGCASCAG-3′ and the reverse primer 5′-GGACTACNNGGGTATCTAAT-3′.
Post-amplification, DNA concentration in all samples was gauged, and random samples underwent agarose gel testing to confirm the presence of the 16SrRNA gene. The anticipated 16SrRNA concentration in each sample should be equal to or greater than 2nM. Prior to sequencing, the Nextera XT DNA library preparation kit (Illumina, CA, USA) was used for indexing.
Sequencing was conducted on the Illumina MiSeq platform (Illumina, CA, USA).
Bioinformatic Analysis
The analysis was conducted with Quantitative Observation of Microbial Ecology (QIIME) version 1.7.0. Raw data from all samples underwent filtering, trimming, merging, and denoising through the USEARCH plugin. In cases where the quality of one read was subpar (Phred score <20 and duplications <50%), only the higher-quality read was considered for analysis.
Annotation of operational taxonomic units (OTUs) followed the Ribosomal Database Project guidelines. OTUs were incorporated into the final analysis if their proportion exceeded 0.1% in each sample, and they had to be present in at least 20% of the samples.
Statistical Analysis
The relative abundance of each microorganism found in samples were presented in percentage. Alpha diversity, measurement of microbial diversity within each sample, was counted using the Shannon index. To compare alpha diversity and the relative abundance of microorganisms across three acne grades, the Kruskal–Wallis test was used.
Statistical significance was considered for p-values less than 0.05. If the comparisons were deemed statistically significant, the Mann–Whitney U-test was then conducted between two groups.
Results
In this study, a total of 144 participants were recruited. The sociodemographic and clinical characteristics of the participants were provided in Table 1. The mean age of participants was 25.6 ± 3.2 years, and the majority were female.
Participants with AV has significantly higher history of AV in first degree family compared to controls (p < 0.05). Most participants wash their face two to three times per day with various soaps either moisturizing soap, non-moisturizing soap, or antiseptics. About 5% of controls have been diagnosed and treated for AV in the past.
There is no significant difference in the mean face sebum excretion amount measured using sebumeter between AV and controls.Table 1Sociodemographic and Clinical Characteristics of the ParticipantsAcne GradesControls (n = 36)Acne Vulgaris (n = 108)p ValueAge (years)25 (19–40)26 (18–39)Genders, n (%)Male15 (41.7)24 (22.2)Female21 (58.3)84 (77.8)Educations, n (%)Middle school to college11 (30.6)47 (43.5)University or higher25 (69.4)61 (56.5)Family history of acne, n (%)Yes8 (22.2)46 (42.6)0.047‡,*No28 (77.8)62 (57.4)Daily facial washing frequency, n (%)<2 times7 (19.4)7 (6.5)0.08§2–3 times28 (77.8)97 (89.8)>3 times1 (2.8)4 (3.7)Soap type, n (%)Moisturizing19 (52.8)49 (45.4)Non-moisturizing15 (41.7)55 (50.9)Antiseptics1 (2.8)3 (2.8)None0 (2.8)1 (0.9)History of acne medications/procedure, n (%)Yes2 (5.6)20 (18.5)0.11§No34 (94.4)88 (82.5)Mean Facial Sebum Excretion (mg/cm)61.6 (10.4–172)77.3 (11–338.4)0.16¶Notes: ‡Using Chi-square test. §Using Fischer’s exact test. ¶Using Mann–Whitney U-test. *Statistically significant.
All samples went through DNA extraction and sequencing processes. Sixteen samples from AV participants and five samples from controls had less than 1,000 reads and cannot be analysed further. The average read in each sample is 63,008.
The average length of each read is 191 bp. The total reads in all samples were 9,073,415. There were 1,451 unique OTUs in total.
The comparison of the relative abundance of observed species and Shannon diversity index between groups was shown in Table 2. The Shannon index was higher in controls compared to AV participants and statistically significant (p < 0.001). The value is below 3 in almost all samples which may signify low bacterial diversity in all samples.
The comparison of C. acnes’ relative abundance was not remarkable. None of our samples showed the presence of S. aureus. The relative abundance of S. epidermidis is higher in AV participants and statistically significant.
Species with relative abundance above 0.1% and can be detected in at least 20% samples were taken into final analysis. Other lesser species constitute more than 50% of the microbiota found in each sample, yet there is no difference in the proportion between both samples. We also did correlation analysis of MFSE and the previous parameters in Table 3 and found significant correlation between MFSE and Shannon Index and relative abundance of S. epidermidis.Table 2Comparison of Shannon Index and Relative Abundance of Microbiome Between Study ParticipantsAcne GradesControls (n = 31)Acne Vulgaris (n = 92)p ValueShannon Index2.96 (1.1–4.2)2.4 (1.2–4.7)<0.001†,*Relative abundance of C. acnes (%)29.5 (0.2–60)31.8 (0.3–85.4)0.83†Relative abundance of S. epidermidis (%)0.1 (0–0.5)0.2 (0–2)0.02†,*Others (%)70 (39.9–99.3)68.2 (14.4–99.7)0.78†Notes: †Presented in median (range), using Mann–Whitney test. *Statistically significant.Table 3Correlation Between Mean Facial Sebum Excretion with Shannon Index and Relative Abundance of Microbiome Among Study ParticipantsMean Facial Sebum Excretion (n = 123)Pearson’s Correlation Coefficientp ValueShannon Index−0.285<0.001*Relative abundance of C. acnes0.0460.613Relative abundance of S. epidermidis0.2330.01*Relative abundance of other bacteria−0.050.583Note: *Statistically significant.
Discussion
Several individual factors might affect the outcome in a study comparing microbiome profile between AV patients and controls. Fortunately, most of those factors are comparable between groups in this study, except for family history of AV. A family history can increase the risk of AV by up to three times, especially if the history comes from parents.4 Family members may share similar skin microbiomes due to genetic factors, environmental exposures, and lifestyle choices.
Song reported family members living under the same roof shared similar skin microbiota. However, this finding is heavily affected by the frequent skin-to-skin contact between individuals.5 Furthermore, it is not possible to segregate individuals based on family relation between participants in this study.
The Shannon index is used to assess alpha diversity, indicating the number and balance of species in a habitat, such as skin samples. A lower index suggests fewer species and dysbiosis.6 While previous studies found no significant differences in dominant microbiota between acne vulgaris (AV) patients and healthy individuals, this study noted an increase in rare bacteria in AV. The Shannon index was lower in AV patients compared to controls, contrasting with some studies that reported no significant difference.
Methodological variations in sample collection may account for these discrepancies (pore stripping vs skin swabbing).2,7 The decrease in diversity in AV may lead to an increase in pathogenic microorganisms. Furthermore, commensal microorganisms have a specific role in educating both innate and adaptive immunity. Disruption of this microbial composition may alter skin conditions to become more pro-inflammatory.
Several microorganisms, namely C.acnes and various Staphylococcus species, are the most abundant inhabitants of human skin. C.acnes is considered a causative bacterium of AV because it can enhance sebaceous gland lipogenesis, keratinocyte differentiation, and local inflammatory.8 In this study, no significant differences were found in the proportion of C.acnes among AV patients and normal. This finding supports the research by Barnard et al, which reported that AV is more related to the presence of increased proportions of certain strains of C.acnes, specifically RT4, RT5, and RT8 strains.2 We did not perform whole genome sequencing in this study, therefore we could not analyze the differences in C.acnes strains.
S. aureus was not detected in any skin samples, aligning with findings from other studies that report low proportions (less than 0.1%) in AV lesions. This bacterium is believed to exacerbate inflammation in AV but is typically identified in throat or nasal swabs where it is more abundant.9,10 In contrast, S. epidermidis was found in greater numbers in AV participants compared to controls. However, the role of S. epidermidis in AV is still unclear.
While it could potentially inhibit pathogenic bacteria like C. acnes through competition or antimicrobial peptide production, this study found no reduction in C. acnes levels in AV patients. Another hypothesis suggests that the increased S. epidermidis may be related to its biofilm formation, which could block sweat glands and provoke inflammation.11,12 Further research, including whole genome sequencing and biofilm analysis, is needed to clarify these findings.
Various other microorganisms were also identified from skin samples in this study, including S. cohnii, C. macginleyi, S. caprae, and S. hominis. There were no overall differences in the number of these microorganisms between AV samples and controls, nor across different severity levels of AV. These microorganisms are commensal bacteria on human skin, but C. macginleyi, S. caprae, and S. capitis can develop into opportunistic pathogens.13,14 The role of these microorganisms in AV has not been extensively explored, but these findings indicate an imbalance in composition.
The composition of skin microorganisms is influenced by physiological factors like sebum and moisture. In this study, samples were taken from the cheeks, which are less seborrheic. A negative correlation was found between skin microbial diversity, measured by the Shannon index, and sebum levels, mirroring findings from a similar study in India.15 Sebum provides substrates for lipophilic bacteria like C. acnes, which can convert oleic acid into propionic and acetic acids that inhibit other bacteria, leading to decreased microbial diversity as sebum levels rise.
However, this study found no correlation between sebum levels and the proportion of C. acnes, likely due to limitations in sampling. Conversely, S. epidermidis showed a positive correlation with sebum levels. While previous research has focused on the genus level, findings indicate that S. epidermidis can metabolize fatty acids in sebum and produce succinic acid, enabling it to thrive in seborrheic environments.
This is supported by reports that S. epidermidis can survive across varying sebum levels.16 Additionally, the significant increase of S. epidermidis relative to C. acnes suggests that the more seborrheic conditions on the cheeks, which come with increased moisture, favor S. epidermidis.15
Acne vulgaris (AV) is a chronic disorder of the pilosebaceous unit that predominantly affects adolescents and is characterized by polymorphic lesions, including comedones, papules, pustules, cysts, nodules, scarring, and dyspigmentation. An acneiform presentation of cutaneous lupus erythematosus (CLE) is extremely rare but may clinically mimic acne.17
The severity of AV may also be associated with dysbiosis of commensal skin flora, particularly Staphylococcus capraeand S. epidermidis.18 The predominant bacterial isolates identified were S. epidermidis (50.5%), C. acnes (11.0%), and S. aureus (7.7%). Among available antibiotics, doxycycline and minocycline demonstrated the highest efficacy, whereas resistance to erythromycin and clindamycin was frequently observed.19
Standard therapy for acne vulgaris in Indonesia, as recommended by the national clinical practice guidelines, was evaluated in a retrospective observational study. We found at baseline, 63.4% of patients presented with moderate acne.20 Most patients received a combination of retinoic acid, benzoyl peroxide, and either topical or oral antibiotics, with several adjuvant such as comedone extraction,21 chemical peeling, pulse dye laser and others. Observing after 12 weeks of treatment, a significant reduction in the median number of non-inflammatory lesions, inflammatory lesions and total lesions significantly reduction (p < 0.001).20 These findings support the implementation of national guidelines and the promotion of antimicrobial stewardship in Indonesia.20
Isotretinoin remains the most effective therapy, targeting all major pathogenic factors of acne, although it requires close monitoring. In contrast, topical tretinoin is safe, well tolerated, and effective not only for AV but also for photoaging.22,23
As adjunctive therapy, nicotinamide combined with zinc and antibacterial adhesive (ABA) has shown efficacy in reducing non-inflammatory lesions within two weeks, suggesting a potential role in improving early treatment outcomes.24 Meanwhile, although serum vitamin D levels in the study population were low (mean 17.29 ng/mL), no significant correlation was observed with acne severity.25
Conclusion
In conclusion, this study highlights significant differences in the skin microbiome of AV patients compared to healthy individuals, particularly reflected in the Shannon index, which indicates decreased microbial diversity in AV. Despite no notable changes in the proportion of C. acnes, the increase in rare bacteria and the higher prevalence of S. epidermidis in AV patients, suggests complex interactions that may influence inflammation and skin health. The findings underscore the need for further investigation into specific strains of C. acnes, the role of S. epidermidis and its implications for AV pathogenesis.
Additionally, the study reveals how physiological factors like sebum levels can affect microbial composition, with increased sebum potentially favoring S. epidermidis. Overall, these insights contribute to understanding the dysbiosis associated with AV and serve as foundation for future research and therapeutic approaches.