Introduction
Acne vulgaris (AV), commonly known as acne, is a prevalent skin condition affecting approximately 85% of adolescents and young adults worldwide1,2. It can persist into the third decade of life3, impacting appearance and leading to complications such as dyspigmentation, scarring, and disfigurement, significantly affecting a patient’s quality of life. Acne is associated with low self-esteem, stress, anxiety disorders, depression, suicidal ideations, and unemployment4, imposing a substantial burden on healthcare services and financial resources globally.
The pathogenesis of acne involves four primary domains: sebum production quantity, follicular hyperkeratinization, skin inflammation, and microbial dysbiosis. Cutibacterium acnes, a Gram-positive, aerotolerant anaerobic bacterium, is a key member of the skin microbiota in individuals with AV and those with healthy skin5. Imbalance in the growth and loss diversity6 of C. acnes disrupts the skin microbiota equilibrium7, contributing to AV pathogenesis by increasing sebum production, altering sebum composition, and promoting follicular hyperkeratization and microcomedone formation8.
In addition, C. acnes triggers innate immune responses and proinflammatory cytokine cascades, further exacerbating inflammation. Moreover, C. acnes disrupts the balance between M1 and M2 macrophage functions, favoring the M2 phenotype9. This shift impairs antimicrobial activity, hampers immune defenses, increases bacterial burden, and contributes to pathological scarring.
Given their potent antibacterial and anti-inflammatory effects against C. acnes, oral and topical antibiotics such as tetracyclines, macrolides, and lincosamides remain the treatment of choice for moderate and severe AV.
In response to the escalating drug resistance crisis, primarily due to the widespread misuse of antibiotics, including inadequate dosing and prolonged treatment courses10, the resistance rates of C. acnes have surged globally in the past decade11,12. Reports indicate a spectrum of resistance, from single agents like macrolides, clindamycin, tetracycline, and, more recently, levofloxacin, to highly multidrug-resistant C. acnes strains. This alarming trend has necessitated a shift to second-line antibiotics, such as fluoroquinolones and trimethoprim/sulfamethoxazole (TMP/SMX), which offer broad-spectrum activity against C. acnes.
However, resistance to these antibiotics has also been documented, with the prevalence of resistant strains expected to rise across various antimicrobial classes used in AV treatment13. In Thailand, where topical and systemic antibiotics are readily available over the counter, the dispensing of medications by locally registered pharmacists aims to reduce hospital workload. In our clinical practice, we frequently encounter treatment failures within a few months of commencing first-line antibiotic therapy for AV.
Our center’s data revealed that12 all C. acnes isolates were resistant to at least one commonly used acne antibiotic, TMP/SMX, with nearly half classified as pan-drug resistant.
In addition to antimicrobial resistance, disruption of the skin microbiome caused by medications such as retinoids and antibiotics, as well as skincare products14, presents a major challenge in the treatment of AV. Recent approaches15 have increasingly focused on maintaining overall skin health while improving the severity of acne. The repurposing of existing therapeutics16 and bioactive strategies17,18 to target acne pathogenesis is gaining more attention.
Photodynamic therapy (PDT), when combined with an appropriate photosensitizer (PS), demonstrates clear efficacy and instant solutions. PDT offers a multitargeting approach to AV pathogenesis, addressing key elements such as the management of drug-resistant C. acnes, even in its biofilm state19, and the reduction of sebum production20–23. Unlike conventional treatments, PDT also provides unique anti-inflammatory and immunomodulatory effects.
It eradicates pathogenic C. acnes strains independently of antibiotic resistance24,25 and contributes to the restoration of the intrafollicular microbiome26–28 and normalization of the wound-healing process29,30, resulting in minimal acne scarring. The appropriate pairing of a PS and red light in PDT facilitates the high-yield production of powerful cytotoxic singlet oxygen molecules (1O2*) and other reactive oxygen species (ROS), selectively at target sites, leading to localized direct bacterial killing and beneficial immune modulation. Various PSs have been employed in PDT for AV treatment, including 5-aminolevulinic acid (ALA)31, methyl-aminolaevulinic acid, curcumin32,33, and methylene blue34.
Chlorophyll, a relatively inexpensive, lipid-soluble magnesium porphyrin complex, has been anciently utilized topically35 for various skin conditions and more recently as a PS in PDT, primarily in oncological applications due to its affinity for tumor tissues36. Water-soluble derivatives, including chlorin e6 derivatives37 and chlorophyllin metal complexes (CHL-M)38, provide greater molecular stability, enhanced ROS production, and high affinity for biomolecules, including microorganisms, thereby enhancing antimicrobial efficacy39–41. Beyond their antimicrobial potential, CHL-M compounds also exhibit anti-inflammatory42, antioxidant, wound healing43, and antiaging properties44.
Their ability to selectively accumulate within the pilosebaceous unit makes them suitable for the treatment of AV45,46 and possibly other hair follicle-origin conditions47, such as rosacea47, with the additional advantage of improving skin quality43.
We employ our proprietary formulation of sodium iron chlorophyllin (CHL-Fe)-based PS48, along with RL, for photodynamic therapy (CH-PDT) at King Chulalongkorn Memorial Hospital, a specialised photodermatology service. CH-PDT is utilized for chronic wounds, photorejuvenation, recalcitrant inflammatory skin dermatoses, and AV. Due to the rise in drug-resistant C. acnes and limitations of conventional antibiotics and other AV treatments, we aimed to investigate the efficacy of CH-PDT in reducing drug-resistant C. acnes isolates identified in our earlier study12.
A computer-based random-number generator was used to select 10 strains of drug-resistant C. acnes from the established database (Table 1; Supplementary Table 1 provides detailed patient profiles for each isolate).
Table 1Drug-resistant profiles of selected Cutibacterium acnes strains.Antibiotic resistance profileTetracyclinesClindamycinErythromycinBactrimDoxycyclineTetracyclineDR-CA1/MDR-CA1//MDR-CA2//MDR-CA3//MDR-CA4///MDR-CA5Resistant to doxycycline or tetracycline//XDR-CA1///XDR-CA2///PDR-CA1/////PDR-CA2/////
To develop an optimal CH-PDT protocol with the minimum effective concentration–fluence dose combination, three sequential experiments were conducted (Fig. 1):
In Experiment 1, the viability of pan-drug-resistant C. acnes strain 1 (PDR-CA1) was assessed following treatment with varying CHL-Fe concentrations (0, 1.875, 3.75, 7.5, and 15 mg/dL), RL fluences (0, 15, 30, 45, and 60 J/cm2), and combination treatments (CH-PDT) with CHL-Fe concentrations of 0, 3.75, 7.5, and 15 mg/dL and light fluence of 0, 3.75, 7.5, and 15 J/cm2.
Experiment 2 involved testing pan-drug-resistant C. acnes strain 2 (PDR-CA2) with the same CH-PDT combinations as in Experiment 1—CHL-Fe concentrations of 0, 3.75, 7.5, and 15 mg/dL, and RL fluences of 0, 3.75, 7.5, and 15 J/cm2—to validate reproducibility and determine the minimum lethal dose.
In Experiment 3, the efficacy of CH-PDT was assessed on the remaining 8 C. acnes strains (drug-resistant C. acnes strain (DR-CA), multidrug-resistant C. acnes strains 1–4 [MDR-CA1, MDR-CA2, MDR-CA3, MDR-CA4], and extremely drug-resistant C. acnes strains 1–2 [XDR-CA1, XDR-CA2]) using the minimum lethal dose identified in Experiment 2: CHL-Fe 7.5 mg/dL combined with RL 3.75 J/cm2. A treatment dose was considered lethal if it resulted in a reduction of more than 3 log10 (≥ 99.9%) in colony-forming units (CFU)/mL compared to the untreated bacterial control.
Fig. 1Overview of the experimental procedures for CH-PDT on bacterial populations.
Results
Experiment 1
Effect of various CHL-Fe concentrations on PDR-CA1 (Table 2)
There was a minimal decrease in PDR-CA1 viability after treatment with CHL-Fe concentrations of 1.875, 3.75, 7.5, and 15 mg/dL compared to the untreated group (p > 0.05). None of the tested concentrations exhibited bactericidal activity, defined as a ≥ 3 log10 reduction in CFU/mL (Fig. 2a).
Effect of various RL fluences on PDR-CA1 (Table 2)
A slight decrease in viable counts was observed with increasing RL fluence. Statistically significant reductions in viability were observed at 45 J/cm2 (log10 CFU change: −0.15; 95% confidence interval [CI]: −0.27, 0.03) and 60 J/cm2 (log10 CFU change: −0.25; 95% CI: −1.42, 0.91) compared to the untreated control group (Fig. 2b).
However, none of the red light (RL) doses achieved bactericidal activity, as all reductions were less than 3 log10 CFU/mL.
Fig. 2Effect of various CHL-Fe concentrations and red light fluences on PDR-CA1.
Table 2Mean log10 CFU/mL reduction in PDR-CA1 following treatment with CHL-Fe alone and red light alone, compared to baseline.CHL-Fe concentration (mg/dL)Fluence (J/cm2)PDR-CA11.8753.757.515Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)−0.02(− 0.31 to0.28)0.265−0.06(− 0.27 to−0.14)0.06−0.10(− 0.20 to0)0.06−0.15(− 0.30 to0.03)0.04Fluence (J/cm2)CHL-Fe concentration (mg/dL)PDR-CA115304560Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)−0.06(− 0.27 to0.14)0.06−0.10(− 0.20 to0)0.06−0.15(− 0.27 to0.03)0.04−0.25(− 1.42 to0.91)0.04
Effect of CHL-Fe-mediated PDT (CH-PDT) on PDR-CA1 (Fig. 3a)
Since CHL-Fe or RL alone did not show significant bactericidal activity in previous experiments, the lowest effective RL fluence of 15 J/cm2 was selected as the reference point. Lower fluences of 7 and 3.5 J/cm2 were also tested in combination with various CHL-Fe concentrations. The results demonstrated concentration- and fluence-dependent antibacterial effects of CH-PDT on PDR-CA1 (Fig.
3). A combination of CHL-Fe 3.75 mg/dL with RL ≥ 7 J/cm2 resulted in significant bacterial killing (log10 CFU reduction: −6.5; 95% CI: −6.80, − 6.38). Similarly, the lowest RL fluence of 3.5 J/cm2 paired with CHL-Fe concentrations ≥ 7 mg/dL achieved bactericidal activity (log10 CFU reduction: −3.94; 95% CI: −4.30, − 3.41) (Table 3).
Experiment 2
Effect of CHL-Fe-mediated PDT (CH-PDT) on PDR-CA2 (Fig. 3b)
Similar to the findings in PDR-CA1, CH-PDT demonstrated a dose-dependent bactericidal effect against PDR-CA2. Combining the lowest CHL-Fe concentration of 3.75 mg/dL with RL fluences of 7 J/cm2 or higher resulted in a significant reduction in bacterial viability (log10 CFU reduction: −6.66; 95% CI: −7.83, − 5.11). In addition, when the lowest RL fluence of 3.5 J/cm2 was combined with CHL-Fe concentrations of 7 mg/dL or higher, a similarly effective bactericidal outcome was achieved (log10 CFU reduction: −3.57; 95% CI: −5.97, − 3.20) (Table 3).
Fig. 3Effect of various CHL-Fe concentrations combined with red light fluences on PDR-CA1 and PDR-CA2.
Table 3Mean log10 CFU/mL reduction for PDR CA1 and PDR CA2 following CH-PDT compared to baseline.CHL-Fe concentration(mg/dL)Fluence (J/cm2)0 (control)3.75PDR-CA1PDR-CA2PDR-CA1PDR-CA2Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)−0.02(− 0.06 to0.02)0.200.02(0.01 to0.04)0.030.01(− 0.26 to0.28)0.43−0.03(− 0.16 to0.12)0.153.5−0.04(− 2.35 to2.35)0.20−0.04(− 0.25 to0.15)0.06−2.25(− 3.50 to−1.01)< 0.001−2.35(− 4.23 to−3.57)< 0.0017−0.05(− 0.59 to0.42)0.16−0.05(− 0.41 to0.30)0.09−6.5(− 6.80 to−6.38)< 0.001−6.66(− 7.83 to−5.11)0.00115−0.07(− 0.36 to 0.25)0.06−0.07(− 0.32 to0.19)0.02−7.80(− 7.81 to−7.76)< 0.001−7.72(− 8.27 to−7.14)< 0.001CHL-Fe concentration(mg/dL)Fluence (J/cm2)7.515PDR-CA1PDR-CA2PDR-CA1PDR-CA2Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)−0.01(− 0.23 to0.25)0.85−0.06(− 0.39 to 0.29)0.08−0.04(− 0.07 to 0.02)0.17−0.05(− 2.95 to1.7)0.173.5−3.94(− 4.30 to−3.41)0.001−3.57(− 5.97 to−3.20)< 0.001−4.52(− 5.71 to−3.50)< 0.001−4.62(− 7.83 to−5.11)< 0.0017−6.96(− 8.27 to−5.65)0.003−7.33(− 15.16 to0.92)0.003−7.04(− 11.65 to−3.19)0.003−7.49(− 12.37 to−2.35)0.00115−7.80(− 7.81 to−7.76)< 0.001−7.72(− 8.27 to−7.14)< 0.001−7.80(− 7.81 to−7.76)< 0.001−7.72(− 8.27 to−7.14)< 0.001
Experiment 3
Effect of CHL-Fe-mediated PDT (CH-PDT) on the remaining 8 C. acnes strains (Fig. 4a-h)
CH-PDT, using a single low-dose nontoxic combination of CHL-Fe at 7.5 mg/dL and RL at 3.5 J/cm2, effectively reduced viable bacterial counts across all 8 selected C. acnes strains, irrespective of their drug-resistant profiles. The treatment achieved bactericidal efficacy, defined as a reduction of > 3 log10 CFU/mL, in all strains tested. The log10 CFU reductions were as follows: DR-CA (− 5.92; 95% CI: −6.00, − 5.83), MDR-CA1 (− 3.57; 95% CI: −3.91,−3.24), MDR-CA2 (− 3.86; 95% CI: −4.03, − 3.68), MDR-CA3 (− 3.67; 95% CI: −3.99,−3.35), MDR-CA4 (− 3.86; 95% CI: −4.06, − 3.65), MDR-CA5 (− 3.62; 95% CI: −3.83, − 3.42), XDR-CA1 (− 3.76; 95% CI: −3.83, − 3.69), and XDR-CA2 (− 3.77; 95% CI: −4.18, − 3.35) (Tables 4 and 5).
Fig. 4Effect of lethal concentration-fluence dose of CHL-Fe and red light from Experiment 2 on remaining C. acnes strains.
Table 4Mean log10 CFU/mL reduction in C. acnes strains (DR-CA, MDR-CA1, MDR-CA2, and MDR-CA3) following CH-PDT compared to baseline.CHL-Fe concentration(mg/dL)Fluence (J/cm2)0 (control)DR-CAMDR-CA1MDR-CA2MDR-CA3Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)0.01(− 0.06 to0.08)0.76−0.03(− 0.14 to0.06)0.25−0.04(− 0.08 to − 0.01)0.03−0.03(− 0.07 to0.01)0.103.5−0.05(− 0.10 to−0.01)0.04−0.02(− 0.14 to0.10)0.55−0.06(− 1.12 to − 0.0001)0.05−0.07(− 0.09 to−0.05)0.004CHL-Fe concentration(mg/dL)Fluence (J/cm2)7.5DR-CAMDR-CA1MDR-CA2MDR-CA3Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)−0.06(− 0.08 to−0.02)0.03−0.06(− 0.08 to−0.04)0.005−0.05(− 0.08 to−0.01)0.02−0.07(− 0.13 to−0.004)0.0443.5−5.92(− 6.00 to−5.83)< 0.001−3.57(− 3.91 to − 3.24)< 0.001−3.86(− 4.03 to−3.68)< 0.001−3.67(− 3.99 to − 3.35)< 0.001
Table 5Mean log10 difference for MDR-CA4, MDR-CA5, XDR-CA1, and XDR-CA2 compared to baseline.CHL-Fe concentration(mg/dL)Fluence (J/cm2)0 (control)MDR-CA4MDR-CA5XDR-CA1XDR-CA2Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)0.01(− 0.06 to0.08)0.76−0.03(− 0.14 to0.06)0.25−0.04(− 0.08 to − 0.01)0.03−0.03(− 0.07 to0.01)0.103.5−0.05(− 0.10 to−0.01)0.04−0.02(− 0.14 to0.10)0.55−0.06(− 1.12 to − 0.0001)0.05−0.07(− 0.09 to−0.05)0.004CHL-Fe concentration(mg/dL)Fluence (J/cm2)7.5MDR-CA4MDR-CA5XDR-CA1XDR-CA2Mean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-valueMean difference(95% CI)P-value0 (control)−0.05(− 0.07 to − 0.03)0.01−0.05(− 0.10 to0.01)0.06−0.08(− 0.15 to−0.02)0.03−0.08(− 0.19 to0.03)0.093.5−3.86(− 4.06 to−3.65)0.20−3.62(− 3.83 to−3.42)< 0.001−3.76(− 3.83 to−3.69)< 0.001−3.77(− 4.18 to−3.35)< 0.001
Discussion
PDT is a comprehensive and multitargeted approach for treating AV by addressing various key aspects of its pathogenesis in a single modality, including the reduction of sebum production by suppressing lipid synthesis and shrinking sebaceous glands22,23, while effectively eradicating pathogenic C. acnes strains24,25. In addition, PDT helps restore the intrafollicular microbiome26–28. A significant advantage of PDT is its ability to normalize the wound-healing process29,30, thereby minimizing postacne scarring.
PDT is well-tolerated31 across diverse patient demographics, including different age groups and conditions49. Selecting the appropriate PS type and concentrations, along with the right RL fluence, for optimizing PDT outcomes in AV. Research has focused on enhancing chlorophyll-a properties to develop novel PS derivatives37 like chlorin e6 trimethyl ester50, CHL, and CHL-M which exhibit favorable properties for AV treatment, including anti-inflammatory42,51, antioxidant, wound healing43, antiaging44, and antimicrobial properties41,50,52,53.
In response to the 2025 Health Emergency Appeal addressing displacement and disease outbreaks, antimicrobial resistance has been identified as a critical global health priority. The widespread use of antimicrobials in acne treatment has led to the emergence of resistance in commensal skin bacteria, including Staphylococcus epidermidis and Cutibacterium species like C. avidum and C. granulosum54. Resistance in C. acnes is particularly concerning as it reduces treatment effectiveness and allows drug-resistant strains to persist postantibiotic treatment55, increasing the risk of acne recurrence.
Additionally, C. acnes is now recognized as an opportunistic pathogen causing infections in various medical fields such as neurosurgery56, ophthalmic57, and orthopedics58, particularly in patients with prosthetic heart valves59 and joint implants. These infections, often atypical and associated with high mortality rates, are challenging to diagnose and treat due to C. acnes’ ability to form biofilms, leading to chronic or treatment-resistant infections.
Jeon et al.60. have demonstrated an effective dose-dependent antimicrobial and anti-inflammatory effect of chlorin e6-mediated photodynamic therapy (Ce6-PDT) against three strains of C. acnes. Using minimal Ce6 concentrations (0.49–0.89 µg/mL) in combination with halide light at 30,000 lx (emission spectrum 600–900 nm), they achieved over 99% growth inhibition (MIC99) of all tested strains. The antimicrobial activity of Ce6-PDT was found to be superior to that of 5-aminolevulinic acid-mediated PDT.
Ce6 alone exhibited inherent antimicrobial activity with MIC99 values ranging from 7.81 to 31.25 µg/mL against the selected C. acnes strains. In contrast, 5-aminolevulinic acid, at equivalent concentrations, showed no inhibitory effect against either C. acnes or Staphylococcus aureus. Similarly, de Annunzio et al.61. reported a concentration- and fluence-dependent antimicrobial effect of Ce6-PDT against C. acnes using Ce6 at 2.62 µM (approximately 15.63 mg/dL) in combination with RL at 3.25 J/cm2 (peak emission spectrum at 660 nm, 153 mW/cm2), achieving a reduction of 8.2 log10 CFU/mL in bacterial viability.
The study further demonstrated that Ce6-PDT produced comparable antibacterial effects to curcumin-mediated PDT (curcumin 14 µM, approximately equivalent to 0.516 mg/dL and RL at 3.12 J/cm2), while being superior to methylene blue-mediated PDT (methylene blue 156 µM, approximately 4.99 mg/dL, and RL 90 J/cm2).
In the present study, we demonstrated the efficacy of sodium iron CH-PDT in eradicating 10 clinically isolated, drug-resistant strains of C. acnes. These isolates were obtained from a biofilm established during a cross-sectional study12 at the dermatology clinic of a university hospital. The study focused on cases of AV that were unresponsive to standard antibiotic treatments.
All selected isolates demonstrated resistance to one or more commonly used antibiotics for AV treatment (Table 1). CHL-Fe was selected as the in-house PS at our centre due to its stability48 and superior production of various ROS, particularly 1O2* and hydroxyl radical (HO•), following visible light (VIS) irradiation, compared with other metal chlorophyllin complexes (CHL-M), including magnesium and copper derivatives38. The possible mechanism of action of CH-PDT primarily involves the type II photochemical pathway, as CHL-Fe predominately produces 1O2* opposed to HO• upon VIS irradiation.
Furthermore, regarding Fenton chemistry, Fe2+ (ferrous iron) reacts with hydrogen peroxide to generate additional HO•, potentially contributing to the overall oxidative damage62. Another advantage of chlorophylls and their derivatives is their lipophilic or amphiphilic nature, which facilitates preferential localization in lipid-rich environments. This promotes prolonged 1O2* lifetimes, enhancing both the efficiency and persistence of oxidative damage to targeted cells, especially within sebaceous follicles.
Extended lifetimes enable 1O2* to diffuse further from its point of generation63, expanding the area of oxidative damage to include neighboring structures such as sebaceous glands, microbial membranes, and biofilms.
Microorganisms with reduced antibiotic susceptibility often require higher PS concentrations or RL fluences to achieve microbial inactivation64, as the quantity of ROS generated is dependent on both the PS concentration and light fluence38. We investigated the individual and combined effects of CHL-Fe (up to 15 mg/dL) and RL fluences (up to 60 J/cm2) on pan drug-resistant C. acnes strain 1 (PDR-CA1). Neither CHL-Fe alone nor RL alone, at any dose tested, exhibited bactericidal activity.
However, CH-PDT using CHL-Fe at 7 mg/dL in combination with the lowest RL fluence tested (3.5 J/cm2) effectively eradicated PDR-CA1. This fixed-dose combination also demonstrated consistent efficacy across all tested strains, regardless of their resistance classification from drug-resistant (DR-CA) to pan drug-resistant (PDR-CAs). A previous study using Ce6-PDT achieved an 8.2 log10 CFU/mL reduction in C. acnes viability with a higher concentration of Ce6 (15.63 mg/dL) and a similar RL fluence (3.25 J/cm2), although the antimicrobial susceptibility of the tested strain was not reported61.
Our results suggest that CHL-Fe may offer superior or comparable antimicrobial efficacy at a lower concentration (Table 6), supporting its potential as an effective PS for clinical PDT applications in drug-resistant C. acnes.
Table 6Comparison of the efficacy of antimicrobial photodynamic therapy using various photosensitizers across different protocols reported in studies over the past 15 years.YearAuthorsTarget microorganismPhotosensitizersLight propertyEfficacyTarget microorganismsNumber of strainsAntimicrobial susceptibility resultsSource of microorganismsType of photosensitizerPeak excitation spectrum (nm)Concentration (mg/dL)Light sourcesEmission spectrum (nm)Light doses (J/cm2)Intensity (mW/cm2)/Wideband light settingsMIC99bAnti-biofilm effectsPhotodynamic therapyPS Alone (mg/dL)Light Alone(J/cm2)PS concentration (mg/dL)Light fluence (J/cm2)Our studyMahajindawong et al.PDR C. acnes2YesClinical isolatesSodium iron chlorophyllin404, 629a0, 1.875, 3.75, 7.5, 15Red light LEDs6330, 3.75, 7.5, 15, 30, 45, and 60657.53.75No bactericidal activityNo bactericidal activityNDXDR C. acnes1404, 6290, 7.56330, 3.757.53.75No bactericidal activityNo bactericidal activityNDMDR C. acnes4404, 6296330, 3.757.53.75No bactericidal activityNo bactericidal activityNDDR C. acnes1404, 6296330, 3.757.53.75No bactericidal activityNo bactericidal activityND2022Wang T et al., 2022P. acnes ATCC69191NDLaboratory strainsHexyl-amino levulinicacid405, 6355 and 50Red light LEDs63010860NDNDNDNDSignificantly disrupted thebiofilm structure to a greater extent than observed forthe other groups2021Barroso RA et al., 2021P.acnes (ATCC6919)1NDLaboratory strainsHypericin480, 5600.5, 1.5Red light LEDs660100, 166NDNDNDNDNDSignificantly reduced thebiofilm formation to a greater extent than observed forthe other groups2019Bui HTH et al., 2019P. acnes (VTCC0218)1NDLaboratory strainsChlorin e6 trimethylester403, 6570.025, 0.05; 0.075; 0.1;0.125; 0.15; 0.175; 0.20; 0.225; 0.25; 0.275; 0.30Halogen light600–90030 min12 V/50 W,30,000 lx0.12530 minNDNDNDS. aureus VTCC 01731NDLaboratory strains2018De Annunzio SR et al., 2018P. acnes(ATCC® 6919™)1NDLaboratory strainsChlorin-e65002.62, 5.25, 10.5, 21, 42µMRed light LEDs6603.25, 7.5, 1515315.633.25NDNDNDCurcumin45014, 27, 54µMBlue light LEDs455 and 4923.12, 6.25, 12.51510.5163.12NDNDNDMethylene blue66039, 78, 156µMRed light LEDs66030, 60, 901534.9990NDNDNDEnterococcus faecalis (ATCC®29,212™)1NDLaboratory strainsChlorin-e650020, 42, 84, 126, 168µMRed light LEDs66030, 60, 45, 90153250.5360NDNDNDCurcumin45017, 34, 68, 136µMBlue light LEDs455 and 49212.5, 18.75, 251512.5125NDNDNDMethylene blue66039, 78, 156, 312µMRed light LEDs66030, 60, 901539.9890NDNDND2018Choi SH et al., 2018P. acnes(American Type Culture Collection no. 6919)1NDLaboratory strainsIndocyanine green787, 650–8402.0, 0.5 or 0.1Yellow light LEDs8300, 2.5, 10, 40ND0.140No bactericidal activityNo bactericidal activityND2.0, 0.5 or 0.1Yellow light LEDs8050, 2.5, 10, 40ND0.140No bactericidal activityNo bactericidal activityNDMethyl aminolevulinate405, 6350.1, 0.5 and 2.0Red light LEDs6300, 2.5, 10, 40NDfrom 040No bactericidal activity40ND2015Jeon YM et al., 2015P. acnes(KCTC3314,KCTC3320,and KCTC5527)3NDLaboratory strainsChlorin e6402–405, 653–6620, 0.0015, 0.0031, 0.0061, 0.0122, 0.0244, 0.0488, 0.0977, 0.195, 0.395, 0.7813, 1.56, 3.12, 6.25, 12.5, 25, 50Halogen lamp600–90030 min12 V/50 W,30,000 lx> 0.049–0.09830 min> 0.781–3.125No bactericidal activityNDStaphylococcus aureus subsp. aureus (KCTC 1927)1NDLaboratory strains> 0.09830 min> 3.125No bactericidal activityNDP. acnes(KCTC3314,KCTC3320,and KCTC5527)3NDLaboratory strains5-aminolevulinicacid405, 635Almost no antibacterial effect against P. acnes and S. aureus using the same amount as Ce6No bactericidal activityNDStaphylococcus aureus subsp. aureus (KCTC 1927)1NDLaboratory strainsAlmost no antibacterial effect against P. acnes and S. aureus using the same amount as Ce6No bactericidal activityND2011Choi MS et al., 2011P. acnes (ATCC 6919)1NDLaboratory strains5-aminolevulinicacid4050.1Blue light LEDs4150, 3.125, 6.25, 12.5, 25, 50, 75, 100, 125eND0.125 cNo bactericidal activityNo bactericidal activity dNDP. acnes (ATCC 6919)1NDLaboratory strains6350.1Red light LEDs635NDNo bactericidal activityNo bactericidal activity up to 125 dNo bactericidal activityNo bactericidal activity dNDND Not demonstrated.a Supplementary Fig. 1.b MIC99 is the concentration of Ce6 at which microbial growth was inhibited by more than 99%. >3Log10 reduction.c Less cumulative doses when illuminated in consecutive days.d Different outcomes when illuminated in consecutive days.e Cumulative doses over 3 days.
The strengths of our study include the use of clinically isolated, highly drug-resistant C. acnes strains with documented antimicrobial susceptibility profiles for the 5 most commonly prescribed antibiotics for AV. This approach enhances the clinical relevance of our findings, particularly for referral centers in Thailand and potentially across broader regions in Asia. Similar to many countries65–68 in Asia, we are witnessing a concerning increase in antimicrobial-resistant pathogens associated with common skin infections, including dermatophytosis, onychomycosis, and AV, rendering these conditions increasingly difficult to treat.
Another strength of this study is the meticulous and stepwise evaluation of CHL-Fe alone, RL alone, and their combinations across various concentrations and fluences. This methodological approach allowed us to demonstrate that CH-PDT exhibited potentially superior efficacy compared to Ce-6-PDT in eradicating drug-resistant C. acnes strains. A primary limitation of our study is that all C. acnes isolates were obtained from Thai patients treated at a single university teaching hospital12.
Therefore, the optimal lethal CHL-Fe concentration and RL fluences (7.5 mg/dL and 3.5 J/cm², respectively) may not be generalizable to other geographic regions where C. acnes strains may vary in genetic characteristics, virulence, and antimicrobial susceptibility profiles.
Our study demonstrated the effective eradication of highly drug-resistant C. acnes isolates in a dose-dependent manner using CHL-Fe mediated PDT, regardless of their antibiotic resistance profiles. A single treatment dose of CHL-Fe at 7.5 mg/dL combined with RL at 3.5 J/cm2 was sufficient to achieve bactericidal effects in laboratory experiments. In addition to its application in AV treatment, CHL-Fe-mediated PDT holds promise for broader medical use, particularly in eradicating C. acnes colonization to prevent opportunistic infections.
Further research is needed to explore its potential role in infection prophylaxis across various surgical and implant-related fields.
Methods
Bacterial collection and antibiotic susceptibility test
The study protocol was approved by the Institutional Research Committee, Faculty of Medicine, Chulalongkorn University (IRB No. 1037/64, MDCU-IBC004/2022), and conducted in accordance with relevant guidelines and regulations, including the Declaration of Helsinki (2008). C. acnes isolates were obtained from a previous study between March 2019 and April 202012 at the dermatology clinic of King Chulalongkorn Memorial Hospital.
Informed consent was obtained from all subjects and/or their legal guardians prior to sample collection. Samples were collected using sterile techniques from facial lesions, such as inflammatory papules or pustules, closed comedones, and open comedones, in patients with mild to severe AV. The C. acnes isolates were cultured and identified under anaerobic conditions.
Antimicrobial susceptibility profiles were determined following the Clinical and Laboratory Standards Institute guidelines for tetracycline and clindamycin resistance and the European Committee on Antimicrobial Susceptibility Testing guidelines for doxycycline, erythromycin, and TMP/STX resistance. All C. acnes isolates were preserved in tryptic soy broth (Thermo Fisher Scientific) supplemented with 20% glycerol and stored at the Microbiology Laboratory, Department of Microbiology, Chulalongkorn University.
Bacterial strain selection and culture conditions
Ten strains of drug-resistant C. acnes were randomly selected from the database using a computer-based random number generator. The selection included one drug-resistant strain (DR-CA), five multidrug-resistant strains (MDR-CA1, MDR-CA2, MDR-CA3, MDR-CA4, and MDR-CA5), two extensively drug-resistant strains (XDR-CA1 and XDR-CA2), and two pan-drug-resistant strains (PDR-CA1 and PDR-CA2), as shown in Table 1. Each strain was cultivated on Brucella blood agar plates (BD BBL) and incubated anaerobically using the AnaeroPack system (Mitsubishi Gas Chemical Company, Tokyo, Japan) at 37 °C for 48 h.
Subsequently, a bacterial suspension was prepared in sterile 0.9% NaCl solution to achieve a McFarland turbidity standard of 0.5, corresponding to approximately 1.5 × 108 CFU/mL.
Photosensitizer Preparation and red-light source
Sodium iron chlorophyllin (CHL-Fe; TAMA Biochemical, Tokyo, Japan) was dissolved in sterile normal saline and stored in the dark at 25 °C. The maximal absorption wavelengths of CHL-Fe were observed at 404 and 629 nm (Supplementary Fig. 1), and these absorbance spectral properties remained stable for at least 30 days postpreparation (Supplementary Fig.
2). All procedures involving CHL-Fe were conducted in a dark room under dim yellow light (λ = 596 nm) to minimize premature PS activation. An incoherent RL source from light-emitting diodes (HEALITE II®, Lutronic Corp., Boston, MA USA, and Goyang, South Korea) with a peak emission at 633 nm and a power intensity of 65 mW/cm2 at an effective distance of 4 cm was used for illumination.
The device is calibrated annually by Thailand’s National Institute of Metrology to ensure the accuracy of spectral irradiance, emission stability, and uniformity across the treatment area. The emission spectrum and peak wavelength were determined using relative spectral irradiance measurements based on the comparison of irradiance values at the peak wavelength. These measurements were performed using a standard spectroradiometer capable of acquiring spectral irradiance data across the wavelength range of 220–1,022 nm.
Effect of CHL-Fe concentration on C. acnes population (Fig. 1)
The CHL-Fe solution was diluted with sterile 0.9% NaCl to achieve final concentrations of 1.875, 3.75, 7.5, and 15 mg/dL. A 50 µl aliquot of CHL-Fe was added to the wells of a 96-well microtiter plate for each concentration. C. acnes isolates were adjusted to the 0.5 McFarland standard using sterile 0.9% NaCl.
Then, 50 µl of the C. acnes suspension was added to each well containing the different CHL-Fe concentrations, resulting in CHL-Fe/C. acnes (CH/Ca) mixtures. A dilution of C. acnes with 0.9% NaCl alone was used as the negative control. The CH/Ca mixtures were incubated at 37 °C for 10 min.
The viable bacterial count (CFU/mL) was then determined by plating the mixtures onto Brucella agar plates with 5% sheep blood (BD BBL) and incubating them at 37 °C for 48 h in a dark and anaerobic environment.
Effect of red light on the C. acnes population (Fig. 1)
A 50 µl aliquot of each C. acnes suspension (0.5 McFarland standard), prepared in sterile 0.9% NaCl without CHL-Fe, was transferred into a transparent 96-well plate. The plates were placed 4 cm below the illumination head of an incoherent red light-emitting diode (HEALITE II®, Lutronic Corp., Boston, MA USA, and Goyang, South Korea) with a peak emission spectrum at 633 nm and an intensity of 65 mW/cm2. Irradiations were performed using light fluences of 3.75, 7, 15, 30, 45, and 60 J/cm2, corresponding to exposure durations of 0.56, 1.47, 3.50, 7.41, 11.32, and 15.23 min, respectively.
Each plate was covered and exposed to only one specific fluence. Bacterial survival was determined immediately following irradiation by assessing CFU/mL.
Effect of CH-PDT on the C. acnes population
To develop an optimal CH-PDT protocol using the minimum effective concentration–fluence dose combination, three sequential experiments were conducted. Direct pH monitoring prior to irradiation was not performed; however, all treatments were performed using freshly prepared CHL-Fe in 0.9% NaCl to maintain consistent experimental conditions. All treatments were conducted under static conditions to ensure uniform light exposure across the surface of the bacterial suspensions.
In Experiment 1, the viability of PDR-CA1 strain was assessed following treatment with varying CHL-Fe concentrations (0, 1.875, 3.75, 7.5, and 15 mg/dL), RL fluences (0, 15, 30, 45, and 60 J/cm2), and combination treatments (CH-PDT) with CHL-Fe concentrations of 0, 3.75, 7.5, and 15 mg/dL and light fluences of 0, 3.75, 7.5, and 15 J/cm2.
Experiment 2 involved testing PDR-CA2 with the same CH-PDT combinations as in Experiment 1—CHL-Fe concentrations of 0, 3.75, 7.5, and 15 mg/dL and RL fluences of 0, 3.75, 7.5, and 15 J/cm2 —to validate reproducibility and determine the minimum lethal dose.
In Experiment 3, the efficacy of CH-PDT was assessed on the remaining 8 C. acnes strains (DR-CA, MDR-CA1, MDR-CA2, MDR-CA3, MDR-CA4, XDR-CA1, and XDR-CA2) using the minimum lethal dose identified in Experiment 2: CHL-Fe 7.5 mg/dL combined with RL 3.75 J/cm2.
Assessment of post-treatment viability and determination of lethal doses
Following a 10 min incubation at 37 °C, the treated C. acnes suspensions were transferred from the wells, serially diluted, and plated onto Brucella agar supplemented with 5% sheep blood (BD BBL). Viable bacterial counts in CFU/mL were determined after 48 h of anaerobic incubation at 37 °C. All experiments were conducted in triplicates in a dark room under dim yellow light (λ = 596 nm) to minimize premature activation of the photosensitizer.
The lethal dose was defined as the treatment condition that caused sufficient direct damage to destroy the majority of the viable bacterial population. For this study, a lethal treatment dose was considered to be a reduction of more than 3 log10 (≥ 99.9%) in CFU/mL compared to the untreated C. acnes control.
Statistical analysis
The log10 mean changes were compared between treatment groups and their corresponding controls using the paired t-test. Data were expressed as log10 means with 95% CIs. All statistical analyses were conducted using Stata version 13.1 (StataCorp, College Station, Texas, USA).
A p-value of < 0.05 was considered statistically significant.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1