(~)
`......
`doi: 10.1111/j.1346-8138.2011.01321.x
`
`THE JOURNAL OF
`
`DERMATOLOGY
`.Journal of Dermatolo£y 2012; 39:350-354
`
`ORIGINAL ARTICLE
`Glycolic acid chemical peeling improves inflammatory acne
`eruptions through its inhibitory and bactericidal effects on
`Propionibacterium acnes
`
`Yuko TAKENAKA,1 Nobukazu HAYASHI,2 Mikiko TAKEDA,3 Sayaka ASHIKAGA,3
`Makoto KAWASHIMA1
`
`1Department of Dermatology, Tokyo Women’s Medical University, 2Department of Dermatology, Toranomon Hospital, Tokyo, and
`3Department of Research and Development, Rohto Pharmaceutical, Osaka, Japan
`
`ABSTRACT
`
`Glycolic acid chemical peeling is effective for treating comedones, and some clinical data show that it also improves
`inflammatory eruptions. The purpose of this study was to identify the mechanism of glycolic acid chemical peeling to
`improve inflammatory acne. To assess growth inhibitory and bactericidal effects of glycolic acid on Propionibacterium
`acnes in vitro, we used an agar diffusion method and a time-kill method. To reveal bactericidal effects in vivo, we estab-
`lished an agar-attached method which correlated well with the ordinary swab-wash method, and we used the agar-
`attached method to compare the numbers of propionibacteria on the cheek treated with glycolic acid chemical peeling.
`Our results show that 30% glycolic acid (at pH 1.5, 3.5 and 5.5) formed growth inhibitory circles in the agar diffusion
`method, but the diameters of those circles were smaller than with 1% nadifloxacin lotion or 1% clindamycin gel. In the
`time-kill method, 30% glycolic acid (at pH 1.5 and 3.5) or 1% nadifloxacin lotion reduced the number of P. acnes to less
`than 100 CFU/mL within 5 min. In contrast, in 30% glycolic acid (at pH 5.5) or in 1% clindamycin gel, P. acnes survived for
`more than 4 h. Chemical peeling with 35% glycolic acid (at pH 1.2) decreased the number of propionibacteria on the
`cheeks of patients compared with untreated controls (P < 0.01). Our results demonstrate that glycolic acid has moderate
`growth inhibitory and bactericidal effects on P. acnes, and that chemical peeling with glycolic acid works on inflammatory
`acne via those effects.
`
`Key words: acne vulgaris, agar-attached method, chemical peeling, glycolic acid, Propionibacterium acnes.
`
`INTRODUCTION
`
`Acne is a chronic inflammatory skin disease which is experienced
`by more than 90% of the population1 and which affects emotional
`aspects of acne patients’ quality of life.2 The primary eruption of
`acne is a comedone, which is induced by abnormal cornification of
`the hair infundibulum and hypersecretion of sebum. Propionibac-
`terium acnes, which prefers anaerobic and lipid-rich conditions,
`increases in number and induces inflammation.3 In consequence,
`comedones develop into papules and pustules? Severe inflamma-
`tion sometimes causes cystic and/or nodular acne and forms atro-
`phic or hypertrophic scars, which remain permanently,
`Glycolic acid chemical peeling is accepted as a useful modality
`to treat acne.5’6 It improves not only comedonal acne but also
`inflammatory acne.7’8 The proposed mechanism of glycolic acid
`chemical peeling is usually thought to be the correction of abnormal
`keratinization in the infundibulum, which works directly on comedo-
`nes and consequently improves inflammatory eruptions. However,
`
`experiments performed by Atzori et aLr and by our group8 showed
`significant improvement of inflammatory eruptions from the first
`application. We assume that glycolic acid might have some bacteri-
`cidal effects which improve inflammatory acne eruptions. The aim
`of our study was to characterize the bactericidal effects of glycolic
`acid in vitro and in vivo.
`
`METHODS
`
`Patients
`Acne patients of all severities were enrolled for comparison of the
`agar-attached method with the swab-wash method. Patients with
`mild inflammatory acne, who had not received any oral or topical
`antibiotics for at least 1 month before participation, were enrolled
`for the chemical peeling experiments. These experiments were
`approved by the ethical committee of the Tokyo Women’s Medical
`University, and each patient received a full explanation and signed a
`written informed consent form before the experiments.
`
`Correspondence: Yuko Takenaka, M.D., Department of Dermatology, Tokyo Women’s Medical University, 8 1 Kawada cho, Shinjuku ku, Tokyo
`162 8666, Japan. Email: y takenaka@derm.twmu.ac.jp
`Received 21 April 2011; accepted 22 May 2011.
`
`350
`
`© 2011 Japanese Dermatological Association
`
`1 of 5
`
`Almirall EXHIBIT 2035
`Amneal v. Almirall
`IPR2018-00608
`
`

`

`Glycolic acid antibacterial effect on P. aches
`
`Materials
`Glycolic acid (Sigma-Aldrich, St Louis, MO, USA) was dissolved in
`distilled water and the pH was adjusted with 1 N NaOH. Thirty per-
`cent glycolic acid (at pH 1.5, 3.5 and 5.5) was used for the in vitro
`study. One percent nadifloxacin lotion (Otsuka Pharmaceutical,
`Tokyo, Japan) and 1% clindamycin gel (Sato Pharmaceutical,
`Tokyo, Japan) were purchased. Commercially available materials
`for glycolic acid chemical peeling, including 35% glycolic acid
`(pH 1.2) and the neutralizer, were obtained from Tokiwa Pharma-
`ceutical (Osaka, Japan). Anaerobic conditions were generated in
`small sealed bags with Anaeromeit-J (Nissui Pharmaceutical,
`Tokyo, Japan) which chemically absorbs oxygen. P. acnes (ATCC
`no. 6919) for in vitro study was purchased from the American Type
`Culture Collection (Rockville, MD, USA).
`
`Evaluation of the bactericidal effects of glycolic acid
`chemical peeling on P. acnes in vivo
`After washing the entire face of each subject with soap, microorgan-
`isms were obtained from both cheeks using the agar plate method.
`A randomly selected side of the cheek of each patient was treated
`with 35% glycolic acid (pH 1.2) for 2 min or less until erythema
`appeared, then the reaction was stopped with a neutralizer. After
`washing the entire face with enough water and cooling for 2 min
`with a paper mask soaked with icy water, microorganisms on the
`skin surface were obtained again in the same way. All agar plates
`were incubated in anaerobic conditions at 37°C for 48 h. The num-
`bers of colonies were then counted and decreasing rates of colonies
`were calculated to statistically compare untreated and treated sides
`with glycolic acid chemical peeling using the paired Student’s t-test.
`
`Growth inhibitory effects of glycolic acid in vitro
`The agar diffusion method, which is the modified disk diffusion
`method, was used to assess growth inhibitory effects. Bacterial
`suspensions, which contained 1 x 106 CFU/mL P. acnes, were
`inoculated onto each Gifu Anaerobic Medium (GAM) plate. Eight-
`millimeter diameter holes were formed on each plate and 0.1 g of
`each tested material was placed in it. After incubation in anaerobic
`conditions at 37°C for 48 h, the diameter of each growth inhibitory
`circle was measured,
`
`Bactericidal effects of glycolic acid in vitro
`The time-kill method was used to estimate the bactericidal effects
`of glycolic acid. We incubated 1 x 106 CFU/mLP. acnes with each
`tested material for 5 min, 30 min, 60 min and 4 h. These samples
`were then diluted with 0.9% sodium chloride, and inoculated onto
`agar plates. After incubation in anaerobic conditions at 37°C for
`48 h, the numbers of surviving colonies were counted,
`
`Establishment of the agar-attached method
`To measure the number of P. acnes on the cheeks of each subject,
`we used an agar-attached method, which is a modified stamp
`method, to count bacteria on the skin surface. We attached an agar
`plate (Petan-Check; Eiken Chemical, Tokyo, Japan) on the cheek for
`5 s, and then incubated it in anaerobic conditions at 37°C for 48 h,
`after which the numbers of colonies on each plate were counted,
`To confirm the reliability of the agar-attached method for count-
`
`ing P. acnes on the skin of acne patients, we compared it with the
`swab-wash method, which is ordinarily used to evaluate propioni-
`bacteria related with acne symptoms. For the swab-wash method,
`4 cm2 of each patient’s cheek adjacent to the place where the sam-
`pie for the agar-attached method was taken, was scrubbed with a
`sterile swab moistened with phosphate-buffered saline (pH 7.4)
`containing 0.05% Tween-20 (Sigma-Aldrich). The swab-tip was
`broken off into the wash fluid, and mixed vigorously, after which the
`wash fluid was serially diluted with distilled water. One hundred
`microliters of the diluted and undiluted wash fluids were spread over
`the surface of the GAM plates, and after incubation in anaerobic
`conditions at 37°C for 3 days, the numbers of colonies were
`counted. Logarithms of the numbers of colonies obtained from the
`swab-wash method and the agar-attached method were calcu-
`lated, and their correlations were assessed,
`
`RESULTS
`
`Growth inhibitory effects of glycolic acid in vitro
`Growth inhibitory circles were observed in all plates treated with
`30% glycolic acid, and the lower pH of 30% glycolic acid showed
`the greater inhibitory effect (Fig. 1). The averages of triplicate results
`of zone-of-inhibition diameters are summarized in Figure 2. The
`clindamycin gel and the nadifloxacin lotion showed larger zone-
`of-inhibition diameters than all 30% glycolic acid solutions.
`
`Bactericidal effects of glycolic acid in vitro
`The effects of glycolic acid on P. acnes growth are summarized in
`Table 1. The numbers of P. acnes were reduced to less than
`100 colony-forming units/mL within 5 min when treated with 30%
`glycolic acid at pH 1.5, 3.5 or with 1% nadifloxacin lotion. In con-
`trast, 30% glycolic acid at pH 5.5 or 1% clindamycin gel did not eli-
`cit any apparent decrease in colony counts, meaning that those two
`materials did not have a bactericidal effect.
`
`Comparison between the swab-wash method and
`the agar-attached method
`Thirty-two patients were enrolled in this part of the study. The Ioga-
`rithms of numbers of colonies obtained from their cheek skin using
`the swab-wash method and the agar-attached method are shown
`in a scatter graph (Fig. 3). The correlation rate was 0.72 (P < 0.01),
`which is an excellent correlation and confirmed the reliability of the
`agar-attached method.
`
`Effects of glycolic acid chemical peeling on
`P. acnes in vivo
`Nine female acne patients, aged 25-35 years, were enrolled in this
`part of the study. There were no apparent differences in the num-
`bers of colonies of propionibacteria taken from the left or right
`sides of the faces before chemical peeling. After the glycolic acid
`chemical peeling, the colonies on the treated sides dramatically
`decreased compared with the untreated sides. Representative
`results are shown in Figure 4. Each treated side had much lower
`numbers of colonies compared with the untreated side, and as a
`consequence, the decreases on the treated sides were statistically
`higher than on the untreated sides (P < 0.01) (Fig. 5). Furthermore,
`we picked up seven of these colonies from patients’ agar plates,
`
`© 2011 Japanese Dermatological Association
`
`351
`
`2 of 5
`
`

`

`Y. Takenaka et al.
`
`Figure 1. Agar diffusion method. All plates with 30% glycolic acid formed growth inhibition circles to Propionibacterium acnes. (a) pH 1.5,
`(b) pH 3.5, (c) pH 5.5.
`
`,°
`
`~Cg% gIy~ohc ~og gl’~collc
`ac,d~olu~on
`a~Idsolution
`]pH 1.5]
`ItoH 3-~I
`
`~OK glycol~:
`acid~olu~on
`
`1~
`
`1%
`
`nacllllo~acm
`
`die,dam~t~_i~
`
`IPH5.51
`
`lalion
`
`gel
`
`I~
`
`a’$i
`
`3~-
`
`E 2
`
`!,
`il
`1
`
`o.5
`]
`
`0"~
`
`p = ,[},’,g2
`
`~5.
`
`~
`
`I’5
`
`~’’
`
`2~5
`
`[!. 1(~I"
`
`Aaalr alL’LIl~lled II~e~Od
`
`(cid:128)
`
`2O
`
`[,o
`,
`
`~3 0
`
`Figure 2. Growth inhibitory effect of glycolic acid. Solutions of 30%
`glycolic acid (pH 1.5, 3.5 and 5.5) formed zones of inhibition with
`diameters smaller than those elicited by clindamycin gel or nadifloxa-
`cin lotion.
`
`and confirmed that all of them are P, acnes using Gram staining,
`catalase test and Rap ID ANA II system (AMCO, Tokyo, Japan).
`
`DISCUSSION
`
`Our clinical study on the efficacy of 30% glycolic acid chemical
`peeling for acne vulgaris revealed the rapid improvement not only of
`
`Figure 3. Correlation between the swab-wash method and the
`agar-attached method. A scatter graph of logarithms of numbers of
`Propionibacterium acnes obtained using the swab-wash method and
`the agar-attached method shows a significant correlation of 0.72
`(P < 0.01).
`
`comedones but also of inflammatory eruptions, which prompted us
`to study the antibacterial effect of glycolic acid.
`First of all, we performed an in vitro study to check the antimicro-
`bial effects of glycolic acid. The agar diffusion method indicated that
`all of the examined glycolic acids have growth inhibitory effects
`
`Table 1. Bactericidal effect of glycolic acid. The numbers of Propionibacterium acnes were reduced to <100 CFU/mL within 5 min after
`treatment with 30% glycolic acid (at pH 1.5 or 3.5)
`
`30% glycolic acid solution (pH 1.5) (CFU/mL)
`30% glycolic acid solution (pH 3.5) (CFU/mL)
`30% glycolic acid solution (pH 5.5)
`1% nadifloxacin lotion (CFU/mL)
`1% clindamycin gel
`
`CFU, colony-forming units.
`
`352
`
`5 min
`
`<100
`<100
`1.1 x 106
`<100
`4.4 x 106
`
`30 min
`
`<100
`<100
`3.3 x 105
`<100
`3.8 x 105
`
`60 min
`
`<100
`<100
`3.3 x 105
`<100
`3.3 x 105
`
`4 h
`
`<100
`<100
`3.2 x 105
`<100
`2.2 x 105
`
`© 2011 Japanese Dermatological Association
`
`3 of 5
`
`

`

`Glycolic acid antibacterial effect on P. acmes
`
`Case 1
`
`Case 2
`
`Figure 4. Bactericidal effect of glycolic acid chemical peeling. Representative results showing the bactericidal effects of glycolic acid chemical
`peeling. (a,e) Treated sides before the experiment; (b,f) treated sides after the experiment; (c,g) untreated sides before the experiment;
`(d,h) untreated sides after the experiment.
`
`~
`
`M~,.
`
`~ ~ 4 ~ ~
`
`~o
`
`g
`-o
`
`ioo ~7~ 974
`
`0
`
`DNA gyrase, has both growth inhibitory and bactericidal effects.
`Based on these studies, we confirmed the effects of glycolic acid
`on P. acnes, and lower pH of glycolic acid has more growth inhibi-
`tory and bactericidal effects, although we still do not know how it
`works.
`Next, we estimated the antimicrobial effects of glycolic acid
`in vivo. The swab-wash method is generally used to count colonies
`10
`of propionibacteria on the skin surface to determine the numbers
`of propionibacteria quantitatively on the skin of acne patients. This
`method requires complicated steps. You have to make special
`wash-fluid, to collect samples, to dilute them, to sow them on agar
`plates, to culture them in anaerobic condition and to count the num-
`bers of colonies. On the other hand, the newly established agar-
`’°° ’°° ¯ P< 0.01 attached method requires only three steps to attach the plate on the
`cheeks, to culture them in anaerobic condition and to count the
`colonies. To determine the amount of P. acnes in infundibulum is
`important and the agar-attached method estimates only propioni-
`bacteria on the surface. Our experiments revealed that the agar-
`attached method is reliable because it correlates well with the
`swab-wash method which has already been established to measure
`the quantities of microbacteria in acne patients. The agar-attached
`method is much easier and more convenient than the swab-wash
`method.
`Our results using the agar-attached method clearly show
`that 35% glycolic acid (pH 1.2) chemical peeling reduces the
`number of propionibacteria on the cheeks of acne patients.
`These results might come from the effects of acidic condition,
`but the most important point is that glycolic acid in practical
`concentration is safe and effective for inflammatory acne and
`comedonal acne.
`
`Figure 5. Decreasing rate of colonies after glycolic acid chemical
`peeling. The bars show the decreasing rates of colonies of both sides
`of the faces of all of nine cases and their mean. The decreasing rate
`of the treated side was statistically higher than the untreated side
`(P < 0.01).
`
`on P. acnes, but its effects are milder than the effects of topical
`antibiotics. The time-kill method revealed that 30% glycolic acids
`(at pH 1.5 and 3.5) have a bactericidal effect. In this study, the
`clindamycin gel did not show any bactericidal effect on P. acnes,
`Clindamycin works on bacteria by binding preferentially to the 23 S
`subunit of the bacterial ribosome and inhibiting bacterial protein
`synthesis.9 As a consequence, the mechanism of action of clinda-
`mycin is not bactericidal, but is bacteriostatic. On the other hand,
`nadifloxacin, which inhibits the configuration of supercoiled DNA by
`
`© 2011 Japanese Dermatological Association
`
`353
`
`4 of 5
`
`

`

`Y. Takenaka et al.
`
`In consequence, glycolic acid has antimicrobial effect on
`
`P, acnes in vitro and reduces the numbers of P, acnes on the
`
`cheeks of acne patients in vivo. Recently, reports of antibiotic-
`resistant P, acnes are increasing.11-14 To prevent antibiotic-resis-
`
`tant bacteria, authorities recommend reducing the widespread
`
`use of antibiotics, not to use topical and oral antibiotics simulta-
`
`neously and to use benzoyl peroxide (BPO). But BPO has some
`
`skin irritation and is not permitted for acne in some countries
`
`including Japan. Our data suggest that glycolic acid could be
`
`one alternative to BPO to reduce antibiotic-resistant P, acnes,
`
`especially when BPO is not available,
`
`REFERENCES
`
`1 Hayashi N, Kawashima M, Watanabe S et al. An epidermiological study of
`acne vulgaris in Japan by questionnaire. Jpn J Dermatol B 2001 ; 111 :
`1347-1335. (In Japanese.)
`2 Hayashi N, Higaki Y, Kawamoto K, Kamo T, Shimizu S, Kawashima M. A
`cross-sectional analysis of quality of life in Japanese acne patients using
`the Japanese version of Skindex-16. J Dermato12004; 31 : 971-976.
`3 Kurokawa I, Willian Danby F, Qiang Ju et aL New developments in our
`understanding of acne pathogenesis and treatment. Exp Dermato12009;
`18:821-832.
`4 Do "FI’, Zarkhin S, Orringer JS et aL Computer-assisted alignment and
`tracking of acne lesions indicate that most inflammatory lesions arise from
`comedones and de novo. J Am Acad Dermato12008; 58: 603-608.
`
`5 Kim SW, Moon SE, Kim JA, Eun HC. Glycolic acid versus Jessner’s solu-
`tion: which is better for facial acne patients? A randomized prospective
`clinical trial of split-face model therapy. Dermatol Surg 1999; 25:
`270-273.
`6 Wang CM, Huang CL, Hu CT, Chan HL. The effect of glycolic acid on the
`treatment of acne in Asian skin. DermatolSurg 1997; 23: 23-29.
`7 Atzori L, Brundu MA, Orru A, Biggio P. Glycolic acid peeling in the treat-
`ment of acne. J EurAcadDermatol Venereo11999; 12:119-122.
`8 Hayashi N, Kawashima M. The usefulness of chemical peeling with 30%
`glycolic acid (pH 1.5) for acne vulgaris. Jpn J Clin Dermatol 2003; 13:
`1213-1216.(In Japanese.)
`9 Ross JI, Eady AE, Cove JH et al. Clinical resistance to erythromycin and
`clindamycin in cutaneous Propionibacterium isolated from acne patients
`is associated with mutations in 23S rRNA. Antimicrob Agents Chemo-
`ther 1997; 41:1162-1165.
`10 Keyworth N, Millar MR, Holland KT. Swab-wash method for quantitation
`of cutaneous microflora. J Clin Microbio12003; 28:941-943.
`11 Kurokawa I, Nishijima S, Kawabata S. Antimicrobial susceptibility of Propi-
`onibacterium acnes isolated from acne vulgaris. Eur J Dermatol 1999; 9:
`25-28.
`12 Ishida N, Nakaminami H, Noguchi N, Kurokawa I, Nishijima S, Sasatsu M.
`Antimicrobial susceptibilities of Propionibacterium acnes isolated from
`patients with acne vulgaris. Microbiol Immuno12008; 52:621-624.
`13 Eady EA, Gloor M, Leyden JJ. Propionibacterium acnes resistance:
`a worldwide problem. Dermatology 2003; 206: 54-56.
`14 RossJI, SnellingAM, EadyEAetaL Phenotypicand genotypiccharacter-
`ization of antibiotic-resistant Propionibacterium acnes isolated from acne
`patients attending dermatology clinics in Europe, the U.S.A., Japan and
`Australia. Br J Dermato12001 ; 144: 339-346.
`
`354
`
`© 2011 Japanese Dermatological Association
`
`5 of 5
`
`