J Med Dent Sci 2010； 57： 65－74

Original Article

Quantitative PCR of Propionibacterium acnes DNA in samples aspirated from sebaceous follicles on the normal skin of subjects with or without acne

Yuriko Miura1, Ikuo Ishige2, Naomi Soejima3, Yoshimi Suzuki3, Keisuke Uchida3, Seiji Kawana4 and Yoshinobu Eishi3

1) Fukasawa Dermatology Clinic, Kanagawa, Japan2) Division of Stem Cell Therapy, Center for Stem Cell and Regenerative Medicine, The Institute of Medical Science, University of Tokyo, Tokyo, Japan3) Department of Human Pathology, Graduate School of Medical Sciences, Tokyo Medical and Dental University, Tokyo, Japan4) Department of Dermatology, Nippon Medical School, Tokyo, Japan,

Corresponding Author: Yoshinobu Eishi, MD, MedScD, PhDDepartment of Human Pathology, Graduate School of Medical Sciences, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan; Tel: +81-3-5803-5964　Fax： +81-3-5803-0174E-mail: eishi.path@tmd.ac.jpReceived September 30；Accepted November 13, 2009

To elucidate whether people with hair follicles containing many Propionibacterium acnes cells are prone to acne, we developed a novel method to count the number of P. acnes in hair follicles. We sampled sebaceous material in hair follicles by aspiration at a constant negative pressure from the nose, forehead, and upper arm of 86 patients with acne vulgaris and 209 control subjects with healthy skin, including 84 subjects age-matched to the patients. Genome-equivalents of P. acnes in samples were estimated by real-time quantitative PCR (TaqMan). Numbers of P. acnes genome-equivalents were extremely low in control subjects less than 10 years of age and generally higher at greater ages, with much variation in subjects in the same decade of life. In men, the median count was highest in controls aged 15-19 years; in women, it peaked twice, in controls aged 15-19 years and again in those aged 40 years or older. P. acnes counts on the forehead and nose were higher in the acne patients aged 10-14 years than in the age-matched controls in both sexes. The counts at three sites were similar in acne patients and controls aged 15 to 29 years in both sexes.

The results suggest that people with hair follicles containing many P. acnes cells are not particularly prone to acne, except for younger teenagers. Our aspiration method with estimation by real-time PCR can be used to examine the cutaneous microflora of P. acnes.

Key words: acne, microflora, Propionibacterium acnes, quantitative PCR, sebaceous follicles.

Introduction

Propionibacteria are the only anaerobes known to be resident members of the cutaneous microflora of humans. Of the three species, Propionibacterium acnes is the most commonly found, being present in nearly 100% of adults. Findings of microbiologic studies of the skin are greatly influenced by the method used for sample collection. The techniques used in harvesting organisms for quantitative culture have included contact culture, flushing, stripping with adhesive tapes or films, scraping with a scalpel, biopsy, and scrubbing or scraping with a stiff brush or a spatula in a chamber pressed against the skin and containing a fluid. The studies of Williamson and Kligman1 and Shaw et al. 2 selected a scrubbing technique that yields a large number of organisms, and the results were fairly reproducible. In the method, a specimen was prepared by use of scrubbing, with culture of the specimen obtained, and counting of the colonies. In the new method introduced here, we used a

66 J Med Dent SciY. Miura et al.

constant negative pressure to aspirate the contents of sebaceous follicles within a specified time from skin of a specified surface area. The number of P. acnes cells recovered was estimated in terms of the number of genome-equivalents by real-time quantitative PCR (TaqMan). The novel method has two advances; one is the method of aspirating follicles to collect their bacterial contents and the other is the use of real-time PCR to quantify P. acnes genome-equivalents in the samples obtained. By the novel method, we sought to identify age-, sex-, and body-site-related differences in P. acnes colonization of the skin of healthy people, and we compared the bacterial density on the normal skin of subjects with and without acne. The main purpose of this paper is to elucidate whether people with hair follicles containing many P. acnes cells are prone to acne or not by the novel method.

Materials and methods

Subjects Eighty-six Japanese patients (39 male and 47 female subjects) with acne of the face attending our outpatient clinic for treatment between July 1998 and December 2000 were enrolled in the study (Table 1). Ages in the group with acne were 10 to 29 years old and patients were divided into three groups: those in early adolescence (10 to 14 years), those in late adolescence (15 to 19 years), and young adults (20 to 29 years). All acne patients in this study were classified as grade Ⅰ andⅡ, according to Kligman grade 3 that classify the severity of acne papulopustulosa from Ⅰ to Ⅳ. As controls, 209 healthy people (105 male and 104 female subjects) (Table 1), including 84 controls (41 male and 43 female subjects) age-matched to the patients, voluntarily agreed to participate in this study between 1998 and 2000. For controls, the absence of acne on the face and trunk were confirmed by inspection, and that the subject had not recently had acne was confirmed in interviews. Ages in the control group were 0 to 85 years old. The healthy adults were classif ied into age categories of one decade each. Antibiotics (including topical antibacterial agents) or other comedolytic agents were not given to any subject for the two weeks before the experiment. Informed consent for the study was obtained from participants, or from one of their parents when the subject was less than 10 years old, in accordance with the institute

guideline and ethical standards of the Helsinki Declaration.

Sample preparation Sampling procedure is shown in Fig. 1. The entire nose and 16 cm2 of the skin surface of the forehead (the area above the left eyebrow to the hair) and the extensor side of the upper part of the left arm were wiped for 10 s each with a swab moistened with alcohol. Next, the skin surface was uniformly aspirated for 60 s at a constant negative pressure of 40 kPa with a sebum aspirator for home use (‘Esthesienne’ pore aspirator ‘Spot Clear’ EH257, Matsushita Electric Works, Ltd., Tokyo). Acne sites were avoided; samples were aspirated in the area where no eruption (comedones, papules, or pusutules) was found. The matter collected in the opening of the aspirator was transferred to a sterile swab, and the tip of the swab was cut off and put into a 1.5-ml polyethylene tube. Next, 0.5 ml of a DNA extraction agent (Dexpat; Takara Shuzo, Kyoto, Japan) was added to the tube, and the mixture was heated at 100 ℃ for 10 min and centrifuged at 12,000 × g for 10 min at 4 ℃. The supernatant was poured into a sample tube, and 5μl of the sample was used in PCR.

Quantitative PCR P. acnes genome-equivalents were estimated by real-time quantitative PCR as described before.4 The primers for P. acnes were 5'-GCGTGAGTGACGGTAATGGGTA-3' and 5'-TTCCGACGCGATCAACCA-3', and the TaqMan probe was 5'-AGCGTTGTCCGGATTTATTGGGCG-3'. PCR was done in 50 μl of a mixture containing 5 μl of a DNA sample, 5 pmol of each primer needed, 2 pmol of a TaqMan probe, 10 nmol of each of the four deoxynucleotides, 175 nmol of MgCl２, 1.25 U of AmpliTaq Gold (PE Biosystems, Foster City, CA), and 1 × TaqMan buffer A (PE Biosystems) to make 50 μl. Amplification and detection were done with a detection system (GeneAmp 5700, PE Biosystems) with the profile of 95 ℃ for 5 min followed by 50 cycles of 95 ℃ for 15 s and 60 ℃ for 1 min. The P. acnes genome-equivalents in samples were estimated with an internal standard curve prepared with three replicates of four concentrations (10 ng, 1 ng, 100 pg, and 10 pg) of bacterial DNA extracted from P. acnes (ATCC 6919). The concentration of DNA was expressed in terms of the number of bacterial genome-equivalents, calculated as 2.5 × 109 daltons per genome-equivalent. The result of quantitative PCR for each sample was expressed as the number of P. acnes genome-equivalents in 5 μl

67Quantitative PCR of P. acnes DNA from the skin

from a total volume of 500 μl of each sample. The specificity of primers and probes for P. acnes was confirmed by the same procedures with 10 ng of bacterial DNA extracted from other species of propionibacteria including P. granulosum (ATCC 25564), P. avidum (ATCC 25577), and P. lymphophilum (ATCC 27520) and Staphylococcus epidermidis (ATCC 14990) (Data not shown).

Evaluation of16S rRNA operon copy number of P. acnes DNA extracted from two type strains of P. acnes (ATCC 6919 and 11828) and 29 strains of P. acnes isolated from normal skin of 29 healthy persons were

digested by restriction enzyme HindⅢ or KpnⅠ. The DNA fragments, separated by electrophoresis, were transferred to nylon membrane by capillary-blotting. The rRNA operon copy number was determined by southern blot hybridization with 32P-labeled DNA probe according to established protocols. 5 Labeled probe were prepared by labeling to 1078 bp-PCR fragment designed in P. acnes 16S rRNA DNA using Random Primer DNA labeling Kit (Takara bio Inc., Shiga, Japan) according to manufacturer’s instruction. HindⅢ and KpnⅠ have no cleavage site within this probe sequence.

Reproducibility Reproducibility of the results was examined by three

Age category (years) Sex Acne n Mean age (years, range) <10 Ma - 15 3 (0 - 9) F - 15 5 (0 - 9) 10 - 14 M + 10 13 (10 - 14) - 15 13 (10 - 14) F + 15 13 (11 - 14) - 12 13 (10 - 14) 15 - 19 M + 14 17 (15 - 19) - 14 18 (15 - 19) F + 17 18 (15 - 19) - 16 18 (16 - 19) 20s M + 12 25 (20 - 29) - 15 25 (21 - 29) F + 15 24 (20 - 29) - 15 24 (20 - 29) 30s M - 15 33 (30 - 39) F - 13 31 (30 - 34) 40s M - 15 42 (40 - 45) F - 15 45 (40 - 49) 50s M - 12 53 (50 - 57) F - 12 55 (50 - 57)

60 M - 7 71 (60 - 85) F - 6 66 (60 - 73)

aM, male; F, female. -, absent; +, present.

Table 1. Age distribution of healthy subjects and acne patients

68 J Med Dent SciY. Miura et al.

TaqMan measurements each of samples (one each) from five subjects with healthy skin, for intra-assay variation, and by assays of three samples taken from each of the five subjects, with two weeks intervening between samplings, for inter-assay variation.

Statistical analysis For evaluation of differences of the numbers of genome-equivalents between the different ages or different sites, a Kluskal-Walis test and then multiple compar ison by the Mann-Whitney U test wi th Bonferroni’s comparisons were performed. The Mann-Whitney U test was used to compare the numbers of genome-equivalents in samples from groups of different sex, or from different body sites, for subjects with or without acne. For evaluation of the correlation of the numbers of genome-equivalents between different sites, the Spearman rank-order correlation coefficient was calculated. Differences with P-values less than 0·05

were considered to be significant. The software used was Statview 5.0 for Windows (SAS Institute, Cary, NC).

Results

16S rRNA operon copy number of P. acnes Number of genes cording for 16S rRNA of P. acnes isolates from normal skin was examined for a check of reliability in counting the bacterium with real-time PCR targeting 16S rRNA. Southern blot analysis revealed that all of 29 isolates and 2 type strains of P. acnes have three copies of 16S rRNA operon without exception (Fig. 2). The result confirmed that real-time PCR for P. acnes 16S rRNA can be used for counting the bacterium.

Reproducibility of the assays For the repeated measurements of single samples, the mean coefficient of variation (CV) was 20% for the

Figure 1 : Sampling procedure.The skin surface of the forehead was uniformly aspirated for 60 s at a constant negative pressure with a sebum aspirator for home use ①. The matter collected in the opening of the aspirator was transferred to a sterile swab (② and ③) and the tip of the swab was cut off and put into a sterile polyethylene tube ④.

69Quantitative PCR of P. acnes DNA from the skin

nose, 16% for the forehead, and 37% for the upper arm (Fig. 3A). In measurements of samples obtained from the same subject at different times, the mean CV was 47% for the nose, 58% for the forehead, and 38% for the upper arm (Fig. 3B).

P. acnes counts at different age The P. acnes counts for healthy individuals are shown for each age category (Fig. 4). The counts at each site were less than 100 for all subjects of both sexes younger than 10 years of age. In other age categories of both sexes, although P. acnes counts varied widely for each subjects within the same age category, the median counts peaked in both sexes for subjects of 15-19 years. In women, there was a second peak at the ages of 40-49 years in samples from forehead and a second increase at the ages of 40 years or older in the samples from nose.

P. acnes counts at different sites P. acnes counts of each healthy subject at different

sites were compared between nose and forehead, arm and forehead, and arm and nose (Fig. 5). Significant correlation was found in all three combinations. P. acnes counts between nose and forehead were most strongly correlated (n = 209, r = 0.276, P < 0.001). In healthy subjects aged 15 or more years of both sexes, median counts were greatest on the nose, less on the forehead, and least on the upper arm (Table 2). Median differences were often by about a factor of 10. Male subjects had significantly greater P. acnes counts than female subjects at each site; except for the arm, differences were often by a factor of 10.

P. acnes counts of subjects with and without acne We compared the P. acnes counts in subjects between the ages of 10 and 29 years, and with and without acne vulgaris (Table 3). In both sexes aged 10-14 years, the counts from the nose and forehead in subjects with acne were higher than those in subjects without acne, except for the arm. In men and women aged 15 or more years, no significant difference was found in P. acnes

Median (25th and 75th percentiles) of genome-equivalents on

Subjects n Nose Forehead Arm

Male 78 15,500a, b, g (6170-29,400) 1900a, c, h (552-7990) 160b, c, i (40-394)

Female 77 7400d, e, g (524-18,600) 273d, f, h (63-1160) 56e, f, i (20-214)

Superscripts mean that genome-equivalents of each sample were compared between the same superscripts. a-f, hP < 0.001, g P = 0.0074, and iP = 0.081, a-fFriedman test followed by Wilcoxon signed-rank test, g-iMann-Whitney U test

Figure 2 : Evaluation of 16S rRNA copy number of P. acnes isolates from the skin.Southern blot analysis with32P-labeled DNA probe for P. acnes 16S rRNA. Apart from the density of bands, three clear bands were found in all lanes loaded by DNA extracted from type strains (ATCC 6919 and 11828) and 29 isolates (No. 1-29) obtained from the healthy skin.

Table 2. P. acnes counts at three sites for healthy subjects aged 15 or more years

70 J Med Dent SciY. Miura et al.

counts between those with acne vulgaris and those without acne vulgaris at any of the three sites.

Discussion

In this paper, we present a method for quantitating bacteria in sebaceous materials aspirated from pilosebaceous follicles of normal skin using aspiration method and quantitative PCR. Many P. acnes cells were observed within pilosebaceous units along the upper region of the follicles (the infra-infundibulum) when we examined the location in situ of P. acnes on the skin by immunostaining with antibodies specif ic to the bacterium (data not shown). Some of the P. acnes cells in sebaceous follicles can be recovered by scrubbing, but results may depend on the intensity of the rubbing, making quantitative estimation of the cell population difficult. This is the main reason why we sampled sebaceous matter in hair follicles by aspiration at a constant negative pressure.

Real-time quantitative PCR was used instead of culture for counting of the bacteria aspirated from the skin. Bacterial culture takes time and effort. Calculation of genome-equivalents gave results quickly. However, estimation of P. acnes genome-equivalents may not reflect the actual number of live bacteria because dead bacteria not detected by culture will be counted; dead bacteria may not contribute to the pathogenesis of acne. Culture also may give inaccurate results because some anaerobic bacteria may die during specimen preparation. Both methods have disadvantages; it is not clear which method is best for examination of bacterial microflora of the skin. Estimation of P. acnes genome-equivalents with real-time PCR targeting 16S rRNA may be influenced by possible difference in the operon copy number of P. acnes detected.6 We examined 16S rRNA operon copy number of 31 strains of P. acnes and found no difference in the copy number among each isolate. Thus, it was confirmed that P. acnes counts obtained in the study can not be influenced by the matter of strain difference. Reproducibility of the results in estimating the P. acnes count by TaqMan PCR was satisfactory: the CV was less than 20% in samples from the forehead and nose. For the arm, the CV was high, perhaps because the actual number of P. acnes cells there was small. The reproducibility of counts of samples obtained two weeks apart was good. This observation suggested that the population density of P. acnes on the skin was fairly constant for individuals, returning to the original level within two weeks even after sebaceous matter was removed by aspiration. It is likely that P. acnes in sebaceous follicles after aspiration proliferates until the densi ty reaches a level decided by host and environmental factors. In this study, a wide variation of P. acnes counts on the skin was found not only at different ages, but also within the same age category. By culture methods with a swabbing procedure that later became standard, Evans7 reported that a rich or sparse population of P. acnes was a stable characteristic of the skin of the forehead of individuals. Our findings of correlation of P. acnes counts at three different sites also suggest that a rich or sparse population of P. acnes is a characteristic of the skin of individuals. Our f indings of highest counts on the nose, intermediate counts on the forehead, and lowest counts on the upper arm were consistent with a earlier report8

that P. acnes is more abundant at sites of high sebum secretion. Higher counts in male subjects than female

1

10

102

103

104

105

A

Num

ber

ofP

. acn

esge

nom

e

Nose Forehead Arm

a b c d e a b c d e a b c d e1

10

102

103

104

105

Num

ber

ofP

. acn

esge

nom

es

Five subjects, a - e, with healthy skin

B Nose Forehead Arm

Figure 3 : Reproducibility of the assays.Intra-assay (A) and inter-assay (B) results for TaqMan estimation of P. acnes genome-equivalents in samples from the skin of the nose, forehead, and arm of five adults (a; 23-year-old female, b; 32-year-old female, c; 25-year-old male, d; 30-year-old male, and e; 42-year-old male) with healthy skin.

71Quantitative PCR of P. acnes DNA from the skin

1

10

102

103

104

105

<10 10-14 15-19 20s 30s 40s 50s 60<

1

10

102

103

104

105

<10 10-14 15-19 20s 30s 40s 50s 60<

1

10

102

103

104

<10 10-14 15-19 20s 30s 40s 50s 60<

1

10

102

103

104

105

<10 10-14 15-19 20s 30s 40s 50s 60<

1

10

102

103

104

105

<10 10-14 15-19 20s 30s 40s 50s 60<

1

10

102

103

104

<10 10-14 15-19 20s 30s 40s 50s 60<

P = 0.0002

P = 0.0084

Num

ber

of P

. acn

esge

nom

es

Forehead

P = 0.0095

P < 0.0001

P = 0.0345

Num

ber

of P

. acn

esge

nom

es

Nose

P = 0.0441P < 0.0001

Generation, years

Num

ber

of P

. acn

esge

nom

es

Arm

P = 0.0124

P = 0.0197

P = 0.0281

P = 0.0112

P = 0.0454

P = 0.0001

P = 0.0016

P = 0.0116

P = 0.0245

P = 0.0045

P = 0.0003

P = 0.0064

P = 0.0333

Generation, years

Male (n = 105) Female (n = 104)

Figure 4 : P. acnes counts at different ages.TaqMan estimation of P. acnes genome-equivalents in samples from the skin of the nose, forehead, and arm of healthy subjects of both sexes. Numbers of subjects are shown in Table 1. Bars indicate 90% -tile and 10% -tile and boxes indicate 75% -tile and 25% -tile with lines of median values inside the boxes. The medians of each age categories were compared by a Kruskal-Walis test followed by the Mann-Whitney U test with Bonnferroni’s correction. Although significant differences were found in many pairs of age categories, those in pairs of adjacent age category with P-values from < 0·001 to 0·045 are shown by brackets to identify the age category which showed maximal genome-equivalents.

72 J Med Dent SciY. Miura et al.

subjects also may be associated with a difference in sebum secretion, because androgens stimulate sebaceous gland activity.9 The P. acnes count peaked in subjects of both sexes aged 15 to 19 years (late adolescence) in our study, but in the previous cross-sectional study of Leyden et al. 10 reported almost 30 years ago, of an American population, the number of P. acnes cells cultured from the skin was first high when subjects were in their early twenties and was highest in subjects in their late twenties and in their thirties. The reason for the difference (later high counts in the 1975 study) may be in the different methods for sampling and counting, racial differences in the subjects, or different times of sexual maturity due to different social environments. That the P. acnes count in women peaked twice is of interest. The prevalence rate of sarcoidosis, a systemic granulomatous disease of unknown etiology, also has a double peak for Japanese women at the same ages.11 A series of recent studies from Japan has suggested an etiologic link between P. acnes and some cases of sarcoidosis; P. acnes is the only microorganism to be isolated from sarcoid lesions. 12 Many genome-equivalents of P. acnes can be detected in sarcoid lymph nodes by quantitative PCR.4, 13 By hybridization in situ, P. acnes was found in sarcoid granulomas.14 S a r c o i d g r a n u l o m a s m a y f o r m b e c a u s e o f hypersensitivity to antigens of P. acnes indigenous to the affected organ, and local proliferation of the bacter ium may tr igger sarcoid granulomatous inflammation in genetically predisposed individuals.15

Acne is a follicular disorder of the skin occurring in specialized pilosebaceous units on the face and trunk.16 Acne usually appears for the first time in mid-to-late

childhood or in the early teen years. The incidence of acne is the same for the two sexes, but males often have more severe forms of the disease. The peak incidence is the age of 17 years. Most cases of acne eventually undergo complete involution, usually by the late teens or early twenties. Some cases have a delayed onset, appearing for the first time in the third or even the fourth decade. There is a well-entrenched hypothesis that bacteria (P. acnes) are important in the pathogenesis of acne. The central effect of the bacteria in acne is to incite inflammation. Leyden et al. 17 compared the P. acnes count in acne patients and healthy subjects, using the scrub technique, and found a significant difference between subjects with and without acne aged 11-20 years, but no significant difference between those aged 21-25 years. We found a significant difference between subjects with and without acne aged 10-14 years, but no significant difference between those aged 15-29 years. Compared to the Leyden report which found a peak in P. acnes counts spanning the 11-20 years age category in the general population, this study found a peak occurring and concentrated in a younger age category. This change in the general population could explain the shift in significantly different P. acnes counts to a younger age category. Moreover, in the 15-19 years male group lower P. acnes counts were found on the nose and forehead of acne patients relative to those without acne, though this is not statistically significant. The results from the studies of Leyden et al. 10, 17 by culture

and our study by PCR suggested that either P. acnes is not directly involved in the onset of acne, or else it is no more than one of a number of causes. Leyden et al. 17 suggested that the high densities of follicular

1 10 102 103 104 105 106 1 10 102 103 104 105 106 1 10 102 103 104 105 1061

10

102

103

104

105

106

1

10

102

103

104

105

106

1

10

102

103

104

105

106

Forehead Forehead Nose

Nos

e

Arm

Arm

r = 0.276p < 0.001

r = 0.254p < 0.001

r = 0.169p = 0.013

Figure 5 : Correlation of P. acnes counts at different sites. Correlation between P. acnes counts of all subjects (n = 209) with healthy skin at pairs of sites. The x- and y- axes show the number of P. acnes genome-equivalents. Many datum points overlap. The Spearman rank-order correlation coefficient was calculated for the correlation between the different sites.

73Quantitative PCR of P. acnes DNA from the skin

P. acnes in acne patients certainly need not be a primary event in inciting the disease. Larger counts of P. acnes might simply reflect a more ample habitat and nutrients for the growth of this anaerobic bacterium. I n Conc lus ion , our asp i ra t ion method w i th quantification by real-time PCR is simple, and because

culture is not needed, results are obtained quickly. The method probably can be used for counting of various bacteria. People with hair follicles containing many P. acnes cells seem not to be particularly prone to acne, except for younger teenagers.

Median (25th and 75th percentile) of genome-equivalents at three sites of subjects aged:

Age and sex Acne n Nose Forehead Arm

10-14 years

Male With acne 10 1990 (556-8230) 814 (215-8230) 336 (94-1070)

Without acne 15 59 (27-592) 34 (10-76) 37 (11-388)

P-valuea 0.014 0.0027 0.096

Female With acne 15 18,400 (1500-60,200) 681 (157-12,300) 711 (138-1550)

Without acne 12 871 (362-4370) 62 (13-395) 464 (65-736)

P-valuea 0.020 0.029 0.37

15-19 years

Male With acne 14 5860 (3500-30,300) 2340 (870-3350) 892 (237-1760)

Without acne 14 29,300 (6500-62,600) 7190 (1770-14,500) 854 (254-2460)

P-valuea 0.089 0.142 0.5814

Female With acne 17 16,200 (6790-25,300) 4480 (938-9040) 906 (358-1350)

Without acne 16 18,800 (11,000-35,100) 2290 (686-8680) 428 (144-1160)

P-valuea 0.494 0.80 0.25

20s

Male With acnes 12 9990 (6990-34,700) 3020 (1870-10,900) 102 (17-897)

Without acne 15 20,600 (14,600-24,500) 4300 (2490-9140) 148 (21-739)

P-valuea 0.41 0.49 0.85

Female With acnes 15 5420 (2540-15,900) 591 (153-2090) 58 (25-188)

Without acne 15 5420 (427-11,500) 770 (86-1320) 53 (20-230)

P-valuea 0.35 0.79 0.84aMann-Whitney U test.

Table 3. P. acnes counts at different sites for subjects with and without acne vulgaris aged from 10 to 29 years

74 J Med Dent SciY. Miura et al.

Acknowledgements

This work was supported in part by grants from the Ministry of Education, Culture, Sports, Science, and Technology (14370195).

References1. Will iamson P, Kligman AM. A new method for the

quantitative investigation of cutaneous bacteria. J Invest Dermatol. 1965;45:498-503.

2. Shaw CM, Smith JA, McBride ME, Duncan WC. An evaluation of techniques for sampling skin flora. J Invest Dermatol. 1970;54:160-163.

3. Kligman AM, Plewig G. Classification of acne. 1976;17 (3) :520-522.

4. Eishi Y, Suga M, Ishige I, et al. Quantitative analysis of mycobacterial and propionibacterial DNA in lymph nodes of Japanese and European patients with sarcoidosis. J Clin Microbiol. 2000;40:198-204.

5. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning. A Laboratory Mannual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989.

6. Candela M, Vitali B, Matteuzzi D, Brigidi P. Evaluation of the rrn operon copy number in Bifidobacterium using real-time PCR. Lett Appl Microbiol. 2004;38:229-232.

7. Evans CA. Persistent individual differences in the bacterial flora of the skin of the forehead: numbers of propionibacteria. J Invest Dermatol. 1975;64:42-46.

8. McGinley KJ, Webster GF, Ruggieri MR, Leyden JJ. Regional variations in density of cutaneous propionibacterial: correlation of Propionibacterium acnes populations with

sebaceous secretion. J Clin Microbiol. 1980;12:672-675.9. Yamamoto A, Ito M. Sebaceous gland activity and urinary

androgen levels in children. J Dermatol Sci. 1992;4:98-104.

10. Leyden JJ, McGinley KJ, Mills OH, Kligman AM. Age-related changes in the resident bacterial flora of the human face. J Invest Dermatol. 1975;65:379-381.

11. Yamaguchi M, Hosoda Y, Sasaki R, Aoki K. Epidemiological study on sarcoidosis in Japan: recent trends in incidence and prevalence rates and changes in epidemiological features. Sarcoidosis. 1989;6:138-146.

12. Abe C, Iwai K, Mikami R, Hosoda Y. Frequent isolation of Propionibacterium acnes from sarcoidosis lymph nodes. Zentralbl Bakteriol Hyg [A]. 1984;256:541-547.

13. Ishige I, Usui Y, Takemura T, Eishi Y. Quantitative PCR of mycobacterial and propionibacterial DNA in lymph nodes of Japanese pat ients with sarcoidosis . Lancet . 1999;354:120-123.

14. Yamada T, Eishi Y, Ikeda S, et al. In situ localization of Propionibacterium acnes DNA in lymph nodes from sarcoidosis patients by signal amplification with catalysed reporter deposition. J Pathol. 2002;198:541-547.

15. Ebe Y, Ikushima S, Yamaguchi T, et al . Proliferative response of peripheral blood mononuclear cells and levels of ant ibody to recombinant protein from Propionibacterium acnes DNA expression library in Japanese patients with sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 2000;17:256-265.

16. Pochi PE. The pathogenesis and treatment of acne. Annu Rev Med. 1990;41:187-198.

17. Leyden JJ, McGinley KJ, Mil ls OH, Kligman AM. Propionibacterium levels in patients with and without acne vulgaris. J Invest Dermatol. 1975;65:382-384.