Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Deoxyribozyme wikipedia , lookup
Bisulfite sequencing wikipedia , lookup
Evolution of metal ions in biological systems wikipedia , lookup
Transformation (genetics) wikipedia , lookup
Genetic engineering wikipedia , lookup
Real-time polymerase chain reaction wikipedia , lookup
Microbial metabolism wikipedia , lookup
RESEARCH ARTICLE A newly discovered Anaerococcus strain responsible for axillary odor and a new axillary odor inhibitor, pentagalloyl glucose Takayoshi Fujii, Junko Shinozaki, Takayuki Kajiura, Keiji Iwasaki & Ryosuke Fudou Frontier Research Laboratories, Institute for Innovation, Ajinomoto Co. INC., Kanagawa, Japan Correspondence: Takayoshi Fujii, Frontier Research Laboratories, Institute for Innovation, Ajinomoto Co. INC., 1-1 Suzukicho, Kawasaki-ku, Kawasaki-shi 210-8681, Japan. Tel.: +81 44 244 7181; fax: +81 44 244 4757; e-mail: [email protected] Received 9 January 2014; revised 24 March 2014; accepted 19 April 2014. Final version published online 14 May 2014. DOI: 10.1111/1574-6941.12347 MICROBIOLOGY ECOLOGY Editor: Alfons Stams Keywords skin microbiota; axillary odor; 3-hydroxy-3metyl-hexanoic acid; Anaerococcus; pentagalloylglucose. Abstract Skin surface bacteria contribute to body odor, especially axillary odor. We aimed to investigate anaerobic bacteria that had not been previously studied for axillary odor formation. A new anaerobic Anaerococcus sp. A20, that releases 3-hydroxy-3-metyl-hexanoic acid (HMHA, main component of axillary odor) from its glutamyl conjugate, was discovered from axillary isolates. This strain showed strong resistance to the antimicrobial agents, triclosan and 4-isopropyl-3-methylphenol; therefore, we screened plant extracts that inhibit the A20 strain. We discovered that pentagalloyl glucose (PGG) extracted from the Chinese Gall plant exhibited both antibacterial and inhibitory activities against HMHA release by the A20 strain. As the excellent antibacterial activity and inhibitory effect of PGG against HMHA release were seen in vitro, we conducted an open study to evaluate the deodorant effects of PGG on axillary odor. The sensory tests on odor strength showed that application of the PGG solution could reduce axillary odors in vivo. Although there was a small change in axillary microbiota, the microbial count of A20 significantly reduced. These results strongly indicate PGG as a new innovative deodorant material that only affects odor-releasing bacteria in the axillary microbiota. Introduction The human body harbors a complex microbiota. Depending on the location of skin, the structure of skin microbiota is clearly different because of the different types of sweat glands such as the eccrine, apocrine, and sebaceous glands. One of the important functions of microbiota on the skin surface is to protect against infections (Coagen et al., 2008). The main components of the skin microbiota, Staphylococcus epidermidis and Propionibacterium acnes, interact with each other to maintain a slightly acidic skin pH, thereby protecting the skin from infectious pathogenic bacteria (Coagen et al., 2008). However, skin surface bacteria also contribute to body odor; the role of bacteria in axillary odor has been reported since the 1950s (Shelley et al., 1953). Zeng et al. (1991) showed that the main component of axillary odor was 3-methyl-2-hexenoic acid (3M2H), derived from symbiotic Corynebacterium species living on the secretory products of the many apocrine sweat glands present in the axillary region. Natsch et al. (2003) showed that 3-hydroxy-3-methyl-hexanoic acid (HMHA) ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved is another major contributor to axillary odor and has a structure similar to that of 3M2H. This fatty acid derivative is released from precursor, which is conjugates of HMHA with L-glutamine, by the action of bacterial Naacylglutamine aminoacylase (Natsch et al., 2003, 2006). In addition, Taylor et al. (2003) found that there was a clear correlation between the Corynebacterium count and odor strength in the axillary region. Similar to mediumchain (C6-C10) volatile fatty acids (VFA), such as 3M2H and HMHA, shorter-chain (C2-C5) VFAs also make a major contribution to axillary odor. Apart from Corynebacterium, other common skin bacteria, such as Staphylococcus and Microbacterium, also contribute to the production of short-chain VFAs (James et al., 2004a, b). A considerable amount of research has been performed on the involvement of aerobic bacteria, such as Corynebacterium and Staphylococcus, in axillary odor production, but research on anaerobic bacteria, such as Propionibacterium, is lacking. Propionibacterium metabolizes lactic acid and glycerol to produce VFAs such as acetic and propionic acids. As these VFAs make up some of the compoFEMS Microbiol Ecol 89 (2014) 198–207 199 Discovery of new axillary bacteria and new odor inhibitor nents of axillary odor, Propionibacterium is considered a potential contributor. Taylor et al. (2003) used conventional culture methods to determine the relationship between the Propionibacterium count and odor strength in the axillary region but was unable to establish any correlation. In contrast, Hasegawa et al. (2004) reported that more sulfurous odor was derived from sweat incubated under anaerobic conditions rather than under aerobic conditions, suggesting that the anaerobic bacteria in sweat contribute to axillary odor. Analysis with a recently developed next-generation sequencing method (without culturing) has confirmed that the skin microbiota contains Propionibacterium as well as other anaerobic bacteria (Costello et al., 2009). These findings led us to focus on investigating the relationship between anaerobic bacteria and axillary odor by isolating these bacteria from the axilla in healthy humans and evaluating the HMHArelease ability of each strain. In addition, as part of our continuing search for an antimicrobial agent against the newly discovered bacterium that is responsible for axillary odor, we report a new deodorant substance that did not damage skin microbiota but only affected bacteria responsible for axillary odor. Materials and methods Materials Triclosan and 4-isopropyl-3-methylphenol (IPMP) were purchased from Wako Pure Chemical Industries, Ltd (Osaka, Japan). Pentagalloyl glucose (PGG) was prepared by hydrolysis of partially purified tannic acid from the Chinese Gall plant (BREWTAN, SA Ajinomoto Omnichem NV, Louvain-la-Neuve, Belgium) through recrystallization from 2% methyl alcohol aqueous solution. The purity was > 85% by HPLC analysis. Bacterial strains were listed in each result table. Bacterial isolate from the axilla and culture conditions For isolation of anaerobic bacteria from axilla, swabscrubbed samples from 12 healthy people (seven men and five women) were obtained. A sterile cotton swab was dipped for 30 s in 1 mL phosphate-buffered saline (PBS), pH 7.0 and used to scrub 21 cm2 of the armpit. Afterward, the cotton swab was suspended for 30 s in PBS. The sample solutions were spread on BL agar (Nissui Pharmaceutical Co., Ltd, Osaka, Japan) containing 5% horse whole blood (Nippon Bio-Supp., Center, Tokyo, Japan) and were cultured at 37 °C under anaerobic conditions. After overnight incubation, single isolates were obtained and separated based on the shape of colonies. FEMS Microbiol Ecol 89 (2014) 198–207 Nine isolates were used to evaluate the ability to release HMHA from the glutamyl conjugate. HMHA release from glutamyl conjugate by isolated bacteria To evaluate the activity of isolated bacteria, an overnight culture was harvested by centrifugation and resuspended to 2.0 9 1010 cells mL1 in a semi-synthetic medium [per L: 3 g monopotassium phosphate (KH2PO4), 1.9 g dipotassium phosphate (K2HPO4), 0.2 g yeast extract, 0.2 g magnesium sulfate heptahydrate (MgSO4 9 7H2O), 1.4 g sodium chloride (NaCl), 1 g ammonium chloride (NH4Cl), 10 mg manganese (II) chloride (MnCl2), 1 mg iron (III) chloride (FeCl3), and 1 mg calcium chloride (CaCl2)]. The samples were added 0.5-mm glass beads (Yasui Kikai, Osaka, Japan) and mechanically disrupted with a Multi-Beads Shocker (Yasui Kikai) at 2700 r.p.m. for 90 s. After mechanical disruption, a final concentration of 2 mM 3-hydroxy-3-methyl-hexanoic acid-glutamine (HMHA-Gln; Sanyo Chemical Industries, Ltd, Kyoto, Japan) was added to the samples. After 24 h of incubation (36 °C) with shaking at 300 r.p.m., the samples were acidified with 1 M aqueous hydrochloric acid (final concentration, 0.016 M) and extracted with an equal volume of methyl tert-butyl ether (MTBE), and the amount of HMHA released was determined with a GC 353B gas chromatograph (GL Science Inc., Tokyo, Japan) with flame ionization detection (FID). An InterCapTM Pure-Wax column (length of 30 m, inner diameter of 0.25 mm, and film thickness of 0.25 lm; GL Science Inc.) was used, and 4 lL of the MTBE solution was injected in the split injection mode (split ratio, 10 : 1). The temperature of the injection port and FID was 270 °C and of the column oven was 220 °C for 10 min. The amount of HMHA released was expressed as the released HMHA percentage, which was obtained from the peak area of each sample divided by the peak area of 2 mM HMHA (same concentration as the initial concentration of HMHA-Gln). Genomic DNA extraction from the A20 strain The A20 strain was grown on BL agar (containing 5% horse whole blood) and was suspended in a 500 lL extraction buffer (pH 9.0) containing 100 mM Tris-HCl and 40 mM ethylenediaminetetraacetic acid (EDTA) with 10% sodium dodecyl sulfate (SDS). Glass beads (1.0 g, 0.5 mm; Yasui Kikai) were added, and the samples were mixed with a Multi-Beads Shocker (Yasui Kikai) at 2700 r.p.m. for 90 s. The samples were incubated at 70 °C for 10 min. After incubation, 500 lL phenol (saturated with TE buffer; Nacalai Tesque Inc., Kyoto, Japan) ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 200 was added to the samples and mixed for 5 s by vortexing. The samples were centrifuged at 20 400 g for 5 min at room temperature. The supernatant was transferred to new 1.5 mL centrifuge tubes, and 400 lL phenol : chloroform : isoamylalcohol (25 : 24 : 1, pH 6.7; Nacalai Tesque Inc.) was added. The samples were centrifuged at 20 400 g for 5 min at 4 °C, and the supernatant was transferred to new 1.5-mL centrifuge tubes. Afterward, 3 M sodium acetate (0.1 volumes) and an equal volume of cold isopropanol were added and mixed gently. DNA was pelleted by centrifugation at 20 400 g for 30 min at 4 °C and resuspended in a 40 lL TE buffer solution. 16S rRNA gene sequence analyses for the A20 strain PCR was performed using DNA Engine Tetrad 2 (BioRad Laboratories, Inc., CA) in a 50 lL reaction mixture containing 5 lL DNA, 1.25 U TaKaRa Ex TaqTM (Takara Bio Inc., Shiga, Japan), 109 Ex TaqTM buffer, 4 lL dNTP mixture (2.5 mM each), and 10 pmol of each universal primer, 27f (50 -AGAGTTTGATCCTGGCTCAG-30 ) and 1492r (50 -GGTTACCTTGTTACGACTT-30 ) (Dekio et al., 2007). DNA amplification was conducted using the following method: preheating at 95 °C for 3 min; 30 cycles each of denaturation at 95 °C for 30 s, annealing at 50 °C for 30 s, extension at 72 °C for 1.5 min, and final extension at 72 °C for 10 min. The quantitative real-time PCR (qPCR) products were purified by AMpure (Beckman Coulter, Inc., CA). For the sequencing reaction, the BigDye Terminator version 3.1 Cycle Sequencing kit (Life Technologies, CA) was used. The primers were the same as those used for amplification; an additional primer, 520r (50 -CCAGCMGCYGCGGTAA-30 ) was also used. Automated sequence determination was performed using an ABI PRISM 3130xl Genetic Analyzer (Applied Biosystems Inc., CA), and the alignment of the resulting 16S rRNA gene sequences was performed by MEGA 5 (Tamura et al., 2011). The 16S rRNA gene sequence of A20 was deposited under DDBJ accession no. AB853090 in GenBank. Sequence data for phylogenetic trees were retrieved from GenBank/DDBJ/EMBL, aligned by CLUSTALW using MEGA 5 and checked manually. Antimicrobial activity of triclosan, IPMP, and PGG The minimum inhibitory concentrations (MICs) were determined by measuring the test bacterial growth on the basis of absorbance. Staphylococcus, Corynebacterium, and Propionibacterium were cultured in TTL [per L: 30 g tryptic soy broth (Becton Dickinson, Tokyo, Japan) 5 g Tween 80, 1 g lecithin, 15 g agar], Mueller Hinton broth ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved T. Fujii et al. (MH; Becton Dickinson, Tokyo, Japan), and Gifu Anaerobic Broth (GAM Broth; Nissui Pharmaceutical Co.), respectively, overnight at 37 °C. For the A20 strain, the colonies grown in BL agar (containing 5% horse whole blood) were added to nutrient broth (Becton Dickinson, Tokyo, Japan) and cultured for 7 day at 37 °C. Cultured bacteria were harvested by centrifugation (13 000 g for 3 min) and resuspended to 1.0 9 109 cells mL1 in a culture medium. Triclosan, IPMP, and PGG were subjected to twofold dilution with EtOH and dispensed to a 96-well plate (each well, 2 lL). Test bacteria were added to the plate (final bacterial count, 1.0 9 106 cells mL1), and 2 lL of EtOH was used as a control. The plate (final volume 200 lL) was incubated at 37 °C for 2 or 7 day, and OD660 nm was measured. Two-week in vivo test with the 0.5% PGG solution An open study was conducted to evaluate the deodorant effects of PGG on axillary odor (Fig. 1). Nine male inhouse volunteers were asked to spray 1 mL solution containing 0.5% PGG (0.5% PGG solution) into the axilla of the dominant hand and 1 mL solution without PGG (non-PGG solution) into the axilla of the nondominant hand once a day for two consecutive weeks after bathing. Evaluation of axillary odor intensity before and after a 2-week test with the visual analogue scale (VAS) Axillary malodor was measured by three judges trained with the VAS using the cotton shirts worn by test subjects. The cotton shirt (BVD Men’s T-shirts, Fujibo Holdings Inc., Tokyo, Japan) was deodorized with a steam iron (Aquaspeed Ultracord 250) for 15 min before use. The volunteers wore the cotton shirt for 16 h after bathing on the day prior to measurement. After the left and Sampling point I (Before application) Non-deodorant usage for 1 week Skin microbiota analysis of swabscrubbed samples (T-RFLP, qPCR) Sampling point I I (After application) 0.5% PGG or non-PGG Once a day for 2 weeks VAS analysis of underwear shirt Fig. 1. Deodorant test scheme using PGG, and protocol for sampling from the axilla. T-RFLP, qPCR, and VAS. FEMS Microbiol Ecol 89 (2014) 198–207 201 Discovery of new axillary bacteria and new odor inhibitor right axillary regions were cut out from the cotton shirts that had been worn by the subjects, these pieces of cloth were placed in a Ziploc bag (16.5 9 14.9 cm, Asahi Kasei Home Products Corporation, Tokyo, Japan), sealed in an anaerobic jar (AnaeroPack Series, Mitsubishi Gas Chemical company, Inc., Tokyo, Japan), and incubated at 37 °C for 16 h. Samples with two types of odor intensity were used as standard samples for VAS analysis. For odor intensity 5, 10 lL of 10 mM HMHA was added to a rectangular piece from a cotton shirt, while for odor intensity 0, 10 lL of 100% EtOH was added to the cloth. The judges first smelled the cloth with odor intensity 5 and then smelled the cloth with odor intensity 0 prior to evaluating the odor intensity of each test sample. DNA extraction from swab-scrubbed samples obtained from the test subjects Swab-scrubbed samples were obtained by the above methods from the 2-week-test volunteers. The swab-scrubbed sample solutions were centrifuged at 13 000 g for 5 min to obtain pellets. DNA was extracted from the swabscrubbed pellets as described above. Terminal restriction fragment length polymorphism (T-RFLP) analysis PCR was performed using DNA Engine Tetrad 2 (BioRad Laboratories, Inc.) in a 50 lL reaction mixture containing 5 lL dissolved DNA (100 ng), 1.25 U TaKaRa Ex TaqTM, 109 Ex TaqTM buffer, 4 lL dNTP mixture (2.5 mM each), and 10 pmol of each universal primer, 27f and 1492r. Primer 27f was labeled with 6-FAM (6carboxyfluorescein; Applied Biosystems Inc.) for T-RFLP analysis. Amplification was performed using the following method: preheating at 95 °C for 3 min; 30 cycles each of denaturation at 95 °C for 30 s, annealing at 50 °C for 30 s and extension at 72 °C for 1.5 min; and final extension at 72 °C for 10 min. The PCR products were purified by AMpure (Beckman Coulter, Inc.) and digested with 20 U of either HhaI or MspI (Takara Bio Inc.) in a total volume of 10 lL at 37 °C for 12 h. The length of the terminal restriction fragments (T-RFs) was determined based on the standard size markers GS500 ROX and 1000 ROX (Applied Biosystems Inc.) using the ABI PRISM TM 3130xl genetic analyzer (Applied Biosystems Inc.) and GENESCAN analysis software (Applied Biosystems Inc.). Dendrogram analysis was performed using T-RFLP patterns in the BIONUMERICS software (Applied Maths, Sint-Martens-Latem, Belgium). The distances between samples were represented graphically by constructing a dendrogram based on the binary coefficient-dendrogram type (Dice-UPGMA). FEMS Microbiol Ecol 89 (2014) 198–207 qPCR for five skin bacteria The counts of S. epidermidis, S. aureus, Corynebacterium xerosis, P. acnes, and A20 in the 2-week-test samples were estimated by qPCR. Primers specific to C. xerosis were designed for the divIVA gene, and P. acnes and the A20 strain were designed for the gyrB gene in this study. The divIVA gene sequence data of C. xerosis (accession no. AM286228) and gyrB gene sequence data of P. acnes (accession no. CP001977) were obtained from GenBank/ DDBJ/EMBL. The gyrB gene sequence of the A20 strain was cloned from the genome of this strain. To clone the gyrB gene, the primers Anagyrf_1222 (50 -AGACCKG GWATGTATATMGGHC-30 ) and AnagyrR_1222 (50 -KT CTWGGTTCTACCTTYTCWC-30 ) were used. Multiple alignments of each gene sequences were carried out using MEGA 5 to determine the sequence similarity between the closely related species (Fig. 2a). The primer pairs (Table 1) were designed by the LIGHTCYCLER PRONE DESIGN Software 2.0 (Roche Diagnostics GmbH, Mannheim, Germany). The specification was tested for 18 skin bacterial genome DNA, and amplification was detected only in target bacteria (Table 2). qPCR was performed in a 20 lL mixture containing 2 lL DNA sample, 6 pmol of each primer, and 10 lL THUNDERBIRD SYBR qPCR Mix (Toyobo Co., Ltd, Osaka, Japan). Amplification and detection were carried out using LightCycler DX400 (Roche Diagnostics GmbH) with a profile of preheating at 95 °C for 20 s followed by 45 cycles each of denaturation at 95 °C for 5 s, annealing at 60 °C for 15 s and extension at 72 °C for 30 s. The counts of the five bacterial species in the samples were estimated with an internal standard curve prepared with three replicates of four concentrations (1 9 103, 1 9 105, 1 9 107, and 1 9 109 CFU mL1) of each bacterium. The bacteria genomic DNA used for the standard curve was extracted as described above. The counts of the five bacterial species in each sample have been expressed as CFU cm2. Results Isolation and identification of novel anaerobic bacteria contributing to axillary odor Based on colony shape, nine strains of anaerobic bacteria from the axillary region were isolated. To determine their HMHA-releasing activity, the percentage of HMHA release from the precursor molecule HMHA-Gln was measured. Most species, including Propionibacterium, Dermabacter, Finegoldia, Rothia, and Bacillus, did not release any HMHA from HMHA-Gln, but the release percentage from the A20 strain was very high, that is 79.8% (Table 3). The 1484 base-pairs 16S rRNA gene of A20 ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 202 T. Fujii et al. (a) Anaerococcus tetradius CCUG46590 AF542234 100 Anaerococcus lactolyticus CCUG31351 AF542233 Anaerococcus prevotii CCUG41932 AF542232 Anaerococcus vaginalis CCUG31349 AF542229 99 A20 AB853090 100 Anaerococcus octavius NCTC9810 Y07841 0.05 A20 AB853090 (b) 100 99 Anaerococcus sp. 8405254 HM587319 Uncultured Anaerococcus sp. clone ML2-55 DQ847450 Anaerococcus octavius NCTC9810 Y07841 Anaerococcus prevotii CCUG41932 AF542232 100 Anaerococcus tetradius CCUG46590 AF542234 77 Anaerococcus lactolyticus CCUG31351 AF542233 98 Anaerococcus murdochii WAL17230 DQ911243 Anaerococcus vaginalis CCUG31349 AF542229 0.01 Fig. 2. NJ tree based on 16S rRNA gene (a) and gyrB gene (b) sequences showing the relationship of the A20 and other Anaerococcus strains. The bar represents 10 nucleotide substitutions per 1000 sites. Bootstrap values (> 50%) based on 1000 replications are shown at branch nodes. Table 1. Primer pairs used in the qPCR for five skin bacteria Bacterium (gene) S. epidermidis (sodA) S. aureus (nuc) P. acnes (gyrB) C. xerosis (divIVA) A20 (gyrB) Primer SE-F SE-R nuc-F nuc-R PA_gyrBF PA_gyrBR cxdiv-nef1 cxdiv-ner1 A839F A1082R Sequence 0 5 -TCAGCAGTTGAAGGGACAGAT-3 50 -CCAGAACAATGAATGGTTAAGG-30 50 -AGGGATGGCTATCAGTAATG-30 50 -GCTGAGCTACTTAGACTTGAAA-30 50 -CTACCGATCATCCTGATGGTC-30 50 -ACCGGCATCGTAGGAAC-30 50 -GACGAGACCCTGGCCAA-30 50 -GTCTCGGACTCCGTCTTC-30 50 -CATTTATATCTGTTGATTGACAATC-30 50 -CAATTTGAAGGACAAACTAAGGCAA-30 was sequenced and analyzed for homology using the GenBank database. The results showed 96.2% homology with Anaerococcus octavius, a very high homology of 99.8% with Anaerococcus sp. 8405254 (accession no. HM587319) from human osteoarticular origin and a very high homolª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved Reference 0 Iwase et al. (2008) Brakstad et al. (1992) In this study In this study In this study ogy of 99.6% with uncultured Anaerococcus sp. clone ML2-55 (accession no. DQ847450) from human skin. In addition, the A20 strain belonged to the same group as Anaerococcus sp. 8405254 and uncultured Anaerococcus sp. clone ML2-55 and to a different group on the neighFEMS Microbiol Ecol 89 (2014) 198–207 203 Discovery of new axillary bacteria and new odor inhibitor Table 2. Specificity test of A20 specific primer No. Name Strain No. or isolated No. PCR* 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Anaerococcus sp. Anaerococcus hydrogenalis Anaerococcus lactolyticus Anaerococcus murdochii Anaerococcus octavius Anaerococcus prevotii Anaerococcus senegalensis Anaerococcus tetradius Anaerococcus vaginalis Corynebacterium xerosis Corynebacterium coyleae Corynebacterium pseudogenitalium Staphylococcus epidermidis Staphylococcus aureus Propionibacterium acnes Propionibacterium avidum Propionibacterium granulosum Propionibacterium propionicus A20 JCM7635 JCM8140 JCM15630 DSM11663 JCM6508 DSM25366 JCM1964 JCM8138 ATCC373 M8WR5 M8WR2 ATCC14990 ATCC12600 P1 P2 P3 P4 + ATCC, American Type Culture Collection; JCM, Japan Collection of Microorganisms; DSM, German Collection of Microorganisms and Cell Cultures; and other numbers were isolated strains. *A single PCR band was defined as a positive test result (+). Table 3. HMHA-releasing rates of anaerobic isolated bacteria and nine Anaerococcus strains Name Isolated No. or strain No. HMHA release (%) Propionibacteria acnes Propionibacterium avidum Propionibacterium granulossum Propionibacterium propionicus Dermabacter hominis Finegoldia magna Rothia dentocariosa Bacillus firmus Anaerococcus species Anaerococcus hydrogenalis Anaerococcus lactolyticus Anaerococcus murdochii Anaerococcus octavius Anaerococcus prevotii Anaerococcus senegalensis Anaerococcus tetradius Anaerococcus vaginalis P1 P2 P3 P4 D1 F1 R1 B1 A20 JCM7635 JCM8140 JCM15630 DSM11663 JCM6508 DSM25366 JCM1964 JCM8138 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 79.8 0.4 0.0 0.5 1.3 0.2 0.0 0.3 0.3 JCM, Japan Collection of Microorganisms; DSM, German Collection of Microorganisms and Cell Cultures; and other numbers were isolated strains. bor-joining (NJ) phylogenetic tree (Fig. 2b). The released HMHA percentage was measured in eight Anaerococcus strains (A. hydrogenalis JCM7635, A. lactolyticus JCM8140, A. murdochii JCM15630, A. octavius DSM11663, A. senegalensis DSM25366, A. tetradius FEMS Microbiol Ecol 89 (2014) 198–207 JCM1964, A. vaginalis JCM8138, and the A20) to confirm whether the ability to release HMHA was common. The analysis showed that HMHA release occurred only in the A20 strain (Table 3). MICs of antimicrobial agents against bacteria present on the skin, including the newly discovered axillary odor bacterium A20 To evaluate the effectiveness of known antimicrobial agents against the A20 isolate, the MICs of triclosan and IPMP were tested against A20, as well as the skin bacteria S. epidermidis, P. acnes, C. xerosis, and the pathogenic S. aureus (Table 4). Triclosan inhibited the growth of S. epidermidis, P. acnes, and S. aureus to < 0.0001% and of C. xerosis to 0.025%. IPMP reduced the growth of S. epidermidis, S. aureus, and P. acnes to 0.025% and of C. xerosis to 0.1%. However, triclosan and IPMP could not inhibit the A20 strain at the maximum tested concentration of 0.1%. We conducted screening for antimicrobial materials against the A20 isolate from about 500 plant extracts. PGG, a type of polyphenol extracted from the Chinese Gall plant, was found to inhibit the A20 strain. PGG inhibited the A20 isolate at the concentration of 0.025%; triclosan and IPMP did not suppress the isolate when used at the concentration of 0.1%. The inhibition activity of PGG was also stronger than that of IPMP for S. epidermidis, S. aureus, and P. acnes (Table 4). Comparison of the effect of inhibitory concentrations on HMHA release in the case of PGG and known antimicrobial agents The inhibitory concentrations of HMHA release by the A20 cell lysate were measured in triplicate (Fig. 3). To compare the inhibitory activity of PGG, triclosan and IPMP, the average IC50 values were calculated and were found to be 0.0033%, 0.0048%, and 0.011%, respectively. The inhibitory activity of PGG was 1.4 times stronger than the activity of triclosan and 3.3 times stronger than that of IPMP. Reduction of axillary odor strength by PGG application As PGG had an antimicrobial effect for the A20 isolate as well as an inhibitory effect on HMHA release in vitro, its effect on inhibiting axillary odor in vivo was estimated. At the end of the 2-week test, the axillae of the volunteers were carefully observed, and no dermatological problems were detected, leading to the conclusion that PGG did not cause any skin problems. The VAS value for the nine subjects on ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 204 T. Fujii et al. Table 4. Minimum growth inhibitory concentration of PGG, triclosan, and IPMP against five skin bacteria Minimum inhibition concentration (final %) Sample S. epidermidis ATCC14990 S. aureus ATCC12600 C. xerosis ATCC373 P. acnes JCM6425 Anaerococcus sp. A20 PGG Triclosan IPMP 0.0004 < 0.0001 0.0250 0.0016 < 0.0001 0.0250 0.0500 0.0250 0.1000 0.0016 < 0.0001 0.0250 0.0250 > 0.1 > 0.1 S, Staphylococcus; C, Corynebacterium; P, Propionibacterium; ATCC, American Type Culture Collection; JCM, Japan Collection of Microorganisms, and the A20 was isolated strain. 1 * , pentagalloylglucose , triclosan , 4-isopropyl-3-methylphenol 100 0 50 0 0 0.005 0.01 0.015 0.02 0.025 0.03 Concentration (% wt vol–1) Fig. 3. Inhibitory activities of pentagalloylglucose, triclosan, and 4isopropyl-3- methylphenol on HMHA releasing by lysates of the A20 strain. Inhibition percentage is obtained from the peak area of released HMHA of each sample divided by released HMHA without sample treatment. the control treatment side was subtracted from the VAS value of the experimental treatment side to obtain a mean value. This value was 0.2 before treatment and 1.1 after treatment, which is a significant difference (Fig. 4). The negative value of this difference indicates that the axillary odor strength decreased after 2 weeks of treatment. Reduction of the A20 count in the axilla by PGG application Two weeks after applying either the 0.5% PGG solution or the non-PGG solution as a control, bacteria counts of S. epidermidis, C. xerosis, P. acnes, and A20 were quantified by qPCR. In contrast to the in vitro results, the counts of S. epidermidis, C. xerosis, and P. acnes did not differ between the non-PGG solution and the 0.5% PGG solution treatments (Fig. 5). However, significant reduction in the counts of the A20 was observed in 0.5% PGG solution treatments as compared to those for the nonPGG solution (Fig. 5). PGG does not have a major effect on axillary microbiota In contrast to the in vitro result, the 0.5% PGG solution had no effect on the counts of S. epidermidis, C. xerosis, ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved Difference in VAS HMHA production rate (%) 150 –1 –2 Before After –3 Fig. 4. The difference in VAS of axillary odor between before and after PGG application. The difference is obtained from subtracting the non-PGG solution side from the PGG solution side. If the difference is < 0, it is implied that odor intensity of the PGG solution side is reduced after application. The average values of before and after PGG application difference in VAS were 0.2 and 1.1, respectively. Significant difference (P < 0.05) was calculated by the two-way repeated-measures ANOVA. *P < 0.05 vs. before. Markers represent each subject. and P. acnes; therefore, the effect of the PGG application on axillary microbiota was evaluated. The microbiota before and after application were analyzed by T-RFLP and clustered based on T-RFLP profiles digested with MspI and HhaI. After application, most of the samples (except no. 6) were from the same cluster, and no clear clusters corresponding to the presence or absence of PGG were observed (Fig. 6). Similarities in T-RFLP profiles before and after treatment with both 0.5% PGG solution and non-PGG solution were determined. There was no significant difference between the results for the 0.5% PGG solution side and the non-PGG solution side. Discussion The main components of axillary odor are isovaleric acid, 3M2H, HMHA, and 3-methyl-3-mercaptohexan-1-ol (Natsch et al., 2003, 2004, 2006; James et al., 2004b; FEMS Microbiol Ecol 89 (2014) 198–207 205 Discovery of new axillary bacteria and new odor inhibitor Number of bacteria (cells cm–2) * 1.0E+06 , Non-PGG , 0.5% PGG 1.0E+04 1.0E+02 1.0E+00 Fig. 5. Quantification of four axillary bacteria using qPCR after application. Statistically significant differences (P < 0.05) were calculated by the t-test (P = 0.0018). *P < 0.05. S: Staphylococcus, C: Corynebacterium, P: Propionibacterium Starkenmann et al., 2005; Emter & Natsch, 2008). These compounds are produced from odorless precursors in sweat secreted from the apocrine sweat glands through the action of indigenous skin bacteria, mainly Corynebacterium species (Fredrich et al., 2013). These species are particularly responsible for the production of 3M2H and HMHA. These molecules are produced from L-glutamineconjugated precursors by the action of Na-acylglutamine aminoacylase, an enzyme specific to the Cornyebacterium species (Natsch et al., 2003, 2006). There are many reports on the involvement of aerobic bacteria in axillary odor production, but fewer reports exist on the role of anaerobic bacteria (Leyden et al., 1981; Taylor et al., 2003). However, with new sequence analytical methods, such as next-generation sequencing, many anaerobic bacteria that could not be isolated previously have been discovered on the skin surface (Costello et al., 2009). This study focused on anaerobic bacteria and identified a strain that contributes to axillary odor production. We obtained six anaerobic genera with nine isolates from the axilla of healthy subjects by sampling and culture techniques. Using the method of Natsch et al. (2003) which involved measuring HMHA release from HMHAGln, we found that the A20 strain had a strong HMHAreleasing activity (Table 3). The identity of the A20 strain was determined using matches based on 16S rRNA gene Fig. 6. Dendrogram of terminal restriction fragment polymorphism profiles before and after PGG application. The dendrogram was constructed by the binary coefficientdendrogram (Dice-UPGMA). The numerals show the subject number, while PG and nPG show 0.5% PGG solution application and nonPGG solution application, respectively. A and B show before and after application, respectively. FEMS Microbiol Ecol 89 (2014) 198–207 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 Similarity (%) 8 nPG B 8 PG B 8 nPG A 8 PG A 6 PG A 2 nPG B 2 PG B 4 nPG B 1 nPG B 1 PG B 5 nPG B 7 nPG B 7 PG B 5 PG B 6 nPG B 5 nPG A 6 PG B 5 PG A 6 nPG A 9 nPG B 9 PG B 1 PG A 7 nPG A 7 PG A 9 nPG A 9 PG A 2 nPG A 2 PG A 4 nPG A 4 PG A 1 nPG A 4 PG B 3 PG B 3 nPG B 3 nPG A 3 PG A ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 206 similarities, and the strain was found to have a very high homology with Anaerococcus sp. 8405254 of osteoarticular origin (La Scola et al., 2011) and uncultured Anaerococcus sp. clone ML2-55 (accession no. DQ847450) of forearm origin (Gao et al., 2007) (99.8% and 99.6% homology, respectively). According to Stackebrandt and Ebers (2006), a 16S rRNA gene homology > 98% correlates with DNA–DNA hybridization indicating that the isolated A20 strain belongs to the genus Anaerococcus. As of now, anaerobic bacteria have not been identified to contribute to axillary odor. However, the ability of the A20 strain to release HMHA and the high numbers of this species in the axilla, found through qPCR (Fig. 5), suggest that A20 strongly contributes to producing axillary odor. Compared to other skin bacteria, the A20 strain showed strong resistance to the existing antimicrobial agents triclosan and IPMP (Table 4). Hence, we tested various plant extracts for antimicrobial activity against the A20 strain and discovered that PGG has antimicrobial activity against the A20 strain. PGG was also more effective than IPMP against S. epidermidis, S. aureus, P. acnes, and C. xerosis in vitro (Table 4). Our research also showed that PGG reduced the A20 strain count when applied to the skin surface (Fig. 5). In contrast, PGG application did not affect the counts of the other bacteria or the skin microbiota (Fig. 5 and 6). This difference between the effects in vivo and in vitro may be due to the difference in the growth rate of skin bacteria. The concentration of PGG is kept constant in vitro, whereas the concentration is gradually diluted by sweat after application in vivo. Therefore, the antimicrobial effect of PGG is predicted to be gradually weakened after application in vivo. Because aerobic bacteria grow faster than anaerobic bacteria, PGG was not expected to reduce aerobic bacteria with fast growth in vivo due to dilution. Accordingly, only the number of the anaerobic A20 was significantly reduced in vivo. PGG also inhibits HMHA release at a lower concentration than other antimicrobial agents such as triclosan and IPMP (Fig. 3). Considering the reports on PGG as an enzyme inhibitor (Zhang et al., 2009), PGG is suspected to directly inhibit the activities of HMHA-releasing enzymes. Usually, we can feel the body odor after daily activities. Body odors are generated by skin bacteria from the precursors adhering to the clothes after daily activities. Therefore, the using cotton shirts were incubated for 16 h to produce odors. The sensory tests on odor strength revealed that, after 2 weeks of treatment with PGG, the axillary odor significantly declined (Fig. 4). Gas chromatography measurements showed that HMHA concentrations declined in five of the nine subjects (data not shown), implying that PGG applications inhibited axillary odor. In addition, we interviewed the test subjects and asked them about the change in the odor profile, for ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved T. Fujii et al. example, the presence of a sulfurous odor or acidic odor after the test. This analysis revealed a reduction of some kind of odor after the treatment, suggesting that PGG also reduced the axillary odor components other than HMHA. The above results show that PGG application does not damage the skin microbiota and can reduce axillary odor by inhibiting Anaerococcus. Thus, PGG will be an innovative deodorant material that shows odor suppressive activity while still maintaining the skin microbiota. References Brakstad OG, Aasbakk K & Maeland JA (1992) Detection of Staphylococcus aureus by polymerase chain reaction amplification of the nuc gene. J Clin Microbiol 30: 1654–1660. Coagen AL, Nizet V & Gallo RL (2008) Skin microbiota: a source of disease or defence? Br J Dermatol 158: 442–455. Costello EK, Lauber CL, Hamady M, Fierer N, Gordon JI & Knight R (2009) Bacterial community variation in human body habitats across space and time. Science 326: 1694– 1697. Dekio I, Sakamoto M, Hayashi H, Amagai M, Suematsu M & Bennno Y (2007) Characterization of skin microbiota in patients with atopic dermatitis and in normal subjects using 16S rRNA gene-based comprehensive analysis. J Med Microbiol 56: 1675–1683. Emter R & Natsch A (2008) The sequential action of a dipeptidase and a beta-lyase is required for the release of the human body odorant 3-methyl-3-sulfanylhexan-1-ol from a secreted Cys-Gly-(S) conjugate by Corynebacteria. J Biol Chem 283: 20645–22065. Fredrich E, Barzantny H, Brune I & Tauch A (2013) Daily battle against body odor: towards the activity of the axillary microbiota. Trends Microbiol 21: 305–312. Gao Z, Tseng CH, Pei Z & Blaser MJ (2007) Molecular analysis of human forearm superficial skin bacterial biota. P Natl Acad Sci USA 104: 2927–2932. Hasegawa Y, Yabuki M & Matsukane M (2004) Identification of new odoriferous compounds in human axillary sweat. Chem Biodivers 1: 2042–2050. Iwase T, Hoshina S, Seki K, Shinji H, Masuda S & Mizunoe Y (2008) Rapid identification and specific quantification of Staphylococcus epidermidis by 50 nuclease real-time polymerase chain reaction with a minor groove binder probe. Diagn Microbiol Infect Dis 60: 217–219. James AG, Casey J, Hyliands D & Mycock G (2004a) Fatty acid metabolism by cutaneous bacteria and its role in axillary malodour. World J Microbiol Biotechnol 20: 787–793. James AG, Hyliands D & Johnston H (2004b) Generation of volatile fatty acids by axillary bacteria. Int J Cosmet Sci 26: 149–156. La Scola B, Fournier PE & Raoult D (2011) Burden of emerging anaerobes in the MALDI-TOF and 16S rRNA gene sequencing era. Anaerobe 17: 106–112. FEMS Microbiol Ecol 89 (2014) 198–207 Discovery of new axillary bacteria and new odor inhibitor Leyden JJ, McGinley KJ, Holzle E, Labows JN & Kligman AM (1981) The microbiology of the human axilla and its relationship to axillary odor. J Invest Dermatol 77: 413–416. Natsch A, Gfeller H, Gygax P, Schmid J & Acuna G (2003) A specific bacterial aminoacylase cleaves odorant precursors secreted in the human axilla. J Biol Chem 278: 5718–5727. Natsch A, Schmid J & Flachsmann F (2004) Identification of odoriferous sulfanylalkanols in human axilla secretions and their formation through cleavage of cysteine precursors by a C-S lyase isolated from axilla bacteria. Chem Biodivers 1: 1058–1072. Natsch A, Derrer S, Flachsmann F & Schmid J (2006) A broad diversity of volatile carboxylic acids, released by a bacterial aminoacylase from axilla secretions, as candidate molecules for the determination of human-body odor type. Chem Biodivers 3: 1–20. Shelley WB, Hurley HJ Jr & Nichols AC (1953) Axillary odor experimental study of the role of bacteria, apocrine sweat, and deodorants. AMA Arch Derm Syphilol 68: 430– 446. FEMS Microbiol Ecol 89 (2014) 198–207 207 Stackebrandt E & Ebers J (2006) Taxonomic parameters revisited: tarnished gold standards. Microbiol Today 33: 152–155. Starkenmann C, Niclass Y, Troccaz M & Clark AJ (2005) Identification of the precursor of (S)-3-methyl-3-sulfany lhexan-1-ol, the sulfury malodour of human axilla sweat. Chem Biodivers 2: 705–716. Tamura K, Peterson D, Peterson N, Stecher G, Nei M & Kumar S (2011) MEGA 5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731– 2739. Taylor D, Daulby A, Grimshaw S, James G, Mercer J & Vaziri S (2003) Characterization of the microflora of the human axilla. Int J Cosmet Sci 25: 137–145. Zeng X-N, Leyden JJ, Lawley HJ, Sawano K, Nohara I & Preti G (1991) Analysis of characteristic odors from human male axillae. J Chem Ecol 17: 1469–1492. Zhang J, Li L, Kim SH, Hagerman AE & L€ u J (2009) Anti-cancer, anti-diabetic and other pharmacologic and biological activities of penta-galloyl-glucose. Pharm Res 26: 2066–2080. ª 2014 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved