- Email: [email protected]
Cloth colorization caused by microbial biofilm
Available online at www.sciencedirect.com
Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
Cloth colorization caused by microbial biofilm Yuki Tsuchiya a , Jun Ohta b , Yoshiki Ishida c , Hisao Morisaki a,∗ a
Graduate School of Science and Engineering, Ritsumeikan University, Noji-higashi, Kusatsu, Shiga, Japan Graduate School of Agriculture, Kyoto University, Kitashirakawa-oiwake, Sakyo-ku, Kyoto, Kyoto, Japan c R&D Microbiology, Procter & Gamble Japan K.K., Koyo-cho, Higashinada-ku, Kobe, Hyogo, Japan
b
Received 14 December 2007; received in revised form 22 January 2008; accepted 23 January 2008 Available online 15 February 2008
Abstract In this study, cloth disfeaturement was investigated biologically. To clarify whether or not microbes can cause cloth disfeaturement, and to identify the microbes causing the disfeaturement, worn cloth samples were incubated on sweat-ingredient agar medium. Non-sterilized cloth samples became yellow-colored during incubation, and bacterial strains belonging to the genera Bacillus, Brevibacterium, Kocuria, Micrococcus and Staphylococcus were isolated from the yellow-colored parts. Two major isolates close to the genera Bacillus and Micrococcus were inoculated separately or together on cloth samples to examine whether or not these isolates can cause colorization. When the isolate close to Micrococcus was inoculated on its own or mixed with the isolate close to Bacillus, the samples turned yellow to a greater extent and a biofilm-like structure was observed by SEM on the colored areas. In contrast, the isolate close to Bacillus alone barely caused any colorization, and no biofilm-like structure was observed. From the yellow-colored samples, bacterial strains with the same 16S rRNA gene sequences as those of the inoculated strains were re-isolated. These results strongly suggest that the bacterial strain belonging to genus Micrococcus causes cloth colorization by forming a biofilm structure. © 2008 Elsevier B.V. All rights reserved. Keywords: Colorization; Cloth; Micrococcus sp.; Biofilm; Sweat
1. Introduction Clothes become disfeatured when worn and cleaned repeatedly for a long time. The disfeaturement of clothes is considered to be due to the attachment of harmful materials to the fabric, and to fabric friction [1,2]. Especially, lipid-soluble chemicals and Fe ions are known to undergo oxidation, thereby causing the colorization of clothes and other typical disfeaturements [3]. On the other hand, cloth disfeaturement may be caused by microorganisms living on the fabric. Clothes are worn by humans, and come into contact with human skin on which microorganisms exist. In fact, it has been suggested that certain types of bacteria and fungi might cause the disfeaturement of clothes [4]. However, to the best of our knowledge, the relationship between microbes, especially bacteria, and the disfeaturement of clothes has not been studied sufficiently. Humans release about 600 mL of sweat per day on average, although the amount depends on the atmospheric temperature and humidity [5]. Sweat may play an important role in cloth ∗
Corresponding author. E-mail address: [email protected] (H. Morisaki).
0927-7765/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2008.01.028
disfeaturement, because it contains various kinds of nutrients for microbes, such as NaCl, lactic acid, certain kinds of amino acids, etc. [6,7]. In this study, we aimed (1) to clarify whether or not microbes can cause cloth disfeaturement by using the ingredients of sweat as nutrients, and (2) to identify the microbes causing the disfeaturement. For these purposes, we examined the color change in various cloth samples (new or used, sterilized or non-sterilized) during incubation on an agar medium containing sweat ingredients. We also isolated the bacterial strains from the colorized parts of the cloth samples and identified them. Some isolates were examined as to whether they actually turn cloth yellow. In this study, it was revealed that there is a relationship between the bacteria isolated from the cloth samples and cloth colorization. 2. Materials and methods 2.1. Culturing of cloth samples on sweat-ingredient medium Three cloth pieces (100% cotton; 15 cm × 5 cm) were attached on the inside of a T-shirt at the part contacting the back
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
217
Fig. 1. Sample cloth pieces after 15 days of incubation. Non-sterilized cloth pieces ((a) 1-day-worn; (b) 7-day-worn) incubated for 15 days. (c) Sterilized cloth piece (7-day-worn) incubated for 15 days. The colorized areas are separated from the non-colorized areas by a black line.
of a volunteer, and the T-shirt was worn by the volunteer for 7 days successively during the daytime (suspended on a hangar at night). The three cloth pieces were retrieved one by one, after 1, 3 and 7 days, respectively. The cloth pieces were cut into smaller pieces (1.5 cm × 1.5 cm) with sterilized scissors. Half of these smaller pieces were used as intact (non-sterilized) samples, and the remaining half were used as sterilized ones (autoclaved at 121 ◦ C for 20 min). The intact and sterilized samples (worn for 1, 3, and 7 days) were put on a sweat-ingredient agar (0.7 wt%) medium (containing 9.000 g of NaCl, 1.730 g of lactic acid, 1.070 g of urea, 0.200 g of casamino acid, 0.180 g of NH4 Cl, 0.020 g of creatinine, and 0.015 g of uric acid in 1000 mL of 100 mM phosphate buffer (pH 7.0)). The samples were incubated at 27 ◦ C (actual temperature measured at the cloth surface) for up to 15 days. The color change and the magnitude of the colorized areas on the samples were measured. A standard color card (Nihon-ShikikenJigyo, Ltd., Japan) was used to determine the color type and intensity of colorization, and the colorized areas were estimated by means of an image-analysis software (Scion Image, Scion Corporation, U.S.A.). 2.2. Isolation of bacterial strains and 16S rRNA gene analysis After the sample cloth pieces were cultured on the sweatingredient medium, the bacterial strains were isolated as follows. A yellow area of the cloth sample was scratched thoroughly with a platinum loop, and then the loop was streaked on a nutrient broth (NB) agar medium (nutrient broth, 10 g; polypepton, 10 g; NaCl, 5 g; water, 1000 mL (adjusted to pH 7.2); agar, 1.5 wt%). The sweat-ingredient agar medium was inadequate because the colony sizes remained microscopic. Therefore, we used NB agar medium, on which visible colonies were formed. When different colony types appeared, each type of colony was re-streaked on a new NB agar medium until the individual colonies became similar in color, shape and surface appearance. For comparison, we tried to isolate the microbes from the noncolored areas of the cloth samples by the procedure described above. The isolates were identified by 16S rRNA gene sequencing analysis. The DNA of the isolates was extracted by the proteinase-K method. GoTaq Green Master Mix (Promega Corporation, U.S.A.) was used for the PCR. The primers used
for amplification were 25F (5 -AGTTTGATCCTGGCTC-3 ) as the forward primer and 1510R (5 -GGCTACCTTGTTACGA3 ) as the reverse primer. The thermal cycling program was as follows: initial denaturing at 95 ◦ C for 5 min, 30 cycles of 95 ◦ C for 1 min, 52 ◦ C for 1 min, and 72 ◦ C for 1 min, and a final extension at 72 ◦ C for 10 min. The PCR products were purified with a Montage PCR unit (Millipore Corporation, U.S.A.) in accordance with the manufacturer’s instructions. The sequences were determined using an ABI PRISM AVANT 3100 (PE Biosystems, U.S.A.) and the 907R primer (5 -CCGTCAATTCCTTTGAGTTT-3 ). The closest related strains were searched in the DDBJ (http://www.ddbj.nig.ac.jp/Welcome-j.html) database using the BLAST program. All of the partial sequences of the 16S rRNA gene determined in this study have been submitted to the DDBJ database under the accession numbers AB353322–AB353336. 2.3. Re-colorization by bacterial isolates Unused and 7-day-worn cloth pieces were cut into small pieces with sterilized scissors and then autoclaved at 121 ◦ C for 20 min. These samples were then put on sweat-ingredient agar medium. Onto these samples the isolates cultured as follows were inoculated. The isolates were cultured for 14 h at 27 ◦ C in a NB liquid medium until the stationary phase, centrifuged (12500 × g for 30 min), and then suspended in a liquid sweat-ingredient medium. Twenty microliters of the cell suspension were inoculated onto the two kinds of sterilized samples (unused and worn for 7 days) on sweat-ingredient agar medium; 20 L of liquid sweat-ingredient medium were used for the negative control. The samples were incubated at 27 ◦ C for up to 15 days. The color change and the magnitude of the colorized areas on the samples were measured as described above, using the standard color card and the image-analysis software. 2.4. Scanning electron microscopy All cloth samples were placed on a copper plate and observed under a scanning electron microscope (VE-8800, KEYENSE, Japan) at an acceleration voltage of 1.5–1.75 kV.
218
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
Fig. 2. The change in the color intensity (A) and area (B) of the yellow patches on the non-sterilized cloth pieces with incubation. The color intensity was determined by reference to a standard color card (Nihon-Shikiken-Jigyo, Ltd., Japan). The magnitude of the colored area was estimated using an image-analysis software (Scion Image, Scion, U.S.A.). Three kinds of samples (1Y (×), 1-day-worn; 3Y (), 3-day-worn; 7Y (), 7-day-worn) were incubated on sweat-ingredient agar medium. Color intensity level 0 corresponds to white and 10 to deep yellow. The error bars indicate the standard error (upper half only).
3. Results and discussion 3.1. Color change of the cloth samples on sweat-ingredient agar medium The cloth sample pieces (half were sterilized and half were not) worn for 1, 3 and 7 days were put on sweat-ingredient agar (0.7 wt%) medium and incubated at 27 ◦ C for up to 15 days. Yellow patches appeared on the intact (non-sterilized) samples during incubation (Fig. 1a and b), whereas the sterilized samples did not undergo any color change (Fig. 1c). These results suggest microbial participation in the colorization. On the 6th day of incubation, yellow patches were formed on almost all the intact samples. The intensity of colorization and the areas of the colored patches increased with incubation, and reached a
plateau in almost all the samples on the 15th day of incubation (Fig. 2). Among the cloth samples, the 7-day-worn ones showed the highest color intensity and the largest colorization area. On clothes worn for a long period, various materials affecting microbial activity may be attached. The 7-day-worn non-sterilized samples were observed by SEM after 15 days of incubation. Many round bodies and adhering fibril-like structures were observed on the colorized areas, as shown in Fig. 3b and c (the color intensity in 3b is less than that in 3c), which were not observed on the cloth before the incubation (Fig. 3a). This was also true of the 1- and 3-day-worn non-sterilized samples (data not shown). These results show a strong relationship between the yellow colorization of the cloth and the appearance of the round bodies and fibril-like structures.
Fig. 3. Non-sterilized cloth piece (7-day-worn) observed by SEM before incubation (a), and after 15 days of incubation (b and c); a region where the color intensity was low (b), and high (c). Bars: 6.66 m.
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
219
Table 1 The isolates from each of the incubated sample clothes Isolate
Closest relative (accession no.)
Gram stain of the relative
Similarity
Isolates from 1-day-worn cloth samples YI-1-BR-1 Brevibacterium casei (AJ251418) YI-1-BR-2 Brevibacterium casei (AJ251418) YI-1-KO-1 Kocuria kristinae (AF501368) YI-1-MI-1 Micrococcus luteus (DQ870770) YI-1-MI-2 Micrococcus luteus (DQ870770) YI-1-ST-1 Staphylococcus aureus (AY859409)
+ + + + + +
381/384 (99.2) 384/387 (99.2) 320/329 (97.3) 381/387 (98.4) 353/355 (99.4) 499/500 (99.8)
Isolates from 3-day-worn cloth samples YI-3-MI-1 Micrococcus luteus (DQ870770) YI-3-MI-2 Micrococcus luteus (DQ870770)
+ +
347/351 (98.9) 411/413 (99.5)
Isolates from 7-day-worn cloth samples YI-7-BA-1 Bacillus subtilis (AY929251) YI-7-BA-2 Bacillus subtilis (EF382365) YI-7-BA-3 Bacillus subtilis (AY929251) YI-7-KO-1 Kocuria rhizophila (Y16264) YI-7-MI-1 Micrococcus luteus (DQ870770) YI-7-MI-2 Micrococcus luteus (DQ870770) YI-7-MI-3 Micrococcus luteus (DQ870770)
+ + + + + + +
433/435 (99.5) 403/408 (98.8) 401/402 (99.8) 475/476 (99.8) 395/397 (99.5) 367/368 (99.7) 400/401 (99.8)
3.2. Isolation of bacterial strains and identification We tried to isolate the bacterial strains from the yellow patches and the non-colorized areas of the non-sterilized samples (worn for 1, 3, and 7 days) after 15 days of incubation on sweatingredient agar medium. Bacterial strains could be isolated only from the yellow patches. The 16S rRNA gene sequences of these isolates were analyzed. As shown in Table 1, the similarity with known bacterial species was rather high, and all the closest relatives were Gram-positive bacteria. Bacteria belonging to the genera Micrococcus and Bacillus were especially common as the closest relatives of the isolates. Clothes are usually kept dry, although
occasionally they become wet during sweating or washing. Thus, Gram-positive bacteria, which tolerate dryness, may be common in the bacterial flora inhabiting clothes. Bacteria belonging to the genera Micrococcus and Bacillus are usual inhabitants of the human skin surface and the air [8,9,10]. They may be transferred to clothes from surrounding environments, such as the skin and the air, and may cause cloth colorization. 3.3. Re-colorization of the cloth samples The findings described above indicate that clothes are colorized in patches due to the existence and growth of bacteria.
Fig. 4. Photos of cloth pieces after 15 days of incubation. Unused and 7-day-worn samples were sterilized and inoculated with YI-7-MI-2 (left photo), YI-7-BA-2 (middle) or a mixture of these (right). The colorized areas are separated from the non-colorized areas by a black line, or indicated by arrows (for the samples inoculated with YI-7-BA-2).
220
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
Fig. 5. SEM images of cloth pieces after 15 days of incubation. Unused and 7-day-worn samples were sterilized and inoculated with YI-7-MI-2, YI-7-BA-2 or a mixture of these.
We tried to confirm whether or not the isolates from the yellow patches could cause colorization, as follows. The strains YI-7-MI-2 and YI-7-BA-2, which were close to Micrococcus and Bacillus, respectively, and were the most frequently isolated genera, as shown in Table 1, were inoculated (volume: 20 L, containing ca. 4 × 106 cells of each strain) on two kinds of sterilized cloth samples (unused and worn for 7 days). Then, these samples were incubated on sweat-ingredient agar medium. A mixture of the two strains containing 10 L of each strain was also examined. Yellow patches were formed on both kinds of cloth samples (unused and worn for 7 days) during the incubation (Fig. 4). On the 6th day of incubation, large yellow patches were formed on the samples inoculated with the YI-7-MI-2 strain only and on those inoculated with the mixture (YI-7-MI-2 and YI-7-BA2), whereas yellow patches were barely formed on the samples inoculated with the YI-7-BA-2 strain only (Fig. 4). These results strongly suggest that the strain YI-7-MI-2 causes greater color change. The color intensity caused by YI-7-MI-2 appears to be slightly weaker compared with that observed on the intact (non-sterilized) clothes incubated for 15 days (compare Fig. 4
and Fig. 1). On the intact clothes other microbes (although not culturable in the present study) than strain YI-7-MI-2 might affect the colorization. Study on unculturable microbes affecting colorization is a further subject. The color intensity and areas of the yellow patches were greater on the 7-day-worn cloth samples than on the unused ones (data not shown). On worn clothes, various materials available as nutrients for bacteria may be attached. The inoculated cloth samples (unused and worn for 7 days) were observed by SEM. After 15 days of incubation, many round objects were observed covering both types of cloth samples, regardless of the inoculated strain (Fig. 5). In contrast, a smaller number of round objects were observed on the cloth samples just after the inoculation (Fig. 6). The inoculated cells seemed to grow and increase in number with incubation. As described above, large yellow patches formed on the samples inoculated with the YI-7-MI-2 strain and the mixture of the YI-7-MI-2 and YI-7-BA-2 strains, but only tiny yellow patches formed with YI-7-BA-2. Thus, the strain YI-7-MI-2 seems to be the main causal agent of the yellow colorization. This strain is close to the genus Micrococcus, which is reported to produce
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
221
Fig. 6. SEM images of cloth pieces immediately after inoculation with a mixture of YI-7-MI-2 and YI-7-BA-2.
Fig. 7. SEM images of fibril-like structures connecting the cloth fibers in 7-day-worn cloth pieces inoculated with YI-7-MI-2 or with a mixture of YI-7-MI-2 and MI-7-BA-2 after 15 days of incubation. The cloth fibers are indicated by arrows.
yellow pigmentation in intermembranes [8,11]. It seemed that the yellow color staining the cloth samples was the color of the YI-7-MI-2 cells themselves. Fibril-like structures were observed adhering to the cloth fibers in the greater areas of yellow colorization. Moreover, these fibril-like structures were only found when the YI-7-MI-2 strain or the mixture of the YI-7-MI-2 and YI-7-BA-2 strains was inoculated (Fig. 7), but not in the case of the YI-7-BA-2 strain alone. It seemed that the YI-7-MI-2 strain formed biofilms on the yellow areas of the cloth pieces. Biofilms usually contain microbes and polymers [12–14]. These polymers are rich in saccharides and/or proteins [15]. We detected some saccharides and proteinlike substances in the colorized cloth sample (60–80 g/cm2 as the glucose equivalent determined by phenol-sulfuric acid method [16] and 20–35 g/cm2 as the BSA equivalent by Bradford method [17]). Thus, the fibril-like structures seem to be polymers produced by the microbial cells. All of the above findings lead us to deduce that the cell growth and the formation of the biofilm structures strongly correlate with the color change caused by the YI-7-MI-2 strain. 3.4. Isolation of bacterial strains from the re-colorized patches As described above, the YI-7-MI-2 strain isolated from the yellow patches reproduced the yellow color. We isolated the
bacterial strains from the re-colorized patches. The recovered strains were close to the YI-7-MI-2 strain and the YI-MI-2 and YI-7-BA-2 strains when isolated from intensely yellow patches formed by YI-7-MI-2 only and the mixture of the two strains, respectively (the sequence similarity of the 16S rRNA genes was 99–100%); the isolates from the faintly colorized areas were close to the YI-7-BA-2 strain. These results indicate that the YI-7-MI-2 strain, which is close to Micrococcus, is the main agent causing the colorization of the cloth samples under the present experimental conditions, and that the YI-7-BA-2 strain plays only a minor role in cloth colorization. The question of whether or not the isolates can cause colorization under the daily conditions of wearing and/or washing is a subject for further investigation. 4. Conclusions Non-sterilized cloth samples were colorized when cultured on sweat-ingredient agar medium, and bacterial strains belonging to the genera Micrococcus and Bacillus were mainly isolated from the colorized parts. The strain close to Micrococcus caused a strong yellow colorization of the cloth samples when cultured on sweat-ingredient agar medium. Moreover, biofilm-like structures were observed on the yellow-colored areas. The inoculated strain was re-isolated from the yellow-colored areas. These results show, for the first time, that a strain close to Micro-
222
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
coccus causes the yellow colorization of clothes by forming a biofilm-like structure. Acknowledgements The authors thank a volunteer for wearing the sample clothes for a long period. References [1] T. Fujii, Fabric product consumption Sci. 12 (1971) 74 (in Japanese). [2] S. Munk, P. Muench, L. Stahunke, J. Adler-Nissen, P. Schieberle, J. Surfactants Detergents 3 (2000) 505. [3] H. Mutou, Cleaning Mater. Environ. Sci. 23 (1999) 54 (in Japanese). [4] M. Haruda (Ed.), Daily life and Sanitary Microbes, Nanzan-do, Tokyo, 1985, p. 364 (in Japanese). [5] G.P. Talwar, L.M. Srivastava (Eds.), Textbook of Biochemistry and Human Biology, Prentic Hall of India, New Delhi, 2004, p. 871.
[6] S. Ito (Ed.), Chemistry of Sweat, Syogaku-syoin, Japan, 1953, p. 27 (in Japanese). [7] N. Nakayama, Tohoku J. Exp. Med. 161 (1990) 25. [8] G.K. Simslt, E. O’loughlin, J. Appl. Environ. Microbiol. 58 (1992) 3423. [9] J.G. Black (Ed.), Microbilogy, 6th ed., Wiley, Hoboken, 2004 (Chapter 25). [10] H. Gibson, J.H. Taylor, K.E. Hall, J.T. Holah, J. Appl. Microbiol. 87 (1999) 41. [11] B. Sobin, G.L. Stahly, J. Bacteriol. 44 (1942) 265. [12] M.E. Davey, G.A. O’toole, Microbiol. Mol. Biol. Rev. 64 (2000) 847. [13] E. Fujiwara-Nagata, M. Eguchi, Microbes Environ. 18 (2003) 196. [14] M. Yamamoto, H. Murai, A. Takeda, S. Okunishi, H. Morisaki, Microbes Environ. 20 (2005) 14. [15] F. Ahimou, M.J. Semmens, G. Haugstad, P.J. Novak, Appl. Environ. Microbiol. 73 (2007) 2905. [16] M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Revers, F. Smith, Anal. Chem. 28 (1956) 350. [17] M.M. Bradford, Anal. Biochem. 72 (1976) 248.
Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
Cloth colorization caused by microbial biofilm Yuki Tsuchiya a , Jun Ohta b , Yoshiki Ishida c , Hisao Morisaki a,∗ a
Graduate School of Science and Engineering, Ritsumeikan University, Noji-higashi, Kusatsu, Shiga, Japan Graduate School of Agriculture, Kyoto University, Kitashirakawa-oiwake, Sakyo-ku, Kyoto, Kyoto, Japan c R&D Microbiology, Procter & Gamble Japan K.K., Koyo-cho, Higashinada-ku, Kobe, Hyogo, Japan
b
Received 14 December 2007; received in revised form 22 January 2008; accepted 23 January 2008 Available online 15 February 2008
Abstract In this study, cloth disfeaturement was investigated biologically. To clarify whether or not microbes can cause cloth disfeaturement, and to identify the microbes causing the disfeaturement, worn cloth samples were incubated on sweat-ingredient agar medium. Non-sterilized cloth samples became yellow-colored during incubation, and bacterial strains belonging to the genera Bacillus, Brevibacterium, Kocuria, Micrococcus and Staphylococcus were isolated from the yellow-colored parts. Two major isolates close to the genera Bacillus and Micrococcus were inoculated separately or together on cloth samples to examine whether or not these isolates can cause colorization. When the isolate close to Micrococcus was inoculated on its own or mixed with the isolate close to Bacillus, the samples turned yellow to a greater extent and a biofilm-like structure was observed by SEM on the colored areas. In contrast, the isolate close to Bacillus alone barely caused any colorization, and no biofilm-like structure was observed. From the yellow-colored samples, bacterial strains with the same 16S rRNA gene sequences as those of the inoculated strains were re-isolated. These results strongly suggest that the bacterial strain belonging to genus Micrococcus causes cloth colorization by forming a biofilm structure. © 2008 Elsevier B.V. All rights reserved. Keywords: Colorization; Cloth; Micrococcus sp.; Biofilm; Sweat
1. Introduction Clothes become disfeatured when worn and cleaned repeatedly for a long time. The disfeaturement of clothes is considered to be due to the attachment of harmful materials to the fabric, and to fabric friction [1,2]. Especially, lipid-soluble chemicals and Fe ions are known to undergo oxidation, thereby causing the colorization of clothes and other typical disfeaturements [3]. On the other hand, cloth disfeaturement may be caused by microorganisms living on the fabric. Clothes are worn by humans, and come into contact with human skin on which microorganisms exist. In fact, it has been suggested that certain types of bacteria and fungi might cause the disfeaturement of clothes [4]. However, to the best of our knowledge, the relationship between microbes, especially bacteria, and the disfeaturement of clothes has not been studied sufficiently. Humans release about 600 mL of sweat per day on average, although the amount depends on the atmospheric temperature and humidity [5]. Sweat may play an important role in cloth ∗
Corresponding author. E-mail address: [email protected] (H. Morisaki).
0927-7765/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2008.01.028
disfeaturement, because it contains various kinds of nutrients for microbes, such as NaCl, lactic acid, certain kinds of amino acids, etc. [6,7]. In this study, we aimed (1) to clarify whether or not microbes can cause cloth disfeaturement by using the ingredients of sweat as nutrients, and (2) to identify the microbes causing the disfeaturement. For these purposes, we examined the color change in various cloth samples (new or used, sterilized or non-sterilized) during incubation on an agar medium containing sweat ingredients. We also isolated the bacterial strains from the colorized parts of the cloth samples and identified them. Some isolates were examined as to whether they actually turn cloth yellow. In this study, it was revealed that there is a relationship between the bacteria isolated from the cloth samples and cloth colorization. 2. Materials and methods 2.1. Culturing of cloth samples on sweat-ingredient medium Three cloth pieces (100% cotton; 15 cm × 5 cm) were attached on the inside of a T-shirt at the part contacting the back
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
217
Fig. 1. Sample cloth pieces after 15 days of incubation. Non-sterilized cloth pieces ((a) 1-day-worn; (b) 7-day-worn) incubated for 15 days. (c) Sterilized cloth piece (7-day-worn) incubated for 15 days. The colorized areas are separated from the non-colorized areas by a black line.
of a volunteer, and the T-shirt was worn by the volunteer for 7 days successively during the daytime (suspended on a hangar at night). The three cloth pieces were retrieved one by one, after 1, 3 and 7 days, respectively. The cloth pieces were cut into smaller pieces (1.5 cm × 1.5 cm) with sterilized scissors. Half of these smaller pieces were used as intact (non-sterilized) samples, and the remaining half were used as sterilized ones (autoclaved at 121 ◦ C for 20 min). The intact and sterilized samples (worn for 1, 3, and 7 days) were put on a sweat-ingredient agar (0.7 wt%) medium (containing 9.000 g of NaCl, 1.730 g of lactic acid, 1.070 g of urea, 0.200 g of casamino acid, 0.180 g of NH4 Cl, 0.020 g of creatinine, and 0.015 g of uric acid in 1000 mL of 100 mM phosphate buffer (pH 7.0)). The samples were incubated at 27 ◦ C (actual temperature measured at the cloth surface) for up to 15 days. The color change and the magnitude of the colorized areas on the samples were measured. A standard color card (Nihon-ShikikenJigyo, Ltd., Japan) was used to determine the color type and intensity of colorization, and the colorized areas were estimated by means of an image-analysis software (Scion Image, Scion Corporation, U.S.A.). 2.2. Isolation of bacterial strains and 16S rRNA gene analysis After the sample cloth pieces were cultured on the sweatingredient medium, the bacterial strains were isolated as follows. A yellow area of the cloth sample was scratched thoroughly with a platinum loop, and then the loop was streaked on a nutrient broth (NB) agar medium (nutrient broth, 10 g; polypepton, 10 g; NaCl, 5 g; water, 1000 mL (adjusted to pH 7.2); agar, 1.5 wt%). The sweat-ingredient agar medium was inadequate because the colony sizes remained microscopic. Therefore, we used NB agar medium, on which visible colonies were formed. When different colony types appeared, each type of colony was re-streaked on a new NB agar medium until the individual colonies became similar in color, shape and surface appearance. For comparison, we tried to isolate the microbes from the noncolored areas of the cloth samples by the procedure described above. The isolates were identified by 16S rRNA gene sequencing analysis. The DNA of the isolates was extracted by the proteinase-K method. GoTaq Green Master Mix (Promega Corporation, U.S.A.) was used for the PCR. The primers used
for amplification were 25F (5 -AGTTTGATCCTGGCTC-3 ) as the forward primer and 1510R (5 -GGCTACCTTGTTACGA3 ) as the reverse primer. The thermal cycling program was as follows: initial denaturing at 95 ◦ C for 5 min, 30 cycles of 95 ◦ C for 1 min, 52 ◦ C for 1 min, and 72 ◦ C for 1 min, and a final extension at 72 ◦ C for 10 min. The PCR products were purified with a Montage PCR unit (Millipore Corporation, U.S.A.) in accordance with the manufacturer’s instructions. The sequences were determined using an ABI PRISM AVANT 3100 (PE Biosystems, U.S.A.) and the 907R primer (5 -CCGTCAATTCCTTTGAGTTT-3 ). The closest related strains were searched in the DDBJ (http://www.ddbj.nig.ac.jp/Welcome-j.html) database using the BLAST program. All of the partial sequences of the 16S rRNA gene determined in this study have been submitted to the DDBJ database under the accession numbers AB353322–AB353336. 2.3. Re-colorization by bacterial isolates Unused and 7-day-worn cloth pieces were cut into small pieces with sterilized scissors and then autoclaved at 121 ◦ C for 20 min. These samples were then put on sweat-ingredient agar medium. Onto these samples the isolates cultured as follows were inoculated. The isolates were cultured for 14 h at 27 ◦ C in a NB liquid medium until the stationary phase, centrifuged (12500 × g for 30 min), and then suspended in a liquid sweat-ingredient medium. Twenty microliters of the cell suspension were inoculated onto the two kinds of sterilized samples (unused and worn for 7 days) on sweat-ingredient agar medium; 20 L of liquid sweat-ingredient medium were used for the negative control. The samples were incubated at 27 ◦ C for up to 15 days. The color change and the magnitude of the colorized areas on the samples were measured as described above, using the standard color card and the image-analysis software. 2.4. Scanning electron microscopy All cloth samples were placed on a copper plate and observed under a scanning electron microscope (VE-8800, KEYENSE, Japan) at an acceleration voltage of 1.5–1.75 kV.
218
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
Fig. 2. The change in the color intensity (A) and area (B) of the yellow patches on the non-sterilized cloth pieces with incubation. The color intensity was determined by reference to a standard color card (Nihon-Shikiken-Jigyo, Ltd., Japan). The magnitude of the colored area was estimated using an image-analysis software (Scion Image, Scion, U.S.A.). Three kinds of samples (1Y (×), 1-day-worn; 3Y (), 3-day-worn; 7Y (), 7-day-worn) were incubated on sweat-ingredient agar medium. Color intensity level 0 corresponds to white and 10 to deep yellow. The error bars indicate the standard error (upper half only).
3. Results and discussion 3.1. Color change of the cloth samples on sweat-ingredient agar medium The cloth sample pieces (half were sterilized and half were not) worn for 1, 3 and 7 days were put on sweat-ingredient agar (0.7 wt%) medium and incubated at 27 ◦ C for up to 15 days. Yellow patches appeared on the intact (non-sterilized) samples during incubation (Fig. 1a and b), whereas the sterilized samples did not undergo any color change (Fig. 1c). These results suggest microbial participation in the colorization. On the 6th day of incubation, yellow patches were formed on almost all the intact samples. The intensity of colorization and the areas of the colored patches increased with incubation, and reached a
plateau in almost all the samples on the 15th day of incubation (Fig. 2). Among the cloth samples, the 7-day-worn ones showed the highest color intensity and the largest colorization area. On clothes worn for a long period, various materials affecting microbial activity may be attached. The 7-day-worn non-sterilized samples were observed by SEM after 15 days of incubation. Many round bodies and adhering fibril-like structures were observed on the colorized areas, as shown in Fig. 3b and c (the color intensity in 3b is less than that in 3c), which were not observed on the cloth before the incubation (Fig. 3a). This was also true of the 1- and 3-day-worn non-sterilized samples (data not shown). These results show a strong relationship between the yellow colorization of the cloth and the appearance of the round bodies and fibril-like structures.
Fig. 3. Non-sterilized cloth piece (7-day-worn) observed by SEM before incubation (a), and after 15 days of incubation (b and c); a region where the color intensity was low (b), and high (c). Bars: 6.66 m.
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
219
Table 1 The isolates from each of the incubated sample clothes Isolate
Closest relative (accession no.)
Gram stain of the relative
Similarity
Isolates from 1-day-worn cloth samples YI-1-BR-1 Brevibacterium casei (AJ251418) YI-1-BR-2 Brevibacterium casei (AJ251418) YI-1-KO-1 Kocuria kristinae (AF501368) YI-1-MI-1 Micrococcus luteus (DQ870770) YI-1-MI-2 Micrococcus luteus (DQ870770) YI-1-ST-1 Staphylococcus aureus (AY859409)
+ + + + + +
381/384 (99.2) 384/387 (99.2) 320/329 (97.3) 381/387 (98.4) 353/355 (99.4) 499/500 (99.8)
Isolates from 3-day-worn cloth samples YI-3-MI-1 Micrococcus luteus (DQ870770) YI-3-MI-2 Micrococcus luteus (DQ870770)
+ +
347/351 (98.9) 411/413 (99.5)
Isolates from 7-day-worn cloth samples YI-7-BA-1 Bacillus subtilis (AY929251) YI-7-BA-2 Bacillus subtilis (EF382365) YI-7-BA-3 Bacillus subtilis (AY929251) YI-7-KO-1 Kocuria rhizophila (Y16264) YI-7-MI-1 Micrococcus luteus (DQ870770) YI-7-MI-2 Micrococcus luteus (DQ870770) YI-7-MI-3 Micrococcus luteus (DQ870770)
+ + + + + + +
433/435 (99.5) 403/408 (98.8) 401/402 (99.8) 475/476 (99.8) 395/397 (99.5) 367/368 (99.7) 400/401 (99.8)
3.2. Isolation of bacterial strains and identification We tried to isolate the bacterial strains from the yellow patches and the non-colorized areas of the non-sterilized samples (worn for 1, 3, and 7 days) after 15 days of incubation on sweatingredient agar medium. Bacterial strains could be isolated only from the yellow patches. The 16S rRNA gene sequences of these isolates were analyzed. As shown in Table 1, the similarity with known bacterial species was rather high, and all the closest relatives were Gram-positive bacteria. Bacteria belonging to the genera Micrococcus and Bacillus were especially common as the closest relatives of the isolates. Clothes are usually kept dry, although
occasionally they become wet during sweating or washing. Thus, Gram-positive bacteria, which tolerate dryness, may be common in the bacterial flora inhabiting clothes. Bacteria belonging to the genera Micrococcus and Bacillus are usual inhabitants of the human skin surface and the air [8,9,10]. They may be transferred to clothes from surrounding environments, such as the skin and the air, and may cause cloth colorization. 3.3. Re-colorization of the cloth samples The findings described above indicate that clothes are colorized in patches due to the existence and growth of bacteria.
Fig. 4. Photos of cloth pieces after 15 days of incubation. Unused and 7-day-worn samples were sterilized and inoculated with YI-7-MI-2 (left photo), YI-7-BA-2 (middle) or a mixture of these (right). The colorized areas are separated from the non-colorized areas by a black line, or indicated by arrows (for the samples inoculated with YI-7-BA-2).
220
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
Fig. 5. SEM images of cloth pieces after 15 days of incubation. Unused and 7-day-worn samples were sterilized and inoculated with YI-7-MI-2, YI-7-BA-2 or a mixture of these.
We tried to confirm whether or not the isolates from the yellow patches could cause colorization, as follows. The strains YI-7-MI-2 and YI-7-BA-2, which were close to Micrococcus and Bacillus, respectively, and were the most frequently isolated genera, as shown in Table 1, were inoculated (volume: 20 L, containing ca. 4 × 106 cells of each strain) on two kinds of sterilized cloth samples (unused and worn for 7 days). Then, these samples were incubated on sweat-ingredient agar medium. A mixture of the two strains containing 10 L of each strain was also examined. Yellow patches were formed on both kinds of cloth samples (unused and worn for 7 days) during the incubation (Fig. 4). On the 6th day of incubation, large yellow patches were formed on the samples inoculated with the YI-7-MI-2 strain only and on those inoculated with the mixture (YI-7-MI-2 and YI-7-BA2), whereas yellow patches were barely formed on the samples inoculated with the YI-7-BA-2 strain only (Fig. 4). These results strongly suggest that the strain YI-7-MI-2 causes greater color change. The color intensity caused by YI-7-MI-2 appears to be slightly weaker compared with that observed on the intact (non-sterilized) clothes incubated for 15 days (compare Fig. 4
and Fig. 1). On the intact clothes other microbes (although not culturable in the present study) than strain YI-7-MI-2 might affect the colorization. Study on unculturable microbes affecting colorization is a further subject. The color intensity and areas of the yellow patches were greater on the 7-day-worn cloth samples than on the unused ones (data not shown). On worn clothes, various materials available as nutrients for bacteria may be attached. The inoculated cloth samples (unused and worn for 7 days) were observed by SEM. After 15 days of incubation, many round objects were observed covering both types of cloth samples, regardless of the inoculated strain (Fig. 5). In contrast, a smaller number of round objects were observed on the cloth samples just after the inoculation (Fig. 6). The inoculated cells seemed to grow and increase in number with incubation. As described above, large yellow patches formed on the samples inoculated with the YI-7-MI-2 strain and the mixture of the YI-7-MI-2 and YI-7-BA-2 strains, but only tiny yellow patches formed with YI-7-BA-2. Thus, the strain YI-7-MI-2 seems to be the main causal agent of the yellow colorization. This strain is close to the genus Micrococcus, which is reported to produce
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
221
Fig. 6. SEM images of cloth pieces immediately after inoculation with a mixture of YI-7-MI-2 and YI-7-BA-2.
Fig. 7. SEM images of fibril-like structures connecting the cloth fibers in 7-day-worn cloth pieces inoculated with YI-7-MI-2 or with a mixture of YI-7-MI-2 and MI-7-BA-2 after 15 days of incubation. The cloth fibers are indicated by arrows.
yellow pigmentation in intermembranes [8,11]. It seemed that the yellow color staining the cloth samples was the color of the YI-7-MI-2 cells themselves. Fibril-like structures were observed adhering to the cloth fibers in the greater areas of yellow colorization. Moreover, these fibril-like structures were only found when the YI-7-MI-2 strain or the mixture of the YI-7-MI-2 and YI-7-BA-2 strains was inoculated (Fig. 7), but not in the case of the YI-7-BA-2 strain alone. It seemed that the YI-7-MI-2 strain formed biofilms on the yellow areas of the cloth pieces. Biofilms usually contain microbes and polymers [12–14]. These polymers are rich in saccharides and/or proteins [15]. We detected some saccharides and proteinlike substances in the colorized cloth sample (60–80 g/cm2 as the glucose equivalent determined by phenol-sulfuric acid method [16] and 20–35 g/cm2 as the BSA equivalent by Bradford method [17]). Thus, the fibril-like structures seem to be polymers produced by the microbial cells. All of the above findings lead us to deduce that the cell growth and the formation of the biofilm structures strongly correlate with the color change caused by the YI-7-MI-2 strain. 3.4. Isolation of bacterial strains from the re-colorized patches As described above, the YI-7-MI-2 strain isolated from the yellow patches reproduced the yellow color. We isolated the
bacterial strains from the re-colorized patches. The recovered strains were close to the YI-7-MI-2 strain and the YI-MI-2 and YI-7-BA-2 strains when isolated from intensely yellow patches formed by YI-7-MI-2 only and the mixture of the two strains, respectively (the sequence similarity of the 16S rRNA genes was 99–100%); the isolates from the faintly colorized areas were close to the YI-7-BA-2 strain. These results indicate that the YI-7-MI-2 strain, which is close to Micrococcus, is the main agent causing the colorization of the cloth samples under the present experimental conditions, and that the YI-7-BA-2 strain plays only a minor role in cloth colorization. The question of whether or not the isolates can cause colorization under the daily conditions of wearing and/or washing is a subject for further investigation. 4. Conclusions Non-sterilized cloth samples were colorized when cultured on sweat-ingredient agar medium, and bacterial strains belonging to the genera Micrococcus and Bacillus were mainly isolated from the colorized parts. The strain close to Micrococcus caused a strong yellow colorization of the cloth samples when cultured on sweat-ingredient agar medium. Moreover, biofilm-like structures were observed on the yellow-colored areas. The inoculated strain was re-isolated from the yellow-colored areas. These results show, for the first time, that a strain close to Micro-
222
Y. Tsuchiya et al. / Colloids and Surfaces B: Biointerfaces 64 (2008) 216–222
coccus causes the yellow colorization of clothes by forming a biofilm-like structure. Acknowledgements The authors thank a volunteer for wearing the sample clothes for a long period. References [1] T. Fujii, Fabric product consumption Sci. 12 (1971) 74 (in Japanese). [2] S. Munk, P. Muench, L. Stahunke, J. Adler-Nissen, P. Schieberle, J. Surfactants Detergents 3 (2000) 505. [3] H. Mutou, Cleaning Mater. Environ. Sci. 23 (1999) 54 (in Japanese). [4] M. Haruda (Ed.), Daily life and Sanitary Microbes, Nanzan-do, Tokyo, 1985, p. 364 (in Japanese). [5] G.P. Talwar, L.M. Srivastava (Eds.), Textbook of Biochemistry and Human Biology, Prentic Hall of India, New Delhi, 2004, p. 871.
[6] S. Ito (Ed.), Chemistry of Sweat, Syogaku-syoin, Japan, 1953, p. 27 (in Japanese). [7] N. Nakayama, Tohoku J. Exp. Med. 161 (1990) 25. [8] G.K. Simslt, E. O’loughlin, J. Appl. Environ. Microbiol. 58 (1992) 3423. [9] J.G. Black (Ed.), Microbilogy, 6th ed., Wiley, Hoboken, 2004 (Chapter 25). [10] H. Gibson, J.H. Taylor, K.E. Hall, J.T. Holah, J. Appl. Microbiol. 87 (1999) 41. [11] B. Sobin, G.L. Stahly, J. Bacteriol. 44 (1942) 265. [12] M.E. Davey, G.A. O’toole, Microbiol. Mol. Biol. Rev. 64 (2000) 847. [13] E. Fujiwara-Nagata, M. Eguchi, Microbes Environ. 18 (2003) 196. [14] M. Yamamoto, H. Murai, A. Takeda, S. Okunishi, H. Morisaki, Microbes Environ. 20 (2005) 14. [15] F. Ahimou, M.J. Semmens, G. Haugstad, P.J. Novak, Appl. Environ. Microbiol. 73 (2007) 2905. [16] M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Revers, F. Smith, Anal. Chem. 28 (1956) 350. [17] M.M. Bradford, Anal. Biochem. 72 (1976) 248.