Effective cellulose production by a coculture of G xylinus and L. mali

Effective cellulose production by a coculture of G xylinus and L. mali

(Parte 1 de 2)

Effective cellulose production by a coculture of Gluconacetobacter xylinus and Lactobacillus mali

Akira Seto & Yu Saito & Mayumi Matsushige & Hiroki Kobayashi & Yasuyuki Sasaki & Naoto Tonouchi & Takayasu Tsuchida & Fumihiro Yoshinaga & Kenji Ueda & Teruhiko Beppu

Abstract A microbial colony that contained a marked amount of cellulose was isolated from vineyard soil. The colony was formed by the associated growth of two bacterial strains: a cellulose-producing acetic acid bacterium (st-60- 12) and a lactic acid bacterium (st-20). The 16S rDNA-based taxonomy indicated that st-60-12 belonged to Gluconacetobacter xylinus and st-20 was closely related to Lactobacillus mali. Cocultivation of the two organisms in corn steep liquor/sucrose liquid medium resulted in a threefold higher cellulose yield when compared to the st-60-12 monoculture. A similar enhancement was observed in a coculture with various L. mali strains but not with other Lactobacillus spp. The enhancement of cellulose production was most remarkable when sucrose was supplied as the substrate. L. mali mutants for exocellular polysaccharide (EPS) production were defective in promoting cellulose production, but the addition of EPS to the monoculture of st-60-12 did not affect cellulose productivity. Scanning electron microscopic observation of the coculture revealed frequent association between the st-60-12 and L. mali cells. These results indicate that cell–cell interaction assisted by the EPS-producing L. mali promotes cellulose production in st-60-12.


Some types of acetic acid bacteria synthesizecellulose called bacterial cellulose. This substance was traditionally produced as a dairy food known as nata de coco in the Philippines. Bacterialcellulose is expected to be a new industrial material because of its unique properties such as mechanical strength, high purity, and biodegradability (Ross et al. 1991;S utherland 1998). The development of a production system by using a submerged culture of Gluconacetobacter (formerly Acetobacter) spp. has enabled the production of bacterial cellulose on an industrial scale (Toyosaki et al. 1995).

To develop an effective microbial production system for this material, we screened effective cellulose producers and isolated various strains that belonged to Gluconacetobacter xylinus (formerly Acetobacter xylinum). For example, strain BPR2001 produced a marked amount of cellulose (7.7 g/l) by utilizing fructose as a substrate (Toyosaki et al. 1995). Strain BPR3001E, a mutant derivative of BPR2001, produced 9.7 g/l of cellulose from fructose (Ishikawa et al. 1995). The evidence obtained from our study has indicated that effective cellulose producers can be frequently obtained from fruits (Toyosaki et al. 1995).

In the screening study mentioned above, we isolated a colony that contained a marked amount of cellulose; this was obtained from a soil sample that was collected from a vineyard located at Yamanashi City, Japan. The subsequent microbiological characterization of the isolate revealed that it contained two different bacterial strains: a celluloseproducing acetic acid bacterium and a lactic acid bacterium. Here, we describe the characteristics of the two bacteria. Because the amount of cellulose produced by the monoculture of the former was far lower when compared to the coculture, it was assumed that coculture with the latter promoted cellulose production in the former.

Appl Microbiol Biotechnol (2006) 73:915–921 DOI 10.1007/s00253-006-0515-2

The nucleotide sequences of 16S rDNA that are reported in this paper were submitted to GenBank/EMBL/DDBJ under the accession numbers AB016864 (st-20) and AB016865 (st-60-12).

A. Seto : Y. Saito : M. Matsushige : H. Kobayashi : Y. Sasaki : K. Ueda (*): T. Beppu Life Science Research Center, College of Bioresource Sciences, Nihon University, 1866 Kameino, Fujisawa 252-8510, Japan email: ueda@brs.nihon-u.ac.jp

A. Seto : N. Tonouchi: T. Tsuchida : F. Yoshinaga Bio-Polymer Research Company, Ltd, Kanagawa, Japan

Materials and methods Bacterial strains, plasmids, and growth conditions

Gluconacetobacter sp. st-60-12 and Lactobacillus sp. st-20 were isolated in this study. Lactobacillus mali JCM1116 and other Lactobacillus strains were obtained from the Japan Collection of Microorganisms (JCM), Riken (Tsukuba, Japan). The L. mali mutants defective in exocellular polysaccharide (EPS) production were isolated by the standard UV mutagenesis. To obtain a pure culture of lactobacillus, it was cultured in 10 ml of De Man–Rogosa– Sharpe (MRS) broth (DIFCO) in a test tube without shaking at 28 °C. Gluconacetobacter sp. strain 60-12 was cultured in 100 ml of yeast extract-peptone-dextrose (YPD) medium [containing bacto yeast extract (DIFCO), 5 g/l; bactopeptone (DIFCO), 2 g/l; and glucose (Kokusan)] in a Roux flask at 28 °C without shaking. Corn steep liquor (CSL)/sucrose medium was used for cellulose production; it contained 4% (w/v) CSL (Sigma) and 4% (w/v) sucrose (Kokusan). Another CSL/sugar medium containing 4% (w/v) CSL and 4% (w/v) sugar (Kokusan) was also used. The agar medium contained 1% agar (Kokusan). For molecular cloning of 16S rDNA, Escherichia coli JM109 and M13mp19 were used as the host and vector (Maniatis et al. 1982). The growth conditions for E. coli were as described by Maniatis et al. (1982).

16S rDNA phylogeny

Methods for DNA manipulation were performed as described by Maniatis et al. (1982). Restriction endonuclease, T4 DNA ligase, and Taq polymerase were purchased from Takara Shuzo (Kyoto, Japan). Chromosomal DNA was extracted using Insta-gene Matrix (BioRad) according to the manufacturer’s instructions. The oligonucleotide universal primers B8F [5′-CGCGGATCCAGAGTTT GATC(A /C)TG GCTCAG] and B1500D [5′-CGCG GATCCTACCTTGTTACGACTTCACCCCAG] were used to amplify the 16S DNA fragment of Gluconacetobacter sp. st-60-12. Primers LACTO1 [5′-CGCGGATCCGAGAG TTTGATCCTGGCTCAGGACGAA] and LACTO2 [5′- CGCGATCGCTGCT C(C/A/T )A(G/A/T )G TTACCCCACC] were used for the amplification of the 16S DNA fragment of Lactobacillus sp. st-20. The amplified products were cleaved with BamHI and cloned into M13mp19 before nucleotide sequencing. DNA sequencing was performed using a SequiTherm Long Read cycle sequencing kit (Epicentre Technologies)and a DNA sequencer LI-COR model 4,000L. The determined 16S rDNA sequences were submitted to the DDBJ, EMBL, and GenBank databases under the accession numbers AB016865 (G. xylinus st-60-12) and AB016864 (Lactobacillus sp. st-20).

The 16S rDNA sequences of st-60-12 and st-20 were aligned with the sequences of various bacterial species using the Clustal W program (Thompson et al. 1994), and a dendrogram was constructed based on the evolutionary distance values that were calculated using Kimura’s distance (Kimura 1980). To estimate the reliability of the inferred tree, the bootstrap (Felsenstein 1985) was performed with 1,0 trials.

Production and quantification of bacterial cellulose

The st-60-12 strain was precultured at 28 °C for 72 h without shaking in a 300-ml Roux flask containing 100 ml of YPD medium. The culture broth was filtered with sterile cotton gauze to remove the cellulose matrix, and the resultant filtrate containing the bacterial cells was inoculated into the main culture medium at 1% (v/v). The Lactobacillus spp. was precultured without shaking at 37 °C for 24 h in a 20-ml test tube containing 10 ml MRS medium. The culture broth was inoculated into the main culture medium at 1% (v/v). The main culture was incubated at 28 °C for 72 h in a 500-ml baffled Erlenmeyer flask containing 50 ml of CSL/sucrose with rotary shaking at 150 rpm.

For quantification, the cellulose matrix was isolated by filtration by using a nylon mesh. The matrix was successively washed by suspending in appropriate volumes of (1) deionized water, (2) 1 N NaOH, (3) 1 N acetic acid, and (4) deionized water. Each suspended sample was incubated at 80 °C for 20 min. Step 3 was repeated three times. The final preparation was dried by heating at 70 °C for 72 h and weighed. The wash procedure efficiently removed the EPS produced by L. mali JCM1116. The addition of various amounts of the EPS isolated from the culture broth of L. mali JCM1116 (see below) did not affect the amount of cellulose produced by st-60-12.

Sugar composition analysis

The EPS produced by the pure culture of L. mali JCM1116 was collected using ethanol precipitation. The precipitant was dissolved in an appropriate volume of deionized water and incubated at 80 °C for 15 min. The solution was washed by extracting twice with equal volume of chloroform. The solution was dialyzed against deionized water and lyophilized. The sugar composition of EPS was analyzed using the method described by Yamamoto et al. (1995).

Microscopic observation

The bacterial cells were observed using an Axioskop 2 plus optical microscope (Zeiss, Oberkochen, Germany) and a scanning electron microscope (model VE-80; Keyence, Tokyo, Japan) according to the manufacturer’s instructions. The specimen for scanning electron microscopy was prepared using the osmium fixation protocol and critical point drying (Gerhardt et al. 1981).


Isolation of a bacterial colony that contained a marked amount of cellulose

We screened effective cellulose producers using CSL/ sucrose solid medium and isolated a colony from the soil collected from a vineyard in Yamanashi City, Japan (Fig. 1a, colony RS). The colony had a rough and three-dimensional appearance, the typical morphology of cellulose-producing acetic acid bacteria. However, after successive inoculations, the large colony could not be reproduced and two types of small colonies appeared. One colony was rough (Fig. 1a, colony R), while the other was smooth (Fig. 1a, colony S). The colony appearance, cell morphology (for optical micrograph, see Fig. 1b), and acid productivity suggested that the former (st-60-12) was a cellulose-producing acetic acid bacterium and the latter (st-20) was a lactic acid bacterium. It should be noted that the occasional association of the two types of colonies resulted in the formation of the large rough colony that was originally isolated from the vineyard soil.

Taxonomical characterization of the isolates

To taxonomically characterize the bacterial isolates, their 16S rDNA sequences were analyzed. The result indicated that st-60-12 belongs to G. xylinus; this species is known for its ability to produce bacterial cellulose; the percentage sequence identity between st-60-12 and G. xylinus (accession no. AB205219) was 98.8%. On the other hand, the analysis showed that st-20 did not cluster with known species. The closest relative of st-20 that was taxonomically characterized was L. mali (Fig. 2). The percentage sequence identity between st-20 and L. mali (accession no. M58824) was 94.9%. The comparison of physiological properties also suggested a close relationship between st-20 and L. mali (Table 1). The 16S rDNA of st-20 showed 9.2% sequence identity with that of an unpublished isolate (accession no. AB162131).

Fig. 1 Macroscopic and microscopic view of the cellulose-producing isolates. a A rough colony formed by the associated growth of st-60- 12 and st-20 (RS) strains, a rough colony formed by the pure growth of st-60-12 (R), and a smooth colony formed by the pure growth of st- 20 (S). Patches were photographed after 5 days of culturing on CSL/ sucrose agar medium. Bar, 1 cm. b Optical micrograph of st-60-12 cells. Bar,3 μm. c Optical micrograph of st-20 cells. Bar,3 μm

Fig. 2 The 16S rDNA-based phylogeny of st-20. The tree was constructed by the maximum-likelihood method using Clustal W (Thompson et al. 1994). Bootstrap values (1,0 trials) from neighborjoining analysis are shown adjacent to each node

Enhancement of cellulose production by coculturing with Lactobacillus spp.

To examine the effect of cocultivation with st-20 on cellulose production by G. xylinus st-60-12, the strains were cultured in the CSL/sucrose liquid medium and the cellulose yields were quantified (see “Materials and methods”). Figure 3a shows a representative result for the time course of cellulose production. The cellulose yields, particularly those in the coculture, fluctuated, but overall, the yield of cellulose produced by the coculture was threefold higher than that produced by the monoculture of st-60-12. The cellular yield in the culture system also fluctuated, but both st-60-12 and st-20 existed at about 1×107 cells/ml in the planktonic phase of the 5-day-culture.

St-60-12 was also cocultured with various Lactobacillus strains, and the cellulose yields of these cocultures were studied. As shown in Fig. 3b, L. mali strains obtained from the JCM culture collection promoted cellulose production. The most effective strain was L. mali JCM1116. On the other hand, the cocultivation with Lactobacillus agilis, Lactobacillus murinus, Lactobacillus brevis, Lactobacillus coryniformis, Lactobacillus casei, Lactobacillus sake, Lactobacillus plantarum, Lactobacillus amylophilus, and Lactobacillus animalis did not affect cellulose production (data are not shown for the last four strains). The promotion of cellulose production by L. mali JCM1116 was remarkable when the strains were cocultured in the CSL/sucrose medium. On the other hand, the cellulose yields in CSL liquid medium containing glucose, fructose (Fig. 3b), maltose, and galactose (data not shown) were low, and the enhancement of cellulose production was not evident. The cellulose production by the original coculture of st-60-12 and st-20 showed a similar dependence on sugar added to media (data not shown).

The above observation suggested that some metabolites specific to L. mali and its relatives possess an activity to promote cellulose synthesis in st-60-12. Therefore, we added

Table 1 Physiological properties of Lactobacillus sp. st-20 and Lactobacillus mali JCM1116

Characteristics st-20 L. mali JCM1116

Gram reaction Positive Positive Morphology Rods Rods Spore −− Anaerobic growth facultative facultative Motility + + Catalase production − + Esculin hydrolysis + + Gas from gluconate + − Growth at 45 °C −− G+C content of DNA (mol%) 37 32.5 Acid Lactose + + D-Maltose + + Melezitose −− Melibiose −− D-Raffinose −− L-Rhamnose + + Ribose Weak −

Fig. 3 Cellulose production by st-60-12. a Time course of cellulose production by st-60-12 cultured with (filled circles) and without (open circles) st-20. The organisms were cultured in CSL/sucrose liquid medium at 28 °C with shaking (see “Materials and methods”). Values are the means of duplicated measurements. b Cellulose production by st-60-12 cocultured with various Lactobacillus strains. Data are shown for the following: st-60-12 monoculture (lanes 1, 1, and 13),

the cell extract and culture supernatant of L. mali JCM1116 to the st-60-12 monoculture (see “Materials and methods”) and studied their effect on cellulose yield. However, marked enhancement was not observed (data not shown). Similarly, we could not detect any positive effect when the enzymes involved in sucrose metabolism, namely, commercial invertase and levansucrase, were added.

L. mali mutants for EPS production

The above observation suggested that the stimulation of cellulose production depends on some L. mali function; this function is exerted when grown on sucrose. A marked property of L. mali and st-20 grown on CSL/sucrose agar medium was the production of EPS. Sugar composition analysis revealed that the EPS isolated from the supernatant of the L. mali JCM1116 culture consisted of mannose (10.1), glucose (4.9), galactose (3.0), fucose (1.0), and rhamnose (0.5) (the numbers in parenthesis indicates relative amount). The marked EPS production by L. mali was specific to its growth in sucrose-containing medium. Therefore, we expected EPS production to be related to the stimulation of cellulose production.

To study the relationship, we isolated nine EPS-deficient

L. mali JCM1116 mutants by using UV irradiation. Figure 4 shows the representative colonies formed by the mutants on CSL/sucrose agar medium. In monoculture, the mutants formed smooth flat colonies, while the parental colony had a dome shape. The coculture of the parental strain with st-60-12 resulted in the formation of the large rough colony. On the other hand, the colonies formed by the coculture of the mutants with st-60-12 were as small as that formed by the monoculture of st-60-12 (data are shown for the representative three mutants). Coculture of the nine EPS mutants with st-60-12 under the abovementioned submerged culture condition yielded the same level of cellulose as that obtained in the st-60-12 monoculture (data not shown). The result indicated that the mutants were defective in the activity to stimulate cellulose production and suggested that the EPS production was related to the promotion of cellulose production. However, the addition of EPS isolated from the culture supernatant of L. mali JCM1116 to the culture medium did not affect the cellulose yield in the st-60-12 monoculture (data not shown).

Scanning electron microscopic observation

The st-60-12 and L. mali cells in the coculture were observed using a scanning electron microscope (Fig. 5). A notable feature was the frequent occurrence of coaggregated cells; most flocks contained a few elongated cells. Because we could not clearly observe the elongated cells in the Gram-stained samples under optical microscope, it was not clear whether the elongated cells are st-60-12 or L. mali. The coaggregation was not frequently observed in the st-60-12 monoculture and the coculture of st-60-12 with the EPS-deficient mutants of L. mali.


The unique phenomenon discovered in this study indicated that on cocultivation, a specific group of lactic acid bacteria promotes cellulose production in the cellulose producer st- 60-12. To our knowledge, this is the first study reporting effective cellulose production achieved by cocultivating two different bacteria. The 16S rDNA sequence similarity indicated that st-60-12 belongs to G. xylinus, a species that is well known for its ability to produce cellulose. Coculturing with st-20 did not enhance cellulose production in other G. xylinus strains isolated in our previous studies (unpublished observation). Therefore, we currently assume that the stimulatory effect of the lactic acid bacteria is specific to st-60-12. Although the maximum yield achieved by the coculture was lower than that of the hyper producers isolated in our previous studies, we believed it worthwhile to study the characteristics of the isolates in terms of basic microbial physiology and ecology.

Fig. 4 Colony morphology of EPS-deficient L. mali mutants and their associated growth with st-60-12. The horizontal views of the colonies formed by the pure growth of the L. mali strains (upper) and their associated growth with st-60-12 (center) are shown. The top views of the latter colonies (lower) are also shown. The culture was incubated at 28 °C for 5 days on CSL/sucrose agar medium. Bar,5m m

The 16S rDNA phylogeny indicated that st-20 was closely related to L. mali. Originally, this species was identified to comprise major homofermentative lactobacilli in freshly pressed apple juice (Carr and Davies 1970). Several strains of L. mali were formerly known as Lactobacillus yamanashiensis; this was based on the fact that these strains were isolated from wine musts collected at Yamanashi, Japan from where st-20 was isolated (Nonomura 1983; Kaneuchi et al. 1988). This city is known for the large-scale farming of grapes for winemaking. The environment may be suitable for the growth of this type of lactic acid bacteria.

The stimulation of cellulose production in the coculture was remarkable when sucrose was added to the media. The circumstantial evidence suggested that the formation of EPS by L. mali was related to the stimulation of cellulose production. We excluded the possibility of an increase in the cellulose weight due to EPS contamination because it was fully confirmed that the wash conditions resulted in the removal of EPS (see “Materials and methods”). A possible explanation for the enhancement of cellulose production is that certain enzymatic activity(ies) involved in sucrose metabolism and/or the EPS formation in L. mali facilitates the metabolic flow leading to cellulose formation in st-60- 12. However, we were unable to detect any enzymatic activity in the cell extract and culture supernatant of L. mali that stimulated cellulose production.

(Parte 1 de 2)