Isolation of digested sludge-assimilating fungal strains and their potential applications

Authors


Correspondence

Katsuhiko Fujii, Faculty of Agriculture, Yamaguchi University, Yoshida, Yamaguchi 7538515, Japan. E-mail: kfujii@yamaguchi-u.ac.jp

Abstract

Aims

Digested sludge (DS) is a major waste product of anaerobic digestion of sewage sludge and is resistant to biodegradation. In this study, we isolated and characterized DS-assimilating fungi from soil.

Methods and Results

We tried to isolate DS-assimilating strains by enrichment culture using DS as the nutrient source, but microbial growth was not observed in any culture. To eliminate the inhibitory effect of metals in DS on microbial growth, acid-treated DS was subsequently used for enrichment, and eight fungal strains were isolated from the subcultures. At least 10–30% reduction in sludge was observed after 1-week cultivation, and prolonged cultivation led to further sludge reduction. All isolates produced xylanase, chitinase and keratinase. Phylogenetic analysis revealed that the isolates were Penicillium, Fusarium, Chaetomium, Cunninghamella, Neosartorya and Umbelopsis. Some isolates were suggested novel species.

Conclusions

To the best of our knowledge, our study is the first to report the isolation of DS-assimilating strains.

Significance and Impact of the Study

These isolates may be useful for commercial production of microbial enzymes using DS as the substrate. Because xylan, chitin and keratin in sludge–hyphae complexes are considered to be partially depolymerized, this material could also be utilized as a readily available fertilizer.

Introduction

The activated sludge method is used as a biological sewage treatment process worldwide; however, there remain several technical issues, including the production of a huge amount of sewage sludge (Liu and Tay 2001). Anaerobic digestion is generally employed in developed nations to reduce sludge and produce biogas. However, the overall anaerobic digestion-mediated degradation efficiency of sludge is still around 30%–60% (Appels et al. 2008; Radaideh et al. 2010; Rapport et al. 2012), and a large quantity of biodegradation-resistant residue (digested sludge, DS) is produced as a major industrial waste. For instance, approximately 620 million and 170 million tons (dry weight) of sewage sludge containing DS are produced annually in United States and Japan, respectively, most of which is incinerated as useless material (Dufreche et al. 2007; Ministry of the Environment, Japan 2012). Because the lack of final disposal sites has become a serious problem recently, it is imperative to reduce the amount of DS requiring disposal by making use of it.

Thermophilic bacteria that assimilate sewage sludge under aerobic conditions have been reported recently. Bacillus stearothermophilus, isolated by Kim et al. (2002), degraded 37% of total suspended solid (TSS) of sewage sludge in 80 days under optimal conditions. Li et al. (2009) isolated Brevibacillus sp. KH3, which degraded 32·8% of TSS and improved autolysis of unsterilized sludge to 54·8%. Liu et al. (2010, 2011) reported approximately 40–50% degradation of volatile solids in sewage sludge by a thermophilic bacterial flora; however, they did not evaluate the degradation efficiency of TSS. Overall, these thermophiles have been shown to be capable of digesting up to 50% of sewage sludge under optimal conditions, but recalcitrant DS-like residue still remains. Therefore, the isolation of microbes possessing DS-assimilating activity is of considerable interest not only with regard to the study of microbial ecology but also for establishing the technology to transform DS into a value-added resource.

In this study, we aimed to isolate and characterize DS-assimilating microbes.

Materials and methods

Chemicals

Cellulose (Avicel), carboxymethyl cellulose, birchwood xylan, syringaldazine, azure chitin and azure keratin were purchased from Sigma-Aldrich (St. Louis, MO, USA). Yeast nitrogen base (YNB) was obtained from Becton, Dickinson and Company (Franklin Lakes, NJ, USA). Reagents for molecular biology, including KOD FX Neo DNA polymerase, were purchased from Toyobo (Osaka, Japan). All other chemicals were purchased from Wako Pure Chemicals (Kyoto, Japan).

Treatment of DS

Dewatered form of DS was obtained from a municipal sewage treatment plant in Yamaguchi City, Japan. To prepare acid-treated DS (AS), 300 g of DS and 600 ml of 1 mol l−1 H2SO4 were mixed in a 1000 ml beaker and incubated at room temperature for 1 h. The mixture was then filtered through trifold gauze to remove the supernatant, and the filter residue was thoroughly rinsed with tap water until pH of the filtrate reached 6·0. The resultant residue (AS) was then dried completely in a FSP450 dry oven (Advantec, Tokyo, Japan) at 105°C for 24 h. To determine the metal content of DS and AS, 200 mg of dried sludge from each was incubated in 1 ml of conc. H2SO4 for 3 h, after which 17 ml deionized water was added to the mixture, which was then autoclaved at 121°C for 1 h to extract metals. The resulting hydrolysate was filtered through an Advantec GB-140 glass fibre filter, and the metals in the filtrate were analysed using an Optima 8300 ICP-OES spectrometer (PerkinElmer, MA, USA). The organic carbon and nitrogen contents of DS and AS were determined by the Tyurin and Kjeldahl methods (Marumoto et al. 1978; Bremner and Mulvaney 1982).

Sampling location, enrichment cultivation and microbe isolation

To isolate sludge-assimilating microbes, nine soil samples were collected at several locations in Yamaguchi Prefecture, as shown in Table 1, and were screened through a 2-mm sieve. Ten millilitres of deionized water or 0·67% YNB containing 200 mg of dried AS (WA and YA medium, respectively) were used for enrichment, subculture and pure culture of sludge-assimilating microbes. Because WA medium composed only of AS and deionized water, AS was the sole source of carbon, nitrogen, phosphorus, sulfur and mineral salts in WA medium. In contrast, YNB contains 5·0 g l−1 (NH4)2SO4, 1·0 g l−1 KH2PO4, 500 mg l−1 MgSO47H2O, 100 mg l−1 NaCl, 100 mg l−1 CaCl22H2O, 0·5 mg l−1 H3BO3, 0·4 mg l−1 MnSO45H2O, 0·4 mg l−1 ZnSO47H2O, 0·2 mg l−1 FeCl3, 0·2 mg l−1 Na2MoO42H2O, 0·1 mg l−1 KI, 0·04 mg l−1 CuSO45H2O, 2·0 mg l−1 inositol, 0·4 mg l−1 Ca-pantothenate, 0·4mg l−1 niacin, 0·4 mg l−1 thiamine-HCl, 0·4 mg l−1 pyridoxine-HCl, 0·2 mg l−1 riboflavin, 0·2 mg l−1 p-aminobenzoic acid, 2 mg l−1 folate, and 2 mg l−1 biotin, and thus AS was considered as the sole carbon source in YA medium.

Table 1. Soil samples used in this study
Soil samplesSampling locationTotal culturable microbes (CFU per gram of dry soil)
Fern bushYamaguchi University experimental forest/farm (34° 8′ N, 131° 28′ E)6·4 × 106
Cedar forest-14·5 × 106
Cedar forest-23·8 × 106
Bamboo forest-12·3 × 106
Bamboo forest-21·4 × 106
Drained rice paddy0·7 × 106
Soy bean field1·1 × 106
Ore yardAbandoned mine located in a cedar forest area of Yamaguchi (34º 14′ N, 131° 20′ E)0·9 × 106
Mining gallery1·6 × 106

The sieved soil sample (5 g) was suspended in 30 ml sterile water and then vortexed vigorously for 1 min. The resultant soil suspension (100 μl) was inoculated to both media and cultivated for 2 weeks at 30°C with shaking at 150 rev min−1. Soil cultures showing microbial growth (turbidity) were subcultured 3 times, and aliquots (50 μl each) of their successive subcultures were streaked onto 0·5% (w/v) glucose–YNB agar (solidified with 1·5% agar) and incubated for 1 week at 30°C to isolate microbes. Emerging single colonies were isolated and examined for further studies, as described below.

For determination of the number of colony forming units of total culturable microbes in the soil samples, the soil suspension described above was serially diluted (1/10, 1/100 and 1/1000), inoculated directly onto 0·5% (w/v) glucose–YNB agar and cultivated statistically at 30°C for 1 week.

Sludge-assimilating activity and enzyme assay

The sludge-assimilating activity of the soil culture and pure culture of the isolate was examined by monitoring the change in dry weight of the culture residue, a complex of AS and cultivated microbial hyphae (sludge–hyphae complex), remaining in each culture after 1–4 weeks of cultivation. The culture residue was recovered by centrifugation of cultures at 3000 × g for 10 min using an H-19F benchtop centrifuge (Kokusan, Tokyo, Japan), dried in an FSP450 dry oven at 105°C for 24 h and weighed using an HR-150AZ electric microbalance (A & D, Tokyo, Japan).

To examine enzyme activity, the isolates were cultivated in 10 ml of WA for 1 week at 30°C, with shaking at 150 rev min−1. The culture supernatants of each isolate were separated from the sludge–hyphae complex by centrifugation at 3000 × g for 10 min and then filtered through a 0·2 μm-Omnipore filter (Millipore, MA, USA). The filtrate was used as an enzyme solution. Exocellulase, endocellulase and xylanase activities were determined using Avicel, carboxymethyl cellulose and birchwood xylan as substrates, respectively, by the DNS method (Ghose 1987). Laccase activity was determined using syringaldazine as a substrate, according to the method of Leonowicz and Grzywnowicz (1981). Chitinase and keratinase activities were determined using azure chitin and azure keratin as substrates, according to the methods of Ramı́rez et al. (2004) and Riffel et al. (2007), respectively.

Statistical analyses of the sludge assimilation and enzyme activities were performed using Student's t-test.

Phylogenetic study of isolates

Cell mycelia of isolates were obtained from 10 ml pure cultures in YM broth (10 g l−1 glucose, 3 g l−1 yeast extract, 3 g l−1 malt extract, 5 g l−1 peptone). DNA of the isolates was extracted using a Master Pure Yeast DNA Purification Kit (Epicentre Biotechnologies, Madison, WI, USA) and was used as a PCR template. PCR was performed to amplify internal transcribed spacer regions (approximately 500 bp, including ITS1, 5·8S, and ITS2 regions) of fungal ribosomal DNA with primers ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′), as developed by White et al. (1990). PCR reactions consisted of 30 cycles of 98°C for 10 s, 54°C for 30 s and 68°C for 1 min. PCR products were stored at −30°C until required for sequencing.

Direct sequencing of the amplified DNA fragments was performed using BigDye Terminator v3.1 (Life technologies, CA, USA), and similarities of the sequences obtained with known species were evaluated by comparison with sequence data in the GenBank, EMBL and DDBJ databases using the BLAST algorithm (Altschul et al. 1990). Phylogenetic trees were constructed using the neighbour-joining method in the Clustal W programme (Saitou and Nei 1987; Thompson et al. 1994).

Results

Enrichment culture for soil sample microbes

First, the soil samples were cultured aerobically for 2 weeks in 10 ml of water or YNB containing 200 mg DS to enrich sludge-assimilating microbes; however, microbial growth was not observed in any culture. Therefore, DS was treated by H2SO4 to remove metals and yield AS. While 30% of dry weight for DS was lost by this treatment, 87% of aluminium, 91% of iron, 41% of manganese and 83% of zinc were removed (Table 2). AS remained rich in organic carbon and nitrogen (Table 2), and thus, enrichment of sludge-assimilating microbes was re-attempted using AS as the substrate (WA and YA media). Following 2 weeks of enrichment, microbial growth was observed in 11 of 18 samples (Table 3). Subsequently, the samples showing microbial growth were subcultured three times under the same culture conditions, and six culture samples showing stable microbial growth were finally obtained (Table 3). Their growth in the subcultures was observed in 1 week.

Table 2. Chemical characteristics of digested sludge (DS) and acid-treated DS (AS) used in this study
Sludge sampleMetal content (mg per g-dry wt)Organic carbon (%)Nitrogen (%)
AlCuCdFeMnNiZn
DS8·930·29<0·0119·700·24<0·010·53416·8
AS1·150·29<0·011·690·14<0·010·093410
Table 3. Microbial growth observed during the enrichment culture and subcultures
Soil sampleMediumMicrobial growthIsolated strains
1st enrichment1st subculture2nd subculture3rd subculture
  1. YA, Mineral salts medium supplemented with AS; WA, Deionized water supplemented with AS; +, microbial growth observed; −, no growth observed.

Fern bushYA++ 
WA++++Umbelopsis sp. FernWA (AB813149)
Cedar forest-1YA 
WA++++Penicillium sp. CedarWA1 (AB813150) Penicillium sp. CedarWA2 (AB813151)
Cedar forest-2YA 
WA++++Cunninghamella sp. CedarWA3 (AB813152) Cunninghamella sp. CedarWA4 (AB813153)
Bamboo forest-1YA 
WA 
Bamboo forest-2YA+ 
WA+ 
Drained rice paddyYA 
WA 
Soy bean fieldYA 
WA+ 
Ore yardYA++++Neosartorya sp. OreWA (AB813154)
WA++++Fusarium sp. OreWA (AB813155)
Mining galleryYA++++Chaetomium sp. GalleryYA (AB813156)
WA+ 

Isolation of DS-assimilating microbes

Aliquots of the 6 subculture samples showing stable microbial growth were streaked onto 0·5% glucose–YNB agar and incubated at 30°C. After 7 days of culture, colonies on each agar plate were morphologically identical in appearance. Eight colonies were randomly selected, isolated and repurified on a fresh agar plate; the strains are listed in Table 3. Microscopic observation and colony morphology indicated that the isolates were fungi.

The sludge-assimilating activity of the isolates was examined by pure culture in WA medium for 1 week. All isolates grew in this medium and their growing hyphae were observed by microscopic analysis. To confirm sludge assimilation, we tried to determine the dry weight of the culture residue remaining following 1 week of culture. However, undegraded AS particles were tightly enveloped by growing hyphae in the culture residue, and thus, we could not separate them by centrifugation, filtration or tweezers. Therefore, undegraded sludge and hyphae were recovered together by centrifugation, dried and weighed as a total mass of culture residue after 1 week of cultivation. Approximately, 10–30% reduction in the dry weight of the culture residue was observed (Fig. 1a). Student's t-test (< 0·05) indicated that the weight obtained was significantly different from that obtained from abiotic culture. The residue mass in the 1, 2 and 4-week cultures was monitored, revealing that prolonged cultivation led to further weight reduction (Fig. 1b).

Figure 1.

Degradation of AS by pure culture of the isolates. (a) Relative weight of the culture residue after 1 week of cultivation in the culture. Mean and standard deviation of independent triplicates are shown. Columns with different letters indicate significant difference at < 0.05 (Student's t-test). (b) Time course for sludge assimilation by the isolate pure culture. Relative weight of the culture residue after 1, 2 and 4 weeks of cultivation in the culture is shown. (○) Umbelopsis sp. FernWA(AB813149); (image) Penicillium sp. CedarWA1 (AB813150); (●) Penicillium sp. CedarWA2 (AB813151); (□) Cunninghamella sp. CedarWA3 (AB813152); (image) Cunninghamella sp. CedarWA4 (AB813153); (■) Fusarium sp. OreWA(AB813154); (image) Neosartorya sp. OreYA (AB813155) and (image) Chaetomium sp. GalleryYA (AB813156).

Enzyme activity of sludge-assimilating strains

We examined which enzyme activities of the isolates had contributed to their sludge-degrading activity by assessment of cellulase, hemicellulase (xylanase), chitinase, laccase and keratinase activities. Table 4 shows the enzyme activity in the culture supernatant of each isolate. It was found that no strains showed activity for endocellulase, exocellulase or laccase, but all were positive for xylanase, chitinase and keratinase. Of note, the strain OreWA exhibited activity for both xylanase and chitinase, while OreYA exhibited the strongest keratinase activity among the strains tested. We also examined whether enzyme activity was affected by cultivation temperature. Among test temperatures (25, 30, 37°C), all isolates showed the highest enzyme activity at 30°C, while none grew at 45°C.

Table 4. Enzyme activities in the culture supernatant of each isolate
StrainsEnzyme activity (U/ml)
XylanaseaChitinasebKeratinaseb
  1. The values shown are the mean ± standard deviation of independent experiments performed in triplicate. The values with different letters indicate significant difference at < 0·05 (Student's t-test).

  2. a

    One unit corresponds to the amount of enzyme required to release 1 μmol reducing sugar produced per minute (Ghose 1987).

  3. b

    One unit corresponds to the amount of enzyme required to release azure dye that produced an increase of 0·01 in the absorbance per hour (Ramı́rez et al. 2004; Riffel et al. 2007).

Umbelopsis sp. FernWA (AB813149)0·32 ± 0·30 (a, b)8·90 ± 4·89 (c)10·14 ± 10·02 (g)
Penicillium sp. CedarWA1 (AB813150)0·23 ± 0·32 (a, b)7·38 ± 2·91 (c, d)6·18 ± 5·09 (g)
Penicillium sp. CedarWA2 (AB813151)0·35 ± 0·11 (b)5·42 ± 2·38 (c, d)14·65 ± 11·87 (g)
Cunninghamella sp. CedarWA3 (AB813152)0·05 ± 0·02 (a)20·50 ± 7·18 (e, f)63·24 ± 39·96 (h)
Cunninghamella sp. CedarWA4 (AB813153)0·05 ± 0·03 (a)21·83 ± 8·93 (e, f)64·97 ± 33·08 (h)
Neosartorya sp. OreWA (AB813154)0·32 ± 0·19 (a, b)27·08 ± 9·33 (e)83·86 ± 46·68 (h)
Fusarium sp. OreYA (AB813155)0·03 ± 0·01 (a)15·01 ± 7·65 (c, f)440·22 ± 258·79 (i)
Chaetomium sp. GalleryYA (AB813156)0·07 ± 0·02 (a)15·19 ± 5·54 (c, f)112·83 ± 60·41 (h)

Phylogenetic study of the isolates

The phylogenetic positions of the isolates were determined based on the ITS region sequence. Figure 2 shows the phylogenetic tree for the isolates and their closest known neighbours, as constructed using the neighbour-joining method. The isolates were classified into 6 genera, consisting of 3 ascomycetous and 3 Mucoromycotina genera. The ClustalW programme showed a high DNA sequence similarity (>99%) between OreWA and Neosartorya udagawae, OreYA and Fusarium solani, CedarWA1 and Penicillium janthinellum, CedarWA2 and P. janthinellum, CedarWA3 and Cunninghamella bainieri and CedarWA4 and C. bainieri. In contrast, sequence similarity between FernWA and Umbelopsis isabellina was 98·2% and that between GalleryYA and Chaetomium globosum was 95·4%.

Figure 2.

Phylogenetic tree of the isolates and known related species constructed using the neighbour-joining method, based on internal transcribed spacer (ITS) region sequences. The scale bar represents 0.1 base substitutions per nucleotide. Bootstrap values (Felsenstein 1985) above 70% (of 1,000 samplings) are shown at the internodes.

Discussion

Because biodegradable constituents of sewage sludge are consumed in anaerobic digestion, DS has low biodegradability and thus becomes a major industrial waste product in wastewater treatment plants. Although aerobic thermophiles with sludge-lysing activity were isolated as described earlier, their assimilation was incomplete and undegraded sludge residues still remained after termination of microbial reactions, supporting the biodegradation-resistant nature of DS. Cellulolytic bacteria are reported to reside in anaerobic digesters (O'Sullivan et al. 2005, 2007; Shiratori et al. 2006, 2009); however, their DS-assimilating activity is not well documented. Conversely, Ajwa and Tabatabai (1994) reported that a field application of DS weakly stimulated the respiration and growth of soil microbes, suggesting that some soil microbes may assimilate DS.

We first tried to enrich sludge-assimilating microbes using untreated DS as a substrate, but failed. Because aluminium and ferrous coagulants are generally employed to dewater sewage sludge (Lo et al. 2001), we presumed that these metals in DS would show some inhibitory effect on microbial growth. Hence, acid-treated DS (AS) was prepared and used for enrichment. Enrichment of DS-assimilating microbes was successful using AS, suggesting that removal of these metals from sludge impact their growth. The results obtained in this study showed that DS-assimilation rate of the isolates appears rather slow. It should however be noted that the sludge-assimilating activity of the isolates was underestimated in our experiments because the culture residue containing both undegraded sludge and fungal hyphae was weighed to estimate assimilation activity. If undegraded sludge could be separated from hyphae, estimation of their assimilation activity would be more precise.

Extracellular polymeric substances and cell-wall polysaccharides of microbial biomass, such as cellulose, hemicellulose (e.g. xylan) and chitin, in sludge are also rate-limiting factors in its biodegradation (Martens and Frankenberger 1991; Li et al. 2001). In addition, abundant human hair is found in DS. We therefore examined whether the isolates possess hydrolysing enzymes for those substrates: all the isolates were found to possess xylanase, chitinase and keratinase activities. Considering that the isolates grew by assimilating AS as the sole nutrient source, that they possess a suite of hydrolysing enzyme for sludge degradation, and that the residue in pure culture was further degraded with extended cultivation, these isolates should be sludge-assimilators and are suggested to contribute to sludge mineralization in soil. To the best of our knowledge, there have been no reports to date of the characterization of DS-assimilating microbes, and hence, our study is the first to report the isolation and identification of DS-assimilating strains. Because a temperature of 30°C was found to be optimal for growth and enzyme production of the isolates, they are mesophiles. Because previously reported sludge-lysing bacteria (Kim et al. 2002; Li et al. 2009; Liu et al. 2010, 2011) were thermophiles isolated from activated sludge, our fungal isolates appear to occupy an ecological niche different from the thermophilic degraders.

While many fungal strains were reported as cellulase producers (Kuhad et al. 2011 for review), cellulase activity was not found in any isolates. Because our experiment was confirmed working properly based on positive controls (Tricoderma reesei cellulase and Trametes vercicolor laccase), the isolates appear to lack cellulase. In this study, digested sludge, a residue of anaerobic digestion, was used. While raw activated sludge contains abundant cellulose as well as chitin and xylose, constituents of microbial cell wall, we think the digested sludge contains less cellulose than the raw sludge, because many researchers reported a wide distribution of cellulolytic anaerobes (mainly Clostridia) in anaerobic digesters (O'Sullivan et al. 2005, 2007; Shiratori et al. 2006, 2009). Cellulose should be thus hydrolysed at least to some extent by the cellulolytic anaerobes during the digestion and lead to produce mono- or di-saccharides, substrates for methanogens. Considering these facts together, digested sludge seems not a suitable substrate for isolation of cellulolytic strains. Raw sludge may rather be suitable for isolation of cellulase producers.

Phylogenetic analysis has revealed that the isolates belong to fungal genera found in soil microbiota. There are many reports on xylanase, chitinase and keratinase activities in the genus Penicillium. In particular, because P. janthinellum, the closest neighbour for the strains CedarWA1 and CedarWA2, is known to produce xylanase and chitinase (Milagres et al. 1993; Fenice et al. 1998), CedarWA1 and CedarWA2 are suggested to be P. janthinellum. The strain OreYA is thought to be F. solani because it is known to degrade xylan and keratin (Gupta et al. 2009; Kannahi and Ancy 2012). Although keratinase activity is also reported for C. globosum, the closest neighbour for the strain GalleryYA (Kaul and Sumbali 1999), DNA sequence similarity between them was low (95·4%). Because a difference of around 2% in ITS region DNA sequence between the tested strains suggests heterospecificity (Schoch et al. 2012), GalleryYA appears distinct from C. globosum. In contrast, there are no reports of xylanase, chitinase or keratinase activities for the genera Cunninghamella and Umbelopsis. Thus, the strains CedarWA3, CedarWA4 and FernWA are the first isolates of that genus to demonstrate association with such enzyme activities. Because FernWA made a distinct lineage from the closest known species U. isabellina and the sequence similarity of ITS region DNA between both strains was 98·2%, it is implied that FernWA is a novel species of the genus. Although there is a report of xylanase activity for the genus Neosartorya (N. spinosa), phylogenetic analysis has revealed that the closest known species for OreWA are N. udagawae and N. aureola. Moreover, because OreWA possesses chitinase and keratinase rather than xylanase, this report appears to be the first to document chitinase and keratinase activities in the genus Neosartorya.

Microbial enzymes including xylanase, chitinase and keratinase may become more important in various industries (Patil et al. 2000; Korniłłowicz-Kowalska and Bohacz 2011; Juturu and Wu 2012). Because DS represents a huge amount of waste produced by sewage treatment, the isolates described in this study could be useful in the commercial production of microbial enzymes using DS as substrate. Because xylan, chitin and keratin in the sludge–hyphae complex are thought to be partially depolymerized by the isolate enzymes, DS could also be utilized as a readily available fertilizer. Enhancement of culture conditions should be attempted to facilitate the application of DS-assimilating fungal strains in biotechnology and agriculture industries.

Acknowledgements

This study was funded by a Grant-in-Aid for Scientific Research (C: 24510101) from the Japan Society for the Promotion of Science.

Ancillary