- Email: [email protected]
Development of novel method for screening microorganisms using symbiotic association between insect (Coptotermes formosanus Shiraki) and intestinal microorganisms
JOURNAL OF BIOSCIENCE AND BIOENGINEERING Vol. 103, No. 4, 358–367. 2007 DOI: 10.1263/jbb.103.358
© 2007, The Society for Biotechnology, Japan
Development of Novel Method for Screening Microorganisms Using Symbiotic Association between Insect (Coptotermes formosanus Shiraki) and Intestinal Microorganisms Arata Hayashi,1 Hideki Aoyagi,1 Tsuyoshi Yoshimura,2 and Hideo Tanaka1* Life Science and Bioengineering Laboratory, Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan1 and Laboratory of Innovative Humanohabitability, Research Institute for Sustainable Humanospere, University of Kyoto, Gokasho, Uji, Kyoto 611-0011, Japan2 Received 30 November 2006/Accepted 18 January 2007
It is becoming increasingly difficult to isolate novel useful microorganisms from the natural environment using conventional screening methods based on pure culture techniques. A novel method for screening microorganisms in symbiotic association with insects was developed. This method involves the following two steps. In the first step, the existence of desired microorganisms that grow well by degrading difficult-to-degrade materials in the gut of insects is detected using the survivability of insects as an indicator. In the second step, the desired microorganisms are selected from the surviving insects. The second step is based on an idea that the guts of insects act as continuous-culture systems whereby microorganisms that cannot degrade diet components are washed out whereas those that can degrade diet components are retained and made to multiply in the gut. Coptotermes formosanus Shiraki was fed with an artificial diet containing phenol as a model of lignin-derived and difficult-to-degrade compound. Each C. formosanus feeding on an artificial diet containing 100 mg/l phenol had different levels of adaptation to the toxicity of phenol. About 20% of C. formosanus fed with an artificial diet containing 100 mg/l phenol died within a few days whereas others survived for more than 10 d. The structure of the intestinal microorganisms of the surviving C. formosanus fed with the 100 mg/l phenol artificial diet gradually changed and was very different from that of the bacterial communities obtained from the enrichment culture of wood-feeding C. formosanus using an artificial medium containing phenol as a sole carbon source. Furthermore, Only three species (as DGGE band) were detected from the gut of wood-feeding C. formosanus, whereas 200 times more phenol-degrading microorganisms were detected in the gut of C. formosanus feeding on a phenol artificial diet. Out of these nine species (as DGGE band) of phenol-degrading microorganisms were isolated. The screening method developed in this study can also be applied to various insects, leading to the isolation of various microorganisms that can degrade difficult-to-degrade compounds. [Key words: screening method, insect, Coptotermes formosanus Shiraki, artificial diet, intestinal microorganisms]
method of isolating new microorganisms. In natural ecosystems, most microorganisms exist by maintaining strong relationships with other organisms. Examples of such relationships include termite-intestinal microorganism symbiosis (2), rhizobia-legume symbiosis (3) and lichen symbiosis (4). It is possible to develop an efficient method of screening the desired microorganisms using the symbiotic association between microorganisms and other organisms. It is known that intestinal microorganisms play important roles in the degradation of diet components of insects. The products of microbial degradation are continuously removed by absorption in the gut wall or excretion through the vent. In this way, the gut of insects can be likened to a continuous-culture system whereby microorganisms with low growth rate (those that cannot utilize diet components) are
In the 19th and 20th century, many useful microorganisms were isolated by using pure culture techniques, and this led to the rapid development of industrial microbiology. However, it is becoming increasingly difficult to use pure culture techniques to isolate new microorganisms. This has significantly limited the progress of microbiology. Various molecular techniques have revealed that more than 99% of microorganisms existing in nature cannot be isolated and maintained in pure cultures (1). To overcome this problem, the development of novel screening methods is very important. We have been focusing on symbiotic associations among microorganisms with the aim of developing a novel * Corresponding author. e-mail: [email protected] phone: +81-(0)29-853-7212 fax: +81-(0)29-853-4605 358
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
washed out, whereas microorganisms with high growth rate (those that can efficiently utilize diet components) are retained and become the predominant species in the gut. We have been studying the symbiotic associations between insects and intestinal microorganisms, which are known to be mediated by their diets (diet–insect–intestinal microbial community), using Coptotermes formosanus Shiraki as a model organism. In such studies, C. formosanus is fed with artificial diets and changes in the intestinal microorganisms in response to changes in the components of the diet are monitored. Our preliminary results showed that when the diet component of C. formosanus (wood-feeding insects) was changed from wood to glucose, their intestinal microorganisms gradually changed from wood-degrading to glucose-degrading microorganisms, and finally, they lost their wood-degrading ability (5). This result suggests that if the components of insect diets are changed, the intestinal microbial community in the insect changes and the microorganisms that are capable of degrading the components of the diet become the predominant species. In other words, microorganisms that are capable of degrading a difficult-todegrade compound can be made to multiply in the intestine of an insect that survives by feeding the insect with an artificial diet containing the difficult-to-degrade compound as the sole carbon source. In this study, we developed a novel screening method that can be used to isolate the desired microorganisms (diet component-degrading microorganisms) by controlling the components of the insect diet. C. formosanus was fed with artificial diets containing phenol as a model of the compounds that are difficult to degrade. During the feeding, about 20% of the termites died but some survived. Changes in the microbial community in the gut of the surviving C. formosanus were analyzed by microscopic observation and using the PCR-denaturing gradient gel electrophoresis (PCRDGGE) method. Subsequently, screening of phenol-degrading microbes was performed using both agar medium and membrane filter methods. Furthermore, the number of colonies formed on membrane filters was counted and species diversity was analyzed using the Plate Wash (PW) PCR (6)DGGE method. From these experiments, it was demonstrated that many species of phenol-degrading microorganisms could be isolated using this screening method. This method can be used for the efficient screening of various difficult to degrade compound-degrading microorganisms. MATERIALS AND METHODS Termites Worker caste termites were collected from a laboratory colony of C. formosanus maintained in the Laboratory of Innovative Humano-habitability, Research Institute for Sustainable Humanosphere, Kyoto University, Japan. They were maintained on wood pieces from the Japanese red pine (Pinus densiflora Sieb) at 30 ± 2°C. Preparation of artificial diets and feeding experiments The preparation of artificial diets was performed according to the method described by Tanaka et al. (5). Agarose (15 g/l) (Takara Shuzo, Otsu) was dissolved in distilled water and autoclaved at 121°C for 15 min. Phenol (50 g/l) was sterilized by passing through φ0.22 µm Durapore filter (Millipore, Bedford, MA, USA) and mixed with the sterilized agarose solution to obtain various phenol
359
concentrations (50, 100, 150 and 200 mg/l). The solutions were solidified in a sterilized glass petri dish. As controls, artificial diets containing 15 g/l agarose only, or 15 g/l agarose and the wood powder of Pinus densiflora (less than 100 µm) were prepared. The solidified agarose gels were cut into blocks aseptically (one block = 25 ×25 ×5 mm). Workers caste termites (25 in number) and one block of the artificial diet were placed inside a sterilized polystyrene case (30 ×30 ×10 mm), and incubated at 30°C. As a starvation condition, C. formosanus individuals were placed inside a sterilized polystyrene case under a humid condition (the case was placed inside a sterilized glass petri dish with sufficient sterilized distilled water). During incubation, the number of viable termites was counted periodically and their survival ratios were calculated. Survival ratio represents the number of surviving C. formosanus as a percentage of the initial number (25 termites). The weight of C. formosanus was also measured. Ten C. formosanus individuals were randomly picked and weighted on a chemical balance, and their average weights were calculated. All experiments were performed in triplicate. Observation and enumeration of protozoa and bacteria Different species of intestinal protozoa of C. formosanus were microscopically observed and counted periodically. Three termite workers were fed with any of the diets and collected randomly. Their hindguts were pulled out from their posterior ends and minced into pieces using fine forceps. The hindgut contents were then suspended in 30 µl of Trager U solution (7). The hindgut pieces were macerated gently in the solution to facilitate the release of protozoa. Different species of protozoa were identified using a light microscope. The number of each species of protozoa was counted using a hemocytometer (Kayagaki Irikakogyo, Tokyo). The number of bacteria was also counted using a bacterial counter (Kayagaki Irikakogyo). Cultivation of intestinal microorganisms on agar plates Intestinal microorganisms were cultivated on agar plates to quantify phenol-degrading microorganisms isolated from the intestine of C. formosanus after feeding with various diets for 10 d. The hindguts of the insects were pulled out from their posterior ends using fine forceps. The hindguts were homogenized using a pellet mixer in 1 ml of MP500 medium containing 500 mg/l phenol (8). The gut homogenate (100 µl) was diluted with 900 µl of MP500 medium. This process was repeated several times and the gut homogenate of a desired concentration was prepared. The diluted gut homogenate was spread onto an MP500 agar plate containing MP500 medium in 1.5% agar gel, and incubated at 30°C for 10 d. After incubation, colonies forming on the MP500 agar plate were counted, and colony forming unit per gut (CFU/gut) was calculated. All experiments were performed in triplicate. Cultivation of intestinal microorganisms using membrane filter method When cultivation was performed on an agar plate, it was observed that some impurities in the agar affected the results. Microbes that cannot degrade phenol but are tolerant to 500 mg/l phenol also formed colonies. Intestinal microorganisms were therefore cultivated using the membrane filter method to determine phenol-degrading microorganisms only. This system contained phenol as the sole carbon source. Gut homogenate was prepared as described above. A Milliflex filter funnel unit, HVWP 0.45 µm (Millipore), was set on a suction pump. The MP500 medium (5 ml) was initially poured into a funnel, and 100 µl of the gut homogenate was placed into the funnel and homogeneously mixed with the MP500 medium. The liquid fraction was filtered out and the intestinal microbes were collected on a membrane filter. The membrane filter was set on a liquid medium cassette, MXLMC0120 (Millipore) filled with MP500 medium, and incubated at 30°C for 10 d. After the incubation, colonies forming on the membrane filter were counted, and CFU/gut was calculated. All experiments were performed in triplicate. The ratio of phenol-degrading microorgan-
360
J. BIOSCI. BIOENG.,
HAYASHI ET AL.
isms to the total number of bacteria was calculated using Eq. 1. Ratio = CFU/gut (membrane filter method)/ Total number of bacteria
(1)
Determination of phenol-degrading ability of microorganisms Microbes that formed colonies were cultivated in liquid cultures to determine their phenol-degrading ability. The colonies were picked with a toothpick and inoculated into 5 ml of MP500 medium in L-shaped test tubes (Taitec, Saitama), and incubated on a shaker (30°C, 30 strokes/min) for 2–7 d. After the incubation, the phenol concentration in the liquid culture was measured using Phenol Test Wako (Wako, Osaka), and the microorganisms that degraded phenol were identified as phenol-degrading microbes. Enrichment culture The guts of wood-feeding C. formosanus (five in number) were pulled out from the posterior ends using a pair of fine forceps and suspended in 5 ml of MP500 medium. The gut suspension was poured into an L-shaped test tube (Taitec) and incubated at 30°C and 30 strokes/min. Cell growth was periodically measured using a Spectronic 20A spectrophotometer (Shimadzu). Phenol concentration in the culture was measured using Phenol Test Wako (Wako). When phenol in the culture broth was completely consumed, microorganisms in the culture were collected by centrifugation (10,000 ×g, 5 min). PCR-DGGE analysis Changes in the components of the intestinal microorganisms of C. formosanus fed with artificial diets were analyzed using the PCR-DGGE method (9). The analysis was performed on C. formosanus fed with an agarose artificial diet, a phenol artificial diet or wood for 10 d. Ten C. formosanus individuals were collected randomly from each set, and their hindguts were pulled out from their posterior ends using a pair of fine forceps. The hindguts were homogenized using a Fast DNA kit (Bio 101, Vista, CA, USA), and total DNA of the intestinal microbial cells were extracted. The V3 region of the bacterial 16S rDNA fragment was amplified using 357F-GC (5′-CGCCCGCCGCGC GCGGGCGGGGCGGGGGCACGGGGGGCCTACGGGAGGC AGCAG-3′) and 518R (5′-GTATTACCGCGGCTGCTGG-3′) as PCR primers. Each amplification reaction mixture (20 µl) consisted of 0.5 U Ex Taq DNA polymerase (Takara Shuzo), 1 µl of total DNA solution, 2 µl of 10 ×PCR buffer, 0.25 µM of each primer, and a mixture containing 100 µM deoxynucleoside triphosphate. A touchdown program (9, 10) was implemented as follows: after the initial denaturation at 94°C for 5 min, 30 cycles of 94°C for 1 min and 72°C for 1 min were performed, and then the reaction mixture was kept at 72°C for 7 min. During the reaction cycle, the annealing temperature was decreased by 1°C from 65 to 56°C every second cycle in the first 20 cycles. The annealing temperature was 55°C in the last 10 cycles. The amplicons were purified with QIA quick (Qiagen, Crawley, UK), and DNA concentration was determined by measuring the OD260 of the purified amplicon solutions. For DGGE, 250 ng of the purified amplicons was used. DGGE was performed using a D-code system (Bio-Rad Laboratories, Hercules, CA, USA). Acrylamide (8%) gels were prepared and electrophoresed with 0.5 ×Tris–acetate–EDTA (TAE) buffer (1 ×TAE buffer consists of 0.04 M Tris base, 0.02 M sodium acetate, and 10 mM EDTA [pH 7.4]). The DGGE gel contained 20–70% gradients of urea and formamide in the direction of electrophoresis as denaturants. The 100% denaturant consisted of 40% (v/v) formamide and 7 M urea. DGGE was performed at a constant voltage of 35 V at 60°C for 24 h. A mixture of amplified 16S rDNA fragments was used as a synthetic marker, consisting of amplified 16S rDNA fragments of Arthrobacter atrocyaneus ATCC 13752, clones obtained from seawater samples, and species related to Balneatrix alpica, Agrobacterium gelatinovorum, Methylomicrobium album, Sargasso Sea-proteobacterium, Sulfitobacter pontiacus, Erythrobacter sp. and the plastid of Emiliania huxleyi. The gel was stained with SYBR Gold Nucleic Acid Gel stain (Molecular Probes,
Eugene, OR, USA) and photographed using a UV transilluminator. The DGGE band profile was analyzed using an image-analyzing system (Image master, Amersham Pharmacia Biotech, Uppsala, Sweden), and the density and migration distance of the bands were calculated. The similarity between each DGGE band pattern in each lane was calculated using Eq. 2 (11). Cs = 2j/(a + b)
(2)
Here, j is the number of common bands between gel lanes A and B, and a and b are the total numbers of bands in gel lanes A and B, respectively. From the observation of the images of the DGGE gel, two bands (bands A and B shown in Fig. 7) that were observed only on the lane of C. formosanus fed with the phenol artificial diet were carefully excised with a razor blade under UV illumination and then placed in a 100 µl of Tris-EDTA buffer. DNA was extracted from the gel pieces by incubation at 4°C for 3 d, and then 0.5 µl of the supernanant was used as the template DNA in a touchdown PCR performed with primers GC-2 (5′-GAAGTCATCATGACCGTTC TGGCACGGGGGGCCTA-3′) (10) and 518R. The amplification mixture and conditions of the touchdown PCR were the same as those used for the amplification of DNA for DGGE as described above. The resulting amplicons were electrophoresed again on a DGGE gel to verify the position of the original band. Subsequently, the amplicons were purified using QIA quick (Qiagen). They were then cloned into the Pgem-T Easy Vector System (Promega, Madison, WI USA). Ligation products were transferred into competent cells of Escherichia coli DH5-α. White colonies were randomly picked and screened directly for inserts by colony PCR using primers for the vector (primers T7 and SP6). Plasmid DNA was prepared from the clones using a QIA prep spin miniprep kit (Qiagen). Plasmid DNA was then sequenced according to the direction of insertion by using CEQ 2000 XL (Beckman, Fullerton, CA, USA). All the sequences were analyzed using Seqman (DNAstar, Madison, WI, USA). Subsequently, the sequences were compared with similar sequences of reference organisms by performing a BLAST search (12, 13). Sequence data were aligned with the CLUSTAL W package (14). Plate Wash (PW)-PCR-DGGE analysis It was demonstrated that phenol-degrading microorganisms were increased number in the gut of C. formosanus fed with the phenol artificial diet using the membrane filter method. However, their specific names and diversity were not clarified. Therefore, an efficient analysis of all phenol-degrading microorganisms grown on a membrane filter was performed. All bacterial colonies, which are formed on the membrane filters by cultivating the intestinal microorganisms of C. formosanus fed with wood and the phenol artificial diet were suspended in sterilized distilled water and placed into a 1.5-ml microtube. The suspensions were collected by centrifugation (12,000 ×g, 10 min). The total DNAs of all bacterial colonies were extracted using the Fast DNA kit, and analyzed using the PCR-DGGE method as described above. Nucleotide sequence accession number The nucleotide sequences of partial 16S rRNA genes have been deposited in the DDBJ nucleotide sequences databank under the following accession numbers: band A-1, AB277362; band A-2, AB277363; band A-3, AB277364; band B, AB277365; Wf-1, AB277366; Wf-2, AB277367; Wf-3, AB2773628; Pf-1, AB277369; Pf-2, AB277370; Pf-3, AB277371; Pf-4, AB277372; Pf-5, AB277363; and Pf-6, AB277364.
RESULTS AND DISCUSSION Effects of artificial diets on survival ratio of C. formosanus Figure 1 shows the effects of artificial diet compo-
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
nents on the survival ratio of C. formosanus. The survival ratios of C. formosanus that were fed with wood powder, agarose artificial diet, or starved, were 95%, 80% or 57.3%, respectively, after 20 d. On the other hand, the survival ratios of C. formosanus that were fed with artificial diets containing 50 and 100 mg/l phenol were 38.7% and 17.3%, respectively. All the C. formosanus fed with artificial diets containing 150 and 200 mg/l phenol died within 15 d. Although the termites were observed eating the wood powder, agarose, and artificial diets containing 50 and 100 mg/l phenol, the termites did not eat the artificial diets containing 150 and 200 mg/l of phenol. This implies that only a very small amount of the artificial diet containing 150 and 200 mg/l phenol and thus phenol, were taken into their gut. These results show that the inclusion of phenol in the artificial diet of C. formosanus reduces their survival ratio because phenol is a toxic substance that inhibits the function of the cytoplasm (15). The amount of decrease in the survival ratio of C. formosanus fed with the phenol artificial diet was higher than those in C. fomosanus fed with agarose and C. formosanus that were starved. This result shows that the main reason for the rapid decrease in the number of C. formosanus fed with the phenol artificial diet is not starvation but phenol toxicity. It is interesting to note that although some of the C. formosanus died within a few days, others survived for a long period of time. Thus, it appears that C. formosanus individuals differ in their susceptibility to phenol toxicity although they were from the same parents and lived in the same colony. The survival levels of C. formosanus fed with artificial diets containing phenol may be considered as gradual steps that involve the following stages: (i) their tolerance to toxicity of phenol, (ii) detoxification of phenol due to the presence of some phenol-degrading microorganisms in the gut of C. formosanus feeding on wood, and (iii) increase in their phenol-degrading ability due to the multiplication of phenol-degrading microorganisms in the gut of C. formosanus fed with the phenol artificial diet. It is likely that only C. formosanus individuals that
FIG. 1. Survival ratio of C. formosanus fed with various artificial diets. Symbols: closed circles, wood powder; open circles, agarose; closed squares, 50 mg/l phenol; open squares, 100 mg/l phenol; closed triangles, 150 mg/l phenol; open triangles, 200 mg/l phenol artificial diet; closed diamonds, starvation. Values are means ± SD from three independent experiments.
361
have phenol tolerance can survive initially. Subsequently, phenol-degrading microorganisms multiply in their gut, prolonging their survival. We propose that the survival period of C. formosanus can be used as an indicator to distinguish individuals harboring phenol-degrading microorganisms from the rest. Thus, the screening method developed in this study is an efficient and novel method by which insects that survive on artificial diets containing phenol successfully multiply desired microorganisms in their intestine. These desired microorganisms are then isolated from the surviving insects. Effects of artificial diets on weight of C. formosanus Figure 2 shows the changes in the weight of C. formosanus during the feeding experiments. The initial mean weight of C. formosanus fed with wood (the feeding time on artificial diets was 0 d) was 3.94 mg. After 20 d, the weights of C. formosanus feeding on wood powder, agarose and the artificial diet containing 100 mg/l phenol decreased to 3.42, 2.41 and 2.22 mg, respectively. These results show that the survival of C. formosanus is inhibited by phenol, and that in comparison with wood powder, both agarose and phenol artificial diets decreased the weight of the termites. Effects of artificial diets on intestinal protozoan community of C. formosanus The changes in the total number of intestinal protozoa of C. formosanus fed with wood, agarose and artificial diets containing 100 mg/l phenol are shown in Fig. 3. There was a rapid decrease in the number of intestinal protozoa of C. formosanus fed with both agarose and phenol artificial diets. All the Pseudotrychonympa grassi (large protozoa) and Holomastigotoides hartmanni (middle protozoa) disappeared after 5 d and all the Spyrotrychonympha leidyi (small protozoa) disappeared after 10 d of feeding on these artificial diets. The disappearance of these protozoa was consistent with the decrease in the weight of the termites, as shown in Fig. 2. It has been reported that intestinal protozoa make up approximately 1/3 of the total weight of lower termites (16). In this study, the weights of C. formosanus individuals fed with agarose and phenol artificial diets were reduced to more than 1/3 of their weights. Therefore, aside from the effect of the physical weight of
FIG. 2. Changes in body weight of C. formosanus fed with various artificial diets with time. Symbols: closed circles, wood powder; open circles, agarose; open squares, phenol (100 mg/l) artificial diet. Values are the means ± SD from three independent experiments.
362
HAYASHI ET AL.
FIG. 3. Changes in total number of protozoa in hindgut of C. formosanus fed with various artificial diets with time. Symbols: closed circles, wood powder; open circles, agarose; open squares, phenol (100 mg/l) artificial diet.
FIG. 4. Changes in total number of bacteria in hindgut of C. formosanus fed with artificial diets. Symbols: closed circles, wood powder; open circles, agarose; open squares, phenol (100 mg/l) artificial diet.
the protozoa, their disappearance might have other physiological effects that result in the further decrease in the weight of C. formosanus. Effects of artificial diets on total number of intestinal bacteria in C. formosanus The changes in the total number of intestinal bacteria of C. formosanus fed with wood powder, agarose and artificial diets containing 100 mg/l phenol are shown in Fig. 4. There was a rapid decrease in the number of intestinal bacteria of C. formosanus fed with both agarose and phenol artificial diets. The total number of bacteria was initially 4.0 ×107 cells/gut (wood feeding) but decreased to less than 1/10 (2.8 ×106 cells/gut) when they were fed with a phenol artificial diet. Detection and determination of phenol-degrading microorganisms from intestinal microorganisms of C. formosanus using MP500 agar plate Figure 5 shows the number of CFU/gut obtained by plating the intestinal microorganisms of C. formosanus on an MP500 agar plate after feeding with wood, agarose and 100 mg/l phenol artificial diets for 10 d. The numbers of the CFU/gut of C. formosanus fed with wood, agarose and 100 mg/l phenol were 3.4 × 104, 5.4 ×104 and 1.9 ×105, respectively. Subsequently, 85 microbial colonies from the intestinal microorganisms of C. formosanus fed with the 100 mg/l phenol artificial diet
J. BIOSCI. BIOENG.,
FIG. 5. Number of colony forming units obtained per gut of C. formosanus fed with various artificial diets. The colonies were grown on MP500 agar plate. Values are means ± SD from three independent experiments.
were randomly picked and their phenol-degrading abilities were investigated. The results show that 83 out of the 85 colonies cannot degrade phenol. Thus, although a very high CFU/gut was obtained from the guts of C. formosanus individuals fed with the 100 mg/l phenol artificial diet, most of these colonies (97.6%) showed no phenol degrading ability. Thus, the MP500 agar plate was not appropriate for the detection and quantification of phenol-degrading microorganisms because nonphenol-degrading microorganisms also formed colonies on the MP500 agar plates. With the MP500 agar plate, all the microorganisms that can be cultured are tolerant to 500 mg/l of phenol, and can metabolize phenol, agar or agar-derived oligosaccharides to form colonies. Thus, the MP500 agar plate method cannot be used to isolate phenol-degrading microorganisms only. Detection and quantification of phenol-degrading microorganisms from intestinal microorganisms of C. formosanus using membrane filter method With the conventionally used MP500 agar plate method, microorganisms that did not degrade phenol but consumed other medium components such as agar and agar-derived oligosaccharides also formed colonies. Thus, to detect and quantify phenol-degrading microorganisms only, the membrane filter method using only phenol as the carbon source was used. In this method, agar was not used, and the microorganisms were collected on the membrane filter while the liquid medium was supplied through the membrane. In this way, only microorganisms that degrade and metabolize phenol contained in the liquid medium can grow and form colonies. Figure 6 shows the number of CFU/gut formed on a membrane filter. The intestinal microorganisms of C. formosanus fed with wood, agarose and 100 mg/l phenol artificial diets for 10 d were collected on a membrane filter and saturated with MP500 medium. Few colonies (200 CFU/gut) were detected from the intestinal microorganisms of C. formosanus fed with the wood and agarose artificial diets. On the other hand, colonies at 2.8 ×103 CFU/gut were detected from the intestinal microorganisms of C. formosanus fed with the 100 mg/l phenol artificial diet. Fifty colonies were randomly picked from the colonies and their phenol-degrading ability was investigated. The results showed that all the 50 colonies degraded phenol. Furthermore, the membrane
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
363
FIG. 6. Number of colony forming units obtained per gut of C. formosanus fed with various artificial diets. The colonies were grown on a PVDF membrane filter supplemented with MP500 liquid medium. Values are means ± SD from three independent experiments.
filter method was used to compare the number of CFU/gut of the microorganisms from the gut of C. formosanus fed with the 50 mg/l phenol artificial diet with that of the microorganisms from the C. formosanus fed with the 100 mg/l phenol artificial diet. The number of CFU/gut of C. formosanus fed with the 100 mg/l phenol artificial diet was high (2.8 ×103 CFU/gut), but only few colonies (5.0 ×102 CFU/gut) were detected in the microorganisms from the gut of C. formosanus fed with artificial diets containing 50 mg/l phenol. This implies that the density of phenol-degrading microorganisms in the gut of termites increases with increasing phenol concentration in their diet. Therefore, for the multiplication of phenol-degrading microorganisms in the gut of C. formosanus, more phenol should be added to the artificial diet but the phenol concentration should not be very high as to markedly reduce feeding activity and thus affect survival. Table 1 shows the total number of bacteria and phenoldegrading microorganisms detected using the membrane filter method and the ratio of phenol-degrading microorganisms in gut of C. formosanus fed with wood and phenol artificial diets for 10 d. The ratios of the intestinal phenoldegrading microorganisms in C. formosanus individuals feeding on wood and phenol artificial diets were 5.0 ×10–6 and 1.0 ×10–3, respectively. These two ratios suggested that the number of phenol-degrading microorganisms were increased by about 200-fold by feeding C. formosanus with the phenol artificial diet. These results showed that the membrane filter method is effective in detecting phenol-degrading microorganisms only, and that the number of phenol-degrading microorganisms is increased in the gut of C. formosanus fed with the artificial diet containing 100 mg/l phenol.
FIG. 7. Image of denaturing gradient gel electrophoresis (DGGE) of 16S rDNA of intestinal bacteria of C. formosanus fed with various artificial diets for 10 d, and bacteria obtained by enrichment culture of intestinal microorganisms of C. formosanus. Lanes 1 and 6, Synthetic markers consisting of amplified 16S rDNA fragments; lane 2, wood; lane 3, agarose; lane 4, phenol; lane 5, enrichment culture with MP500. Bands A and B, Two excised bands for sequence.
PCR-DGGE analysis By using the membrane filter method, it was demonstrated that the number of cultivatable phenol-degrading microorganisms increased in the gut of C. formosanus fed with the phenol artificial diet. However, many microorganisms that cannot be cultured may also be present in the gut of such termites. PCR-DGGE, which is generally used for the analysis of the bacterial community structure, was therefore used to analyze all the bacteria in the gut of the termites. Furthermore, the intestinal microorganisms of C. formosanus fed with the phenol artificial diet was compared with those obtained by the conventional enrichment culture. Figure 7 shows the band patterns of the intestinal bacterial community of C. formosanus fed with the three types of diets, and those of the intestinal microorganisms of wood-feeding C. formosanus obtained by the enrichment culture. Figure 8 shows the dendrogram based on the differences and similarity of band patterns among the bacterial community using Sorensen’s equation. The similarity between the bacterial community in the C. formosanus fed with wood and that fed with the phenol artificial diet was
TABLE 1. Comparison of total bacterial number, number of phenol degrading bacteria and ratio of phenol degrading-bacteria in intestine of C. formosanus fed with wood and phenol artificial diets Total number of bacteria (cells/gut) (A) Wood 4.0 ×107 2.8 ×106 Phenol artificial dietb a Obtained using membrane filter method. b Feeding time: 10 d. Diet
Number of phenol-degrading bacteria (CFU/gut)a (B) 2.0 ×102 2.8 ×103
Ratio of phenol degrading bacteria (B/A) 5.0 ×10–6 (C) 1.0 ×10–3 (D)
D/C – 200
364
J. BIOSCI. BIOENG.,
HAYASHI ET AL.
FIG. 8. Cluster analysis (UPGAMA method) of DGGE band pattern obtained from termites fed with various artificial diets (scale represents the percentage of divergence).
35%, whereas the similarity between the bacterial community in the C. formosanus fed with phenol and that fed with agarose artificial diets was 58%. The similarity between the microorganisms of C. formosanus fed with the phenol artificial diet and those obtained by the enrichment culture was only 12%. These results showed that the structure of the bacterial community of C. formosanus varied greatly depending on the type of artificial diet. However, the intestinal bacterial community of C. formosanus fed with the phenol artificial diet was comparatively similar to that of C. formosanus fed with the agarose artificial diet. Feeding C. formosanus with both diets resulted in the disappearance of intestinal protozoa and a decrease in the total number of bacteria (less than 1/10) after 10 d. It is well known that in the gut of C. formosanus, there is a strict symbiotic association between protozoa and many bacteria (see Ref. 2 and references therein). One important function of intestinal protozoa is phagocytosis (16). Thus, the disappearance of protozoa in the gut of C. formosanus fed with an artificial diet containing phenol affected the intestinal bacterial community. In other words, the disappearance of the intestinal protozoa may be one important factor that led to increase in the number of phenol-degrading microorganism in the gut of C. formosanus. Moreover, these results showed that the bacterial community obtained from the gut of C. formosanus fed with the phenol artificial diet was different from those obtained by the enrichment culture using MP500 medium. In this study, the enrichment culture was performed in batches, using an L-shaped test tube under aerobic condition (under anaerobic condition, there were no growth of microorganisms and phenol consumption). Consequently, the culture conditions such as phenol concentration and pH of the medium varied during the cultivation. On the other hand, in the gut of C. formosanus, nutrients were supplied continuously to the intestinal microorganisms (by the feeding activity of C. formosanus), whereas the products of microbial metabolism such as organic acid were continuously removed by absorption into the gut wall or excretion through the vent. Thus, the gut of C. formosanus can be likened to a continuous-
culture system. Microorganisms with low growth rate (those that cannot utilize the diet components) are washed out, whereas microorganisms with high growth rate (those that can efficiently utilize the diet components) are retained and become the predominant species in the gut. The number of bands obtained from the intestinal bacterial community of C. formosanus fed with the phenol artificial diet was larger than that obtained by the enrichment culture (Fig. 7). This result shows that the species of microorganisms from the gut of C. formosanus fed with the phenol artificial diet are more varied than the species of microorganisms obtained by the conventional enrichment culture. It has been reported that there is an oxygen concentration gradient in the gut of termites (17). Thus, this variation in intestinal environment enabled many microorganisms to adapt to each environment (oxygen concentration) and grow in the gut of C. formosanus. The DNA sequences of bands A and B shown in Fig. 7 were analyzed. Bands A and B were the only bands obtained from the intestinal microorganisms of the termites fed with the phenol artificial diet. The alignment of DNA sequences is shown in Fig. 9. Table 2 shows the strains and accession number of the bacteria that had the highest homology with each DNA sequence in bands A and B. One of the sequences in band A had a 98% homology with Pseudomonas nitroreducens (EF107515). This bacterial strain has been reported as 1,2,4-trichlorobenzene degrader. Although bacteria that degrade aromatic compounds such as phenol and 1,2,4-trichlorobenzene hardly exist in the gut of wood-feeding C. formosanus, they increased in number in such a high concentration to be detected as DNA bands when C. formosanus was fed with the phenol artificial diet. Other DNA sequences contained in bands A and B had high homologies with uncultivatable Crostridium and Bacteroides, and their functions are not yet known. Phenol degradation (particularly complete degradation to CO2 and H2O) occurs under aerobic condition, but the above results show that strictly anaerobic bacteria also exist in the intestinal microorganisms of termites fed with phenol artificial diet. Their functions are unknown but they could degrade phenol. Phenol degradation has been reported to occur mainly under aerobic condition (18), but anaerobic degradation has also been reported (19). This suggests that anaerobic degradation of phenol may occur in the gut of C. formosanus by strictly anaerobic bacteria such as Clostridium and Bacteroides. PWPCR-DGGE analysis Figure 10 shows the band patterns of all phenol-degrading bacteria obtained by the membrane filter method from the intestinal microorganisms of C. fomosanus fed with wood and 100 mg/l phenol artificial diet. The alignment of DNA sequences is shown in Fig. 11. Table 3 shows the numbers of DGGE band detected on each lane and the DNA sequences of the DGGE bands. The number of DGGE band detected on each lane corre-
TABLE 2. Sequencing results of DGGE band (Fig. 7) Band A-1 -2 -3 B
Bacterial strain showing highest similarity Uncultured bacteroidaceae bacterium, clone Rs-E83 Uncultured bacterium, clone RsaHw538 Pseudomonas nitroreducens Uncultured anaerobic bacterium, clone B-4C
Accession no. AB088937 AY571428 EF107515 AY953243
Similarity 89% 92% 98% 100%
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
365
FIG. 9. Alignment of partial sequences of 16S rDNA cloned from excised bands from DGGE gel (Fig. 7) and reference bacteria in database (GenBank).
FIG. 10. Image of denaturing gradient gel electrophoresis (DGGE) of 16S rDNA of all bacterial colonies grown on membrane filter from intestinal microorganisms of C. formosanus fed with wood and 100 mg/l phenol. Lane 1, Wood; lane 2, phenol. The arrows show the detected bands and the DNA of the nominated bands were sequenced.
lated with the number of species of the phenol-degrading bacteria. The total number of bacterial species (as DGGE band) detected from C. formosanus fed with wood was three. On the other hand, that from C. formosanus fed with the phenol artificial diet was nine. These results showed that the num-
ber of species of phenol-degrading microorganisms isolated from gut of C. formosanus fed with the phenol artificial diet (9 DGGE bands) was larger than that of the species isolated from the gut of wood-feeding C. formosanus (conventional method) (3 DGGE bands). DGGE Bands Wf-1, Wf-2 and Wf-3 detected in colonies from the intestinal microorganisms of wood-feeding C. formosanus showed high homology with Salmonella paratyphi strain 50973 (homology = 99%), Citrobacter sp. SVUB3 (97%) and Citrobacter sp. SVUB3 (99%), respectively. DGGE bands Pf-1, Pf-2, Pf-3, Pf-4, Pf-5 and Pf-6 detected in colonies from the intestinal microorganisms of C. formosanus fed with the phenol artificial diet showed high homology with uncultured Pseudomonadales bacterium (99%), Pseudomonas sp. pDL01 (100%), uncultured Pseudomonadales bacterium (99%), uncultured Pseudomonadales bacterium (100%), Pseudomonas aeruginosa strain PD100 (99%) and Enterobacteriales bacterium H0407 (100%). From these results, it can be inferred that many species of Pseudomonas, which are known as aromatic compound degraders, were detected in colonies from the intestinal microorganisms of C. formosanus fed with the phenol artificial diet. The main reason for this difference is that few phenol-degrading microorganisms (below the detection limit), which exist in wood-feeding C. formosanus, grow and multiply in the gut (as a continuousculture system) by feeding on phenol artificial diet. Therefore, microorganisms, which are originally present in the gut of insects but in small number, multiply by feeding on a specific diet and thus can be isolated by using this screening method. There was only one PWPCR-DGGE band (in the same position) that is common to wood-feeding and phenol artificial diet-feeding C. formosanus, but the bacterial species were different. It is possible that phenol-degrading bac-
366
HAYASHI ET AL.
J. BIOSCI. BIOENG.,
FIG. 11. Alignment of partial sequences of 16S rDNA cloned from excised bands from DGGE gel (Fig. 10) and reference bacteria in database (GenBank).
TABLE 3. Total numbers of DGGE bands (Fig. 10) and the DNA sequences of bacteria obtained from intestinal microorganisms of C. formosanus fed with wood and 100 mg/l phenol artificial diet Band Strain showing highest similarity Accession no. Similarity Wf-1 Salmonella paratyphi strain 50973 DQ683179 99% Wf-2 Citrobacter sp. SVUB3 AM401577 97% Wf-3 Citrobacter sp. SVUB3 AM401577 99% Pf 9 Pf-1 Uncultured Pseudomonadales bacterium DQ171526 99% Pf-2 Pseudomonas sp. pDL01 AF125317 100% Pf-3 Uncultured Pseudomonadales bacterium DQ171526 99% Pf-4 Uncultured Pseudomonadales bacterium DQ171526 100% Pf-5 Pseudomonas aeruginosa strain PD100 AY825034 99% Pf-6 Enterobacteriales bacterium H0407 DQ822788 98% The membrane filter method was used in this experiment. Wf, Total number bacterial colonies obtained by membrane filter method from woodfeeding C. formosanus; Pf, total number of bacterial colonies obtained by membrane filter method from C. formosanus fed with phenol artificial diet. Wf
Total band no. 3
teria with high growth rate but an initially small number grew and became predominant, whereas phenol-degrading bacteria with low growth rate were washed out even if their
initial number was high. Because all screening experiments carried out in this study were performed under aerobic condition, screening under anaerobic condition may lead to the
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
isolation of more phenol-degrading microorganisms. From these results, the screening method described in this study (increasing the number of microorganisms that degrade a particular compound in the gut of an insect by feeding the insect with the compound as the sole carbon source) can be used to isolate more diverse microorganisms. Some microorganisms that cannot be isolated by conventional enrichment culture may be isolated using this novel method. This method can also be applied to other insects, and by feeding artificial diets containing any target compound, microorganisms that can degrade such a compound can be propagated and subsequently isolated. ACKNOWLEDGMENTS This study was supported in part by Grants-in-Aid for Scientific Research A (no. 16268016) and B (no. 17310042) from the Japan Society for the Promotion of Science (JSPS), and the 21st century COE Programs from the Ministry of Education, Culture Sports, Science and Technology (MEXT).
REFERENCES 1. Amman, R. I., Ludwig, W., and Schleifer, K. H.: Phylogenetic identification and in situ detection of individual microbial cells whithout cultivation. Microbiol. Rev., 59, 143–169 (1995). 2. Ohkuma, M.: Termite symbiotic systems: efficient bio-recycling of lignocellulose. Appl. Microbiol. Biotechnol., 61, 1– 9 (2003). 3. Hirsch, A. M., Lum, M. R., and Downie, J. A.: What makes the rhizobia-legume symbiosis so special? Plant Physiol., 127, 1484–1492 (2001). 4. Ahmadjian, V. and Jacobs, J. B.: Relationship between fungus and alga in the lichen Cladonia cristatella Tuck. Nature, 289, 169–172 (1981). 5. Tanaka, H., Aoyagi, H., Shiina, S., Doudou, Y., Yoshimura, T., Nakamura, R., and Uchiyama, H.: Influence of the artificial diet components on the structure and function of the symbiotic microorganisms community in hindgut of Coptotermes formosanus Shiraki. Appl. Microbiol. Biotechnol., 71, 907–917 (2006). 6. Stevenson, B. S., Eichorst, S. A., Wertz, J. T., Schnidt, T. M., and Breznak, J. A.: New strategies for cultivation and detection of previously uncultured microbes. Appl. Environ.
367
Microbiol., 70, 4787–4755 (2004). 7. Trager, W.: The cultivation of a cellulose-digesting flagellate, Trichomonas termopsidis, and certain other termite protozoa. Biol. Bull., 66, 182–190 (1934). 8. Watanabe, K., Teramoto, M., Futamata, H., and Harayama, S.: Molecular detection, isolation, and physiological characterization of functionally dominant phenol-degrading bacteria in activated sludge. Appl. Environ. Microbiol., 64, 4396–4402 (1998). 9. Muyzer, G., de Waal, E. C., and Uitterlinden, A. G.: Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis pf polymerase chain reactionamplified genes encoding for 16S rRNA. Appl. Environ. Microbiol., 59, 695–700 (1993). 10. Ogino, A., Koshikawa, H., Nakahara, T., and Uchiyama, H.: Succession of microbial communities during a biostimulation process as evaluated by DGGE and clone library analyses. J. Appl. Microbiol., 91, 625–635 (2001). 11. Sorensen, T.: A method of establishing groups of equal amplitude in plant sociology based on similarity of species content. Biol. Skr., 4, 1–34 (1948). 12. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J.: Basic local alignment search tool. J. Mol. Biol., 215, 403–410 (1990). 13. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D. J.: Gapped BLAST and PSI-BLAST: a new generation of protein database each programs. Nucleic Acids Res., 25, 3389–3402 (1997). 14. Thompson, J. D., Higgins, D. G., and Gibson, T. J.: CLUSTAL W: improving the sensitivity of progressive multiple sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22, 4673–4680 (1994). 15. Yap, L. F., Lee, Y. K., and Poh, C. L.: Mechanism for tolerance in phenol-degrading Comamonas testosteroni strain. Appl. Microbiol. Biotechnol., 51, 833–840 (1999). 16. Brune, A. and Friedrich, M.: Microecology of the termite guts: structure and function on a microscale. Curr. Opin. Microbiol., 3, 263–269 (2000). 17. Brune, A., Emerson, D., and Breznak, J. A.: The termite gut microflora as an oxygen sink; microelectrode determination of oxygen and pH gradients in guts of lower and higher termites. Appl. Environ. Microbiol., 61, 2681–2687 (1995). 18. Feist, C. F. and Hegeman, G. D.: Phenol and benzoate metabolism by Pseudomonas putida: regulation of trangential pathways. J. Bacterol., 100, 869–877 (1969). 19. Heider, J. and Fuchs, G.: Anaerobic metabolism of aromatic compounds. Eur. J. Biochem., 243, 577–596 (1997).
© 2007, The Society for Biotechnology, Japan
Development of Novel Method for Screening Microorganisms Using Symbiotic Association between Insect (Coptotermes formosanus Shiraki) and Intestinal Microorganisms Arata Hayashi,1 Hideki Aoyagi,1 Tsuyoshi Yoshimura,2 and Hideo Tanaka1* Life Science and Bioengineering Laboratory, Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan1 and Laboratory of Innovative Humanohabitability, Research Institute for Sustainable Humanospere, University of Kyoto, Gokasho, Uji, Kyoto 611-0011, Japan2 Received 30 November 2006/Accepted 18 January 2007
It is becoming increasingly difficult to isolate novel useful microorganisms from the natural environment using conventional screening methods based on pure culture techniques. A novel method for screening microorganisms in symbiotic association with insects was developed. This method involves the following two steps. In the first step, the existence of desired microorganisms that grow well by degrading difficult-to-degrade materials in the gut of insects is detected using the survivability of insects as an indicator. In the second step, the desired microorganisms are selected from the surviving insects. The second step is based on an idea that the guts of insects act as continuous-culture systems whereby microorganisms that cannot degrade diet components are washed out whereas those that can degrade diet components are retained and made to multiply in the gut. Coptotermes formosanus Shiraki was fed with an artificial diet containing phenol as a model of lignin-derived and difficult-to-degrade compound. Each C. formosanus feeding on an artificial diet containing 100 mg/l phenol had different levels of adaptation to the toxicity of phenol. About 20% of C. formosanus fed with an artificial diet containing 100 mg/l phenol died within a few days whereas others survived for more than 10 d. The structure of the intestinal microorganisms of the surviving C. formosanus fed with the 100 mg/l phenol artificial diet gradually changed and was very different from that of the bacterial communities obtained from the enrichment culture of wood-feeding C. formosanus using an artificial medium containing phenol as a sole carbon source. Furthermore, Only three species (as DGGE band) were detected from the gut of wood-feeding C. formosanus, whereas 200 times more phenol-degrading microorganisms were detected in the gut of C. formosanus feeding on a phenol artificial diet. Out of these nine species (as DGGE band) of phenol-degrading microorganisms were isolated. The screening method developed in this study can also be applied to various insects, leading to the isolation of various microorganisms that can degrade difficult-to-degrade compounds. [Key words: screening method, insect, Coptotermes formosanus Shiraki, artificial diet, intestinal microorganisms]
method of isolating new microorganisms. In natural ecosystems, most microorganisms exist by maintaining strong relationships with other organisms. Examples of such relationships include termite-intestinal microorganism symbiosis (2), rhizobia-legume symbiosis (3) and lichen symbiosis (4). It is possible to develop an efficient method of screening the desired microorganisms using the symbiotic association between microorganisms and other organisms. It is known that intestinal microorganisms play important roles in the degradation of diet components of insects. The products of microbial degradation are continuously removed by absorption in the gut wall or excretion through the vent. In this way, the gut of insects can be likened to a continuous-culture system whereby microorganisms with low growth rate (those that cannot utilize diet components) are
In the 19th and 20th century, many useful microorganisms were isolated by using pure culture techniques, and this led to the rapid development of industrial microbiology. However, it is becoming increasingly difficult to use pure culture techniques to isolate new microorganisms. This has significantly limited the progress of microbiology. Various molecular techniques have revealed that more than 99% of microorganisms existing in nature cannot be isolated and maintained in pure cultures (1). To overcome this problem, the development of novel screening methods is very important. We have been focusing on symbiotic associations among microorganisms with the aim of developing a novel * Corresponding author. e-mail: [email protected] phone: +81-(0)29-853-7212 fax: +81-(0)29-853-4605 358
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
washed out, whereas microorganisms with high growth rate (those that can efficiently utilize diet components) are retained and become the predominant species in the gut. We have been studying the symbiotic associations between insects and intestinal microorganisms, which are known to be mediated by their diets (diet–insect–intestinal microbial community), using Coptotermes formosanus Shiraki as a model organism. In such studies, C. formosanus is fed with artificial diets and changes in the intestinal microorganisms in response to changes in the components of the diet are monitored. Our preliminary results showed that when the diet component of C. formosanus (wood-feeding insects) was changed from wood to glucose, their intestinal microorganisms gradually changed from wood-degrading to glucose-degrading microorganisms, and finally, they lost their wood-degrading ability (5). This result suggests that if the components of insect diets are changed, the intestinal microbial community in the insect changes and the microorganisms that are capable of degrading the components of the diet become the predominant species. In other words, microorganisms that are capable of degrading a difficult-todegrade compound can be made to multiply in the intestine of an insect that survives by feeding the insect with an artificial diet containing the difficult-to-degrade compound as the sole carbon source. In this study, we developed a novel screening method that can be used to isolate the desired microorganisms (diet component-degrading microorganisms) by controlling the components of the insect diet. C. formosanus was fed with artificial diets containing phenol as a model of the compounds that are difficult to degrade. During the feeding, about 20% of the termites died but some survived. Changes in the microbial community in the gut of the surviving C. formosanus were analyzed by microscopic observation and using the PCR-denaturing gradient gel electrophoresis (PCRDGGE) method. Subsequently, screening of phenol-degrading microbes was performed using both agar medium and membrane filter methods. Furthermore, the number of colonies formed on membrane filters was counted and species diversity was analyzed using the Plate Wash (PW) PCR (6)DGGE method. From these experiments, it was demonstrated that many species of phenol-degrading microorganisms could be isolated using this screening method. This method can be used for the efficient screening of various difficult to degrade compound-degrading microorganisms. MATERIALS AND METHODS Termites Worker caste termites were collected from a laboratory colony of C. formosanus maintained in the Laboratory of Innovative Humano-habitability, Research Institute for Sustainable Humanosphere, Kyoto University, Japan. They were maintained on wood pieces from the Japanese red pine (Pinus densiflora Sieb) at 30 ± 2°C. Preparation of artificial diets and feeding experiments The preparation of artificial diets was performed according to the method described by Tanaka et al. (5). Agarose (15 g/l) (Takara Shuzo, Otsu) was dissolved in distilled water and autoclaved at 121°C for 15 min. Phenol (50 g/l) was sterilized by passing through φ0.22 µm Durapore filter (Millipore, Bedford, MA, USA) and mixed with the sterilized agarose solution to obtain various phenol
359
concentrations (50, 100, 150 and 200 mg/l). The solutions were solidified in a sterilized glass petri dish. As controls, artificial diets containing 15 g/l agarose only, or 15 g/l agarose and the wood powder of Pinus densiflora (less than 100 µm) were prepared. The solidified agarose gels were cut into blocks aseptically (one block = 25 ×25 ×5 mm). Workers caste termites (25 in number) and one block of the artificial diet were placed inside a sterilized polystyrene case (30 ×30 ×10 mm), and incubated at 30°C. As a starvation condition, C. formosanus individuals were placed inside a sterilized polystyrene case under a humid condition (the case was placed inside a sterilized glass petri dish with sufficient sterilized distilled water). During incubation, the number of viable termites was counted periodically and their survival ratios were calculated. Survival ratio represents the number of surviving C. formosanus as a percentage of the initial number (25 termites). The weight of C. formosanus was also measured. Ten C. formosanus individuals were randomly picked and weighted on a chemical balance, and their average weights were calculated. All experiments were performed in triplicate. Observation and enumeration of protozoa and bacteria Different species of intestinal protozoa of C. formosanus were microscopically observed and counted periodically. Three termite workers were fed with any of the diets and collected randomly. Their hindguts were pulled out from their posterior ends and minced into pieces using fine forceps. The hindgut contents were then suspended in 30 µl of Trager U solution (7). The hindgut pieces were macerated gently in the solution to facilitate the release of protozoa. Different species of protozoa were identified using a light microscope. The number of each species of protozoa was counted using a hemocytometer (Kayagaki Irikakogyo, Tokyo). The number of bacteria was also counted using a bacterial counter (Kayagaki Irikakogyo). Cultivation of intestinal microorganisms on agar plates Intestinal microorganisms were cultivated on agar plates to quantify phenol-degrading microorganisms isolated from the intestine of C. formosanus after feeding with various diets for 10 d. The hindguts of the insects were pulled out from their posterior ends using fine forceps. The hindguts were homogenized using a pellet mixer in 1 ml of MP500 medium containing 500 mg/l phenol (8). The gut homogenate (100 µl) was diluted with 900 µl of MP500 medium. This process was repeated several times and the gut homogenate of a desired concentration was prepared. The diluted gut homogenate was spread onto an MP500 agar plate containing MP500 medium in 1.5% agar gel, and incubated at 30°C for 10 d. After incubation, colonies forming on the MP500 agar plate were counted, and colony forming unit per gut (CFU/gut) was calculated. All experiments were performed in triplicate. Cultivation of intestinal microorganisms using membrane filter method When cultivation was performed on an agar plate, it was observed that some impurities in the agar affected the results. Microbes that cannot degrade phenol but are tolerant to 500 mg/l phenol also formed colonies. Intestinal microorganisms were therefore cultivated using the membrane filter method to determine phenol-degrading microorganisms only. This system contained phenol as the sole carbon source. Gut homogenate was prepared as described above. A Milliflex filter funnel unit, HVWP 0.45 µm (Millipore), was set on a suction pump. The MP500 medium (5 ml) was initially poured into a funnel, and 100 µl of the gut homogenate was placed into the funnel and homogeneously mixed with the MP500 medium. The liquid fraction was filtered out and the intestinal microbes were collected on a membrane filter. The membrane filter was set on a liquid medium cassette, MXLMC0120 (Millipore) filled with MP500 medium, and incubated at 30°C for 10 d. After the incubation, colonies forming on the membrane filter were counted, and CFU/gut was calculated. All experiments were performed in triplicate. The ratio of phenol-degrading microorgan-
360
J. BIOSCI. BIOENG.,
HAYASHI ET AL.
isms to the total number of bacteria was calculated using Eq. 1. Ratio = CFU/gut (membrane filter method)/ Total number of bacteria
(1)
Determination of phenol-degrading ability of microorganisms Microbes that formed colonies were cultivated in liquid cultures to determine their phenol-degrading ability. The colonies were picked with a toothpick and inoculated into 5 ml of MP500 medium in L-shaped test tubes (Taitec, Saitama), and incubated on a shaker (30°C, 30 strokes/min) for 2–7 d. After the incubation, the phenol concentration in the liquid culture was measured using Phenol Test Wako (Wako, Osaka), and the microorganisms that degraded phenol were identified as phenol-degrading microbes. Enrichment culture The guts of wood-feeding C. formosanus (five in number) were pulled out from the posterior ends using a pair of fine forceps and suspended in 5 ml of MP500 medium. The gut suspension was poured into an L-shaped test tube (Taitec) and incubated at 30°C and 30 strokes/min. Cell growth was periodically measured using a Spectronic 20A spectrophotometer (Shimadzu). Phenol concentration in the culture was measured using Phenol Test Wako (Wako). When phenol in the culture broth was completely consumed, microorganisms in the culture were collected by centrifugation (10,000 ×g, 5 min). PCR-DGGE analysis Changes in the components of the intestinal microorganisms of C. formosanus fed with artificial diets were analyzed using the PCR-DGGE method (9). The analysis was performed on C. formosanus fed with an agarose artificial diet, a phenol artificial diet or wood for 10 d. Ten C. formosanus individuals were collected randomly from each set, and their hindguts were pulled out from their posterior ends using a pair of fine forceps. The hindguts were homogenized using a Fast DNA kit (Bio 101, Vista, CA, USA), and total DNA of the intestinal microbial cells were extracted. The V3 region of the bacterial 16S rDNA fragment was amplified using 357F-GC (5′-CGCCCGCCGCGC GCGGGCGGGGCGGGGGCACGGGGGGCCTACGGGAGGC AGCAG-3′) and 518R (5′-GTATTACCGCGGCTGCTGG-3′) as PCR primers. Each amplification reaction mixture (20 µl) consisted of 0.5 U Ex Taq DNA polymerase (Takara Shuzo), 1 µl of total DNA solution, 2 µl of 10 ×PCR buffer, 0.25 µM of each primer, and a mixture containing 100 µM deoxynucleoside triphosphate. A touchdown program (9, 10) was implemented as follows: after the initial denaturation at 94°C for 5 min, 30 cycles of 94°C for 1 min and 72°C for 1 min were performed, and then the reaction mixture was kept at 72°C for 7 min. During the reaction cycle, the annealing temperature was decreased by 1°C from 65 to 56°C every second cycle in the first 20 cycles. The annealing temperature was 55°C in the last 10 cycles. The amplicons were purified with QIA quick (Qiagen, Crawley, UK), and DNA concentration was determined by measuring the OD260 of the purified amplicon solutions. For DGGE, 250 ng of the purified amplicons was used. DGGE was performed using a D-code system (Bio-Rad Laboratories, Hercules, CA, USA). Acrylamide (8%) gels were prepared and electrophoresed with 0.5 ×Tris–acetate–EDTA (TAE) buffer (1 ×TAE buffer consists of 0.04 M Tris base, 0.02 M sodium acetate, and 10 mM EDTA [pH 7.4]). The DGGE gel contained 20–70% gradients of urea and formamide in the direction of electrophoresis as denaturants. The 100% denaturant consisted of 40% (v/v) formamide and 7 M urea. DGGE was performed at a constant voltage of 35 V at 60°C for 24 h. A mixture of amplified 16S rDNA fragments was used as a synthetic marker, consisting of amplified 16S rDNA fragments of Arthrobacter atrocyaneus ATCC 13752, clones obtained from seawater samples, and species related to Balneatrix alpica, Agrobacterium gelatinovorum, Methylomicrobium album, Sargasso Sea-proteobacterium, Sulfitobacter pontiacus, Erythrobacter sp. and the plastid of Emiliania huxleyi. The gel was stained with SYBR Gold Nucleic Acid Gel stain (Molecular Probes,
Eugene, OR, USA) and photographed using a UV transilluminator. The DGGE band profile was analyzed using an image-analyzing system (Image master, Amersham Pharmacia Biotech, Uppsala, Sweden), and the density and migration distance of the bands were calculated. The similarity between each DGGE band pattern in each lane was calculated using Eq. 2 (11). Cs = 2j/(a + b)
(2)
Here, j is the number of common bands between gel lanes A and B, and a and b are the total numbers of bands in gel lanes A and B, respectively. From the observation of the images of the DGGE gel, two bands (bands A and B shown in Fig. 7) that were observed only on the lane of C. formosanus fed with the phenol artificial diet were carefully excised with a razor blade under UV illumination and then placed in a 100 µl of Tris-EDTA buffer. DNA was extracted from the gel pieces by incubation at 4°C for 3 d, and then 0.5 µl of the supernanant was used as the template DNA in a touchdown PCR performed with primers GC-2 (5′-GAAGTCATCATGACCGTTC TGGCACGGGGGGCCTA-3′) (10) and 518R. The amplification mixture and conditions of the touchdown PCR were the same as those used for the amplification of DNA for DGGE as described above. The resulting amplicons were electrophoresed again on a DGGE gel to verify the position of the original band. Subsequently, the amplicons were purified using QIA quick (Qiagen). They were then cloned into the Pgem-T Easy Vector System (Promega, Madison, WI USA). Ligation products were transferred into competent cells of Escherichia coli DH5-α. White colonies were randomly picked and screened directly for inserts by colony PCR using primers for the vector (primers T7 and SP6). Plasmid DNA was prepared from the clones using a QIA prep spin miniprep kit (Qiagen). Plasmid DNA was then sequenced according to the direction of insertion by using CEQ 2000 XL (Beckman, Fullerton, CA, USA). All the sequences were analyzed using Seqman (DNAstar, Madison, WI, USA). Subsequently, the sequences were compared with similar sequences of reference organisms by performing a BLAST search (12, 13). Sequence data were aligned with the CLUSTAL W package (14). Plate Wash (PW)-PCR-DGGE analysis It was demonstrated that phenol-degrading microorganisms were increased number in the gut of C. formosanus fed with the phenol artificial diet using the membrane filter method. However, their specific names and diversity were not clarified. Therefore, an efficient analysis of all phenol-degrading microorganisms grown on a membrane filter was performed. All bacterial colonies, which are formed on the membrane filters by cultivating the intestinal microorganisms of C. formosanus fed with wood and the phenol artificial diet were suspended in sterilized distilled water and placed into a 1.5-ml microtube. The suspensions were collected by centrifugation (12,000 ×g, 10 min). The total DNAs of all bacterial colonies were extracted using the Fast DNA kit, and analyzed using the PCR-DGGE method as described above. Nucleotide sequence accession number The nucleotide sequences of partial 16S rRNA genes have been deposited in the DDBJ nucleotide sequences databank under the following accession numbers: band A-1, AB277362; band A-2, AB277363; band A-3, AB277364; band B, AB277365; Wf-1, AB277366; Wf-2, AB277367; Wf-3, AB2773628; Pf-1, AB277369; Pf-2, AB277370; Pf-3, AB277371; Pf-4, AB277372; Pf-5, AB277363; and Pf-6, AB277364.
RESULTS AND DISCUSSION Effects of artificial diets on survival ratio of C. formosanus Figure 1 shows the effects of artificial diet compo-
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
nents on the survival ratio of C. formosanus. The survival ratios of C. formosanus that were fed with wood powder, agarose artificial diet, or starved, were 95%, 80% or 57.3%, respectively, after 20 d. On the other hand, the survival ratios of C. formosanus that were fed with artificial diets containing 50 and 100 mg/l phenol were 38.7% and 17.3%, respectively. All the C. formosanus fed with artificial diets containing 150 and 200 mg/l phenol died within 15 d. Although the termites were observed eating the wood powder, agarose, and artificial diets containing 50 and 100 mg/l phenol, the termites did not eat the artificial diets containing 150 and 200 mg/l of phenol. This implies that only a very small amount of the artificial diet containing 150 and 200 mg/l phenol and thus phenol, were taken into their gut. These results show that the inclusion of phenol in the artificial diet of C. formosanus reduces their survival ratio because phenol is a toxic substance that inhibits the function of the cytoplasm (15). The amount of decrease in the survival ratio of C. formosanus fed with the phenol artificial diet was higher than those in C. fomosanus fed with agarose and C. formosanus that were starved. This result shows that the main reason for the rapid decrease in the number of C. formosanus fed with the phenol artificial diet is not starvation but phenol toxicity. It is interesting to note that although some of the C. formosanus died within a few days, others survived for a long period of time. Thus, it appears that C. formosanus individuals differ in their susceptibility to phenol toxicity although they were from the same parents and lived in the same colony. The survival levels of C. formosanus fed with artificial diets containing phenol may be considered as gradual steps that involve the following stages: (i) their tolerance to toxicity of phenol, (ii) detoxification of phenol due to the presence of some phenol-degrading microorganisms in the gut of C. formosanus feeding on wood, and (iii) increase in their phenol-degrading ability due to the multiplication of phenol-degrading microorganisms in the gut of C. formosanus fed with the phenol artificial diet. It is likely that only C. formosanus individuals that
FIG. 1. Survival ratio of C. formosanus fed with various artificial diets. Symbols: closed circles, wood powder; open circles, agarose; closed squares, 50 mg/l phenol; open squares, 100 mg/l phenol; closed triangles, 150 mg/l phenol; open triangles, 200 mg/l phenol artificial diet; closed diamonds, starvation. Values are means ± SD from three independent experiments.
361
have phenol tolerance can survive initially. Subsequently, phenol-degrading microorganisms multiply in their gut, prolonging their survival. We propose that the survival period of C. formosanus can be used as an indicator to distinguish individuals harboring phenol-degrading microorganisms from the rest. Thus, the screening method developed in this study is an efficient and novel method by which insects that survive on artificial diets containing phenol successfully multiply desired microorganisms in their intestine. These desired microorganisms are then isolated from the surviving insects. Effects of artificial diets on weight of C. formosanus Figure 2 shows the changes in the weight of C. formosanus during the feeding experiments. The initial mean weight of C. formosanus fed with wood (the feeding time on artificial diets was 0 d) was 3.94 mg. After 20 d, the weights of C. formosanus feeding on wood powder, agarose and the artificial diet containing 100 mg/l phenol decreased to 3.42, 2.41 and 2.22 mg, respectively. These results show that the survival of C. formosanus is inhibited by phenol, and that in comparison with wood powder, both agarose and phenol artificial diets decreased the weight of the termites. Effects of artificial diets on intestinal protozoan community of C. formosanus The changes in the total number of intestinal protozoa of C. formosanus fed with wood, agarose and artificial diets containing 100 mg/l phenol are shown in Fig. 3. There was a rapid decrease in the number of intestinal protozoa of C. formosanus fed with both agarose and phenol artificial diets. All the Pseudotrychonympa grassi (large protozoa) and Holomastigotoides hartmanni (middle protozoa) disappeared after 5 d and all the Spyrotrychonympha leidyi (small protozoa) disappeared after 10 d of feeding on these artificial diets. The disappearance of these protozoa was consistent with the decrease in the weight of the termites, as shown in Fig. 2. It has been reported that intestinal protozoa make up approximately 1/3 of the total weight of lower termites (16). In this study, the weights of C. formosanus individuals fed with agarose and phenol artificial diets were reduced to more than 1/3 of their weights. Therefore, aside from the effect of the physical weight of
FIG. 2. Changes in body weight of C. formosanus fed with various artificial diets with time. Symbols: closed circles, wood powder; open circles, agarose; open squares, phenol (100 mg/l) artificial diet. Values are the means ± SD from three independent experiments.
362
HAYASHI ET AL.
FIG. 3. Changes in total number of protozoa in hindgut of C. formosanus fed with various artificial diets with time. Symbols: closed circles, wood powder; open circles, agarose; open squares, phenol (100 mg/l) artificial diet.
FIG. 4. Changes in total number of bacteria in hindgut of C. formosanus fed with artificial diets. Symbols: closed circles, wood powder; open circles, agarose; open squares, phenol (100 mg/l) artificial diet.
the protozoa, their disappearance might have other physiological effects that result in the further decrease in the weight of C. formosanus. Effects of artificial diets on total number of intestinal bacteria in C. formosanus The changes in the total number of intestinal bacteria of C. formosanus fed with wood powder, agarose and artificial diets containing 100 mg/l phenol are shown in Fig. 4. There was a rapid decrease in the number of intestinal bacteria of C. formosanus fed with both agarose and phenol artificial diets. The total number of bacteria was initially 4.0 ×107 cells/gut (wood feeding) but decreased to less than 1/10 (2.8 ×106 cells/gut) when they were fed with a phenol artificial diet. Detection and determination of phenol-degrading microorganisms from intestinal microorganisms of C. formosanus using MP500 agar plate Figure 5 shows the number of CFU/gut obtained by plating the intestinal microorganisms of C. formosanus on an MP500 agar plate after feeding with wood, agarose and 100 mg/l phenol artificial diets for 10 d. The numbers of the CFU/gut of C. formosanus fed with wood, agarose and 100 mg/l phenol were 3.4 × 104, 5.4 ×104 and 1.9 ×105, respectively. Subsequently, 85 microbial colonies from the intestinal microorganisms of C. formosanus fed with the 100 mg/l phenol artificial diet
J. BIOSCI. BIOENG.,
FIG. 5. Number of colony forming units obtained per gut of C. formosanus fed with various artificial diets. The colonies were grown on MP500 agar plate. Values are means ± SD from three independent experiments.
were randomly picked and their phenol-degrading abilities were investigated. The results show that 83 out of the 85 colonies cannot degrade phenol. Thus, although a very high CFU/gut was obtained from the guts of C. formosanus individuals fed with the 100 mg/l phenol artificial diet, most of these colonies (97.6%) showed no phenol degrading ability. Thus, the MP500 agar plate was not appropriate for the detection and quantification of phenol-degrading microorganisms because nonphenol-degrading microorganisms also formed colonies on the MP500 agar plates. With the MP500 agar plate, all the microorganisms that can be cultured are tolerant to 500 mg/l of phenol, and can metabolize phenol, agar or agar-derived oligosaccharides to form colonies. Thus, the MP500 agar plate method cannot be used to isolate phenol-degrading microorganisms only. Detection and quantification of phenol-degrading microorganisms from intestinal microorganisms of C. formosanus using membrane filter method With the conventionally used MP500 agar plate method, microorganisms that did not degrade phenol but consumed other medium components such as agar and agar-derived oligosaccharides also formed colonies. Thus, to detect and quantify phenol-degrading microorganisms only, the membrane filter method using only phenol as the carbon source was used. In this method, agar was not used, and the microorganisms were collected on the membrane filter while the liquid medium was supplied through the membrane. In this way, only microorganisms that degrade and metabolize phenol contained in the liquid medium can grow and form colonies. Figure 6 shows the number of CFU/gut formed on a membrane filter. The intestinal microorganisms of C. formosanus fed with wood, agarose and 100 mg/l phenol artificial diets for 10 d were collected on a membrane filter and saturated with MP500 medium. Few colonies (200 CFU/gut) were detected from the intestinal microorganisms of C. formosanus fed with the wood and agarose artificial diets. On the other hand, colonies at 2.8 ×103 CFU/gut were detected from the intestinal microorganisms of C. formosanus fed with the 100 mg/l phenol artificial diet. Fifty colonies were randomly picked from the colonies and their phenol-degrading ability was investigated. The results showed that all the 50 colonies degraded phenol. Furthermore, the membrane
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
363
FIG. 6. Number of colony forming units obtained per gut of C. formosanus fed with various artificial diets. The colonies were grown on a PVDF membrane filter supplemented with MP500 liquid medium. Values are means ± SD from three independent experiments.
filter method was used to compare the number of CFU/gut of the microorganisms from the gut of C. formosanus fed with the 50 mg/l phenol artificial diet with that of the microorganisms from the C. formosanus fed with the 100 mg/l phenol artificial diet. The number of CFU/gut of C. formosanus fed with the 100 mg/l phenol artificial diet was high (2.8 ×103 CFU/gut), but only few colonies (5.0 ×102 CFU/gut) were detected in the microorganisms from the gut of C. formosanus fed with artificial diets containing 50 mg/l phenol. This implies that the density of phenol-degrading microorganisms in the gut of termites increases with increasing phenol concentration in their diet. Therefore, for the multiplication of phenol-degrading microorganisms in the gut of C. formosanus, more phenol should be added to the artificial diet but the phenol concentration should not be very high as to markedly reduce feeding activity and thus affect survival. Table 1 shows the total number of bacteria and phenoldegrading microorganisms detected using the membrane filter method and the ratio of phenol-degrading microorganisms in gut of C. formosanus fed with wood and phenol artificial diets for 10 d. The ratios of the intestinal phenoldegrading microorganisms in C. formosanus individuals feeding on wood and phenol artificial diets were 5.0 ×10–6 and 1.0 ×10–3, respectively. These two ratios suggested that the number of phenol-degrading microorganisms were increased by about 200-fold by feeding C. formosanus with the phenol artificial diet. These results showed that the membrane filter method is effective in detecting phenol-degrading microorganisms only, and that the number of phenol-degrading microorganisms is increased in the gut of C. formosanus fed with the artificial diet containing 100 mg/l phenol.
FIG. 7. Image of denaturing gradient gel electrophoresis (DGGE) of 16S rDNA of intestinal bacteria of C. formosanus fed with various artificial diets for 10 d, and bacteria obtained by enrichment culture of intestinal microorganisms of C. formosanus. Lanes 1 and 6, Synthetic markers consisting of amplified 16S rDNA fragments; lane 2, wood; lane 3, agarose; lane 4, phenol; lane 5, enrichment culture with MP500. Bands A and B, Two excised bands for sequence.
PCR-DGGE analysis By using the membrane filter method, it was demonstrated that the number of cultivatable phenol-degrading microorganisms increased in the gut of C. formosanus fed with the phenol artificial diet. However, many microorganisms that cannot be cultured may also be present in the gut of such termites. PCR-DGGE, which is generally used for the analysis of the bacterial community structure, was therefore used to analyze all the bacteria in the gut of the termites. Furthermore, the intestinal microorganisms of C. formosanus fed with the phenol artificial diet was compared with those obtained by the conventional enrichment culture. Figure 7 shows the band patterns of the intestinal bacterial community of C. formosanus fed with the three types of diets, and those of the intestinal microorganisms of wood-feeding C. formosanus obtained by the enrichment culture. Figure 8 shows the dendrogram based on the differences and similarity of band patterns among the bacterial community using Sorensen’s equation. The similarity between the bacterial community in the C. formosanus fed with wood and that fed with the phenol artificial diet was
TABLE 1. Comparison of total bacterial number, number of phenol degrading bacteria and ratio of phenol degrading-bacteria in intestine of C. formosanus fed with wood and phenol artificial diets Total number of bacteria (cells/gut) (A) Wood 4.0 ×107 2.8 ×106 Phenol artificial dietb a Obtained using membrane filter method. b Feeding time: 10 d. Diet
Number of phenol-degrading bacteria (CFU/gut)a (B) 2.0 ×102 2.8 ×103
Ratio of phenol degrading bacteria (B/A) 5.0 ×10–6 (C) 1.0 ×10–3 (D)
D/C – 200
364
J. BIOSCI. BIOENG.,
HAYASHI ET AL.
FIG. 8. Cluster analysis (UPGAMA method) of DGGE band pattern obtained from termites fed with various artificial diets (scale represents the percentage of divergence).
35%, whereas the similarity between the bacterial community in the C. formosanus fed with phenol and that fed with agarose artificial diets was 58%. The similarity between the microorganisms of C. formosanus fed with the phenol artificial diet and those obtained by the enrichment culture was only 12%. These results showed that the structure of the bacterial community of C. formosanus varied greatly depending on the type of artificial diet. However, the intestinal bacterial community of C. formosanus fed with the phenol artificial diet was comparatively similar to that of C. formosanus fed with the agarose artificial diet. Feeding C. formosanus with both diets resulted in the disappearance of intestinal protozoa and a decrease in the total number of bacteria (less than 1/10) after 10 d. It is well known that in the gut of C. formosanus, there is a strict symbiotic association between protozoa and many bacteria (see Ref. 2 and references therein). One important function of intestinal protozoa is phagocytosis (16). Thus, the disappearance of protozoa in the gut of C. formosanus fed with an artificial diet containing phenol affected the intestinal bacterial community. In other words, the disappearance of the intestinal protozoa may be one important factor that led to increase in the number of phenol-degrading microorganism in the gut of C. formosanus. Moreover, these results showed that the bacterial community obtained from the gut of C. formosanus fed with the phenol artificial diet was different from those obtained by the enrichment culture using MP500 medium. In this study, the enrichment culture was performed in batches, using an L-shaped test tube under aerobic condition (under anaerobic condition, there were no growth of microorganisms and phenol consumption). Consequently, the culture conditions such as phenol concentration and pH of the medium varied during the cultivation. On the other hand, in the gut of C. formosanus, nutrients were supplied continuously to the intestinal microorganisms (by the feeding activity of C. formosanus), whereas the products of microbial metabolism such as organic acid were continuously removed by absorption into the gut wall or excretion through the vent. Thus, the gut of C. formosanus can be likened to a continuous-
culture system. Microorganisms with low growth rate (those that cannot utilize the diet components) are washed out, whereas microorganisms with high growth rate (those that can efficiently utilize the diet components) are retained and become the predominant species in the gut. The number of bands obtained from the intestinal bacterial community of C. formosanus fed with the phenol artificial diet was larger than that obtained by the enrichment culture (Fig. 7). This result shows that the species of microorganisms from the gut of C. formosanus fed with the phenol artificial diet are more varied than the species of microorganisms obtained by the conventional enrichment culture. It has been reported that there is an oxygen concentration gradient in the gut of termites (17). Thus, this variation in intestinal environment enabled many microorganisms to adapt to each environment (oxygen concentration) and grow in the gut of C. formosanus. The DNA sequences of bands A and B shown in Fig. 7 were analyzed. Bands A and B were the only bands obtained from the intestinal microorganisms of the termites fed with the phenol artificial diet. The alignment of DNA sequences is shown in Fig. 9. Table 2 shows the strains and accession number of the bacteria that had the highest homology with each DNA sequence in bands A and B. One of the sequences in band A had a 98% homology with Pseudomonas nitroreducens (EF107515). This bacterial strain has been reported as 1,2,4-trichlorobenzene degrader. Although bacteria that degrade aromatic compounds such as phenol and 1,2,4-trichlorobenzene hardly exist in the gut of wood-feeding C. formosanus, they increased in number in such a high concentration to be detected as DNA bands when C. formosanus was fed with the phenol artificial diet. Other DNA sequences contained in bands A and B had high homologies with uncultivatable Crostridium and Bacteroides, and their functions are not yet known. Phenol degradation (particularly complete degradation to CO2 and H2O) occurs under aerobic condition, but the above results show that strictly anaerobic bacteria also exist in the intestinal microorganisms of termites fed with phenol artificial diet. Their functions are unknown but they could degrade phenol. Phenol degradation has been reported to occur mainly under aerobic condition (18), but anaerobic degradation has also been reported (19). This suggests that anaerobic degradation of phenol may occur in the gut of C. formosanus by strictly anaerobic bacteria such as Clostridium and Bacteroides. PWPCR-DGGE analysis Figure 10 shows the band patterns of all phenol-degrading bacteria obtained by the membrane filter method from the intestinal microorganisms of C. fomosanus fed with wood and 100 mg/l phenol artificial diet. The alignment of DNA sequences is shown in Fig. 11. Table 3 shows the numbers of DGGE band detected on each lane and the DNA sequences of the DGGE bands. The number of DGGE band detected on each lane corre-
TABLE 2. Sequencing results of DGGE band (Fig. 7) Band A-1 -2 -3 B
Bacterial strain showing highest similarity Uncultured bacteroidaceae bacterium, clone Rs-E83 Uncultured bacterium, clone RsaHw538 Pseudomonas nitroreducens Uncultured anaerobic bacterium, clone B-4C
Accession no. AB088937 AY571428 EF107515 AY953243
Similarity 89% 92% 98% 100%
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
365
FIG. 9. Alignment of partial sequences of 16S rDNA cloned from excised bands from DGGE gel (Fig. 7) and reference bacteria in database (GenBank).
FIG. 10. Image of denaturing gradient gel electrophoresis (DGGE) of 16S rDNA of all bacterial colonies grown on membrane filter from intestinal microorganisms of C. formosanus fed with wood and 100 mg/l phenol. Lane 1, Wood; lane 2, phenol. The arrows show the detected bands and the DNA of the nominated bands were sequenced.
lated with the number of species of the phenol-degrading bacteria. The total number of bacterial species (as DGGE band) detected from C. formosanus fed with wood was three. On the other hand, that from C. formosanus fed with the phenol artificial diet was nine. These results showed that the num-
ber of species of phenol-degrading microorganisms isolated from gut of C. formosanus fed with the phenol artificial diet (9 DGGE bands) was larger than that of the species isolated from the gut of wood-feeding C. formosanus (conventional method) (3 DGGE bands). DGGE Bands Wf-1, Wf-2 and Wf-3 detected in colonies from the intestinal microorganisms of wood-feeding C. formosanus showed high homology with Salmonella paratyphi strain 50973 (homology = 99%), Citrobacter sp. SVUB3 (97%) and Citrobacter sp. SVUB3 (99%), respectively. DGGE bands Pf-1, Pf-2, Pf-3, Pf-4, Pf-5 and Pf-6 detected in colonies from the intestinal microorganisms of C. formosanus fed with the phenol artificial diet showed high homology with uncultured Pseudomonadales bacterium (99%), Pseudomonas sp. pDL01 (100%), uncultured Pseudomonadales bacterium (99%), uncultured Pseudomonadales bacterium (100%), Pseudomonas aeruginosa strain PD100 (99%) and Enterobacteriales bacterium H0407 (100%). From these results, it can be inferred that many species of Pseudomonas, which are known as aromatic compound degraders, were detected in colonies from the intestinal microorganisms of C. formosanus fed with the phenol artificial diet. The main reason for this difference is that few phenol-degrading microorganisms (below the detection limit), which exist in wood-feeding C. formosanus, grow and multiply in the gut (as a continuousculture system) by feeding on phenol artificial diet. Therefore, microorganisms, which are originally present in the gut of insects but in small number, multiply by feeding on a specific diet and thus can be isolated by using this screening method. There was only one PWPCR-DGGE band (in the same position) that is common to wood-feeding and phenol artificial diet-feeding C. formosanus, but the bacterial species were different. It is possible that phenol-degrading bac-
366
HAYASHI ET AL.
J. BIOSCI. BIOENG.,
FIG. 11. Alignment of partial sequences of 16S rDNA cloned from excised bands from DGGE gel (Fig. 10) and reference bacteria in database (GenBank).
TABLE 3. Total numbers of DGGE bands (Fig. 10) and the DNA sequences of bacteria obtained from intestinal microorganisms of C. formosanus fed with wood and 100 mg/l phenol artificial diet Band Strain showing highest similarity Accession no. Similarity Wf-1 Salmonella paratyphi strain 50973 DQ683179 99% Wf-2 Citrobacter sp. SVUB3 AM401577 97% Wf-3 Citrobacter sp. SVUB3 AM401577 99% Pf 9 Pf-1 Uncultured Pseudomonadales bacterium DQ171526 99% Pf-2 Pseudomonas sp. pDL01 AF125317 100% Pf-3 Uncultured Pseudomonadales bacterium DQ171526 99% Pf-4 Uncultured Pseudomonadales bacterium DQ171526 100% Pf-5 Pseudomonas aeruginosa strain PD100 AY825034 99% Pf-6 Enterobacteriales bacterium H0407 DQ822788 98% The membrane filter method was used in this experiment. Wf, Total number bacterial colonies obtained by membrane filter method from woodfeeding C. formosanus; Pf, total number of bacterial colonies obtained by membrane filter method from C. formosanus fed with phenol artificial diet. Wf
Total band no. 3
teria with high growth rate but an initially small number grew and became predominant, whereas phenol-degrading bacteria with low growth rate were washed out even if their
initial number was high. Because all screening experiments carried out in this study were performed under aerobic condition, screening under anaerobic condition may lead to the
VOL. 103, 2007
DEVELOPMENT OF NOVEL METHOD FOR SCREENING MICROBES
isolation of more phenol-degrading microorganisms. From these results, the screening method described in this study (increasing the number of microorganisms that degrade a particular compound in the gut of an insect by feeding the insect with the compound as the sole carbon source) can be used to isolate more diverse microorganisms. Some microorganisms that cannot be isolated by conventional enrichment culture may be isolated using this novel method. This method can also be applied to other insects, and by feeding artificial diets containing any target compound, microorganisms that can degrade such a compound can be propagated and subsequently isolated. ACKNOWLEDGMENTS This study was supported in part by Grants-in-Aid for Scientific Research A (no. 16268016) and B (no. 17310042) from the Japan Society for the Promotion of Science (JSPS), and the 21st century COE Programs from the Ministry of Education, Culture Sports, Science and Technology (MEXT).
REFERENCES 1. Amman, R. I., Ludwig, W., and Schleifer, K. H.: Phylogenetic identification and in situ detection of individual microbial cells whithout cultivation. Microbiol. Rev., 59, 143–169 (1995). 2. Ohkuma, M.: Termite symbiotic systems: efficient bio-recycling of lignocellulose. Appl. Microbiol. Biotechnol., 61, 1– 9 (2003). 3. Hirsch, A. M., Lum, M. R., and Downie, J. A.: What makes the rhizobia-legume symbiosis so special? Plant Physiol., 127, 1484–1492 (2001). 4. Ahmadjian, V. and Jacobs, J. B.: Relationship between fungus and alga in the lichen Cladonia cristatella Tuck. Nature, 289, 169–172 (1981). 5. Tanaka, H., Aoyagi, H., Shiina, S., Doudou, Y., Yoshimura, T., Nakamura, R., and Uchiyama, H.: Influence of the artificial diet components on the structure and function of the symbiotic microorganisms community in hindgut of Coptotermes formosanus Shiraki. Appl. Microbiol. Biotechnol., 71, 907–917 (2006). 6. Stevenson, B. S., Eichorst, S. A., Wertz, J. T., Schnidt, T. M., and Breznak, J. A.: New strategies for cultivation and detection of previously uncultured microbes. Appl. Environ.
367
Microbiol., 70, 4787–4755 (2004). 7. Trager, W.: The cultivation of a cellulose-digesting flagellate, Trichomonas termopsidis, and certain other termite protozoa. Biol. Bull., 66, 182–190 (1934). 8. Watanabe, K., Teramoto, M., Futamata, H., and Harayama, S.: Molecular detection, isolation, and physiological characterization of functionally dominant phenol-degrading bacteria in activated sludge. Appl. Environ. Microbiol., 64, 4396–4402 (1998). 9. Muyzer, G., de Waal, E. C., and Uitterlinden, A. G.: Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis pf polymerase chain reactionamplified genes encoding for 16S rRNA. Appl. Environ. Microbiol., 59, 695–700 (1993). 10. Ogino, A., Koshikawa, H., Nakahara, T., and Uchiyama, H.: Succession of microbial communities during a biostimulation process as evaluated by DGGE and clone library analyses. J. Appl. Microbiol., 91, 625–635 (2001). 11. Sorensen, T.: A method of establishing groups of equal amplitude in plant sociology based on similarity of species content. Biol. Skr., 4, 1–34 (1948). 12. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J.: Basic local alignment search tool. J. Mol. Biol., 215, 403–410 (1990). 13. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D. J.: Gapped BLAST and PSI-BLAST: a new generation of protein database each programs. Nucleic Acids Res., 25, 3389–3402 (1997). 14. Thompson, J. D., Higgins, D. G., and Gibson, T. J.: CLUSTAL W: improving the sensitivity of progressive multiple sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22, 4673–4680 (1994). 15. Yap, L. F., Lee, Y. K., and Poh, C. L.: Mechanism for tolerance in phenol-degrading Comamonas testosteroni strain. Appl. Microbiol. Biotechnol., 51, 833–840 (1999). 16. Brune, A. and Friedrich, M.: Microecology of the termite guts: structure and function on a microscale. Curr. Opin. Microbiol., 3, 263–269 (2000). 17. Brune, A., Emerson, D., and Breznak, J. A.: The termite gut microflora as an oxygen sink; microelectrode determination of oxygen and pH gradients in guts of lower and higher termites. Appl. Environ. Microbiol., 61, 2681–2687 (1995). 18. Feist, C. F. and Hegeman, G. D.: Phenol and benzoate metabolism by Pseudomonas putida: regulation of trangential pathways. J. Bacterol., 100, 869–877 (1969). 19. Heider, J. and Fuchs, G.: Anaerobic metabolism of aromatic compounds. Eur. J. Biochem., 243, 577–596 (1997).