Bioresource Technology 96 (2005) 15921598

Isolation and characterization of enteric bacteria from the hindgut of Formosan termite
Leith Adams, Raj Boopathy
*
Department of Biological Sciences, Nicholls State University, Thibodaux, LA 70310, USA Received 19 July 2004; received in revised form 6 December 2004; accepted 7 December 2004 Available online 26 February 2005

Abstract The Formosan subterranean termite, Coptotermes formosanus Shiraki, is an aggressive, invasive termite species that has caused billions of dollars of damage across the United States for the past 50 years. Termites depend on intestinal microorganisms for cellulose digestion. Symbiotic microorganisms in the termite gut play key physiological functions such as cellulose and hemicellulose digestion, acetogenesis, hydrogenesis, methanogenesis, sulfate reduction, and nitrogen xation. Additionally, intestinal microbes create suitable conditions for symbiotic protozoans through the production of nutrients and the maintenance of the pH and the anaerobic conditions in the gut. Although extensive research has been done on the symbiotic relationship of these termites and the microbes found in its gut, there is little information available on the role of facultative anaerobes in the gut. We isolated four enteric bacteria from the hindgut of Formosan subterranean termite, C. formosanus. All isolates were facultative anaerobes and G. The isolates were identied as Serratia marcescens, Enterobacter aerogens, Enterobacter cloacae, and Citrobacter farmeri by using BIOLOG assay and fatty acid methyl ester analysis (FAME). Each isolate was characterized using sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis and biochemical study. This is the rst report on the presence of facultative microbes in termite gut. Results of this rst study on facultative microbes in the termite gut indicate that the role of facultative organisms in the Formosan termite gut may be to scavenge oxygen and create anaerobic conditions for the anaerobic microorganisms, which are essential for digestion of cellulose consumed by the termite. 2005 Elsevier Ltd. All rights reserved.
Keywords: Termites; Enteric bacteria; SDS-PAGE; Antibiotic prole; FAME analysis

1. Introduction The termite gut contains endosymbiotic protozoa and bacteria. Previous studies have shown the relationship among these endosymbionts on dierent aspects of cellulose digestion, acetate formation, and the formation of methane. Specically, the protozoa are the initial digester of cellulose (Fig. 1), being able to engulf and metabolize cellulose into acetate, hydrogen and carbon dioxide (Nakashima et al., 2002a,b; Watanabe et al., 2002). Bacteria then utilize the H2 and CO2 to form
*

Corresponding author. Tel.: +1 985 448 4716; fax: +1 985 493 2496. E-mail address: biol-rrb@nicholls.edu (R. Boopathy).

additional acetate (Boga and Brune, 2003; Tholen and Brune, 1999). There are also methanogens present that can use the H2 and CO2 to form methane through interspecies hydrogen transfer to remove excess hydrogen, which can be toxic to the microorganisms at high concentrations (Leadbetter and Breznak, 1996; Schmitt and Brune, 1999). The termite gut must remain completely anaerobic to accommodate anaerobic microbes (Breznak, 1982; Breznak and Brune, 1994; Graber and Breznak, 2004; Graber et al., 2004). Some studies have shown the presence of strict aerobic bacteria in termite gut (Brune et al., 1995); however, the identication of facultative anaerobic bacteria in the Formosan termite gut has

0960-8524/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2004.12.020

L. Adams, R. Boopathy / Bioresource Technology 96 (2005) 15921598

1593

Fig. 1. Microbial ecology of the Formosan termite gut.

not been described in published work. Oxygen can permeate into the gut through the body wall (Brune et al., 1995), and this could be lethal to the anaerobic endosymbionts if not removed. The role of facultative anaerobes may be to scavenge the oxygen that can permeate to the gut, eectively keeping the gut anaerobic. Several studies demonstrate the pathogenic eects of bacteria on immuno-compromised termites. Serratia marcescens has been shown to cause mortality in termites, which have been treated with immuno-suppressing compounds (Connick et al., 2001; Osbrink et al., 2001). The objectives of this study were to isolate, identify, and characterize enteric bacteria specically, facultative anaerobic bacteria in the gut of Formosan subterranean termites. These facultative microbes may have a role to play in the microbial ecology of termite gut by removing oxygen and maintaining anaerobiosis for the strict anaerobic symbionts.

bically with hydrogen/carbon dioxide (80:20) headspaces (Balch and Wolfe, 1976). Cultures were transferred to fresh media approximately every two weeks, and, after three enrichment transfers, pure cultures were isolated. Cultures were streaked for isolation under aerobic and anaerobic conditions. Isolates that grew aerobically were placed in test tubes containing 10-mL of tryptic soy broth (TSB) in test tubes. Isolates that grew anaerobically were placed in 20-mL TSB in anaerobic culture bottles, under anaerobic conditions described by Balch and Wolfe (1976). Pure cultures were maintained in tryptic soy agar (TSA) slants at 4 C until needed for identication and characterization experiments. 2.2. Identication BIOLOG (Hayward, CA) and fatty acid methyl ester (FAME) analysis were used to identify the isolates. BIOLOG system uses a 96 well microplate with a different carbon source in each well (Biolog, 2002). The microplates are specic for Gram positive and Gram negative isolates. Lawns of each isolate were streaked on TSA plates and incubated at 28 C for 24 h. Cultures were then suspended in phosphate buer solution (PBS), and the absorbance was read using a BIOLOG spectrophotometer at the wavelength of 600 nm. Gram-negative bacteria were suspended to a transmittance of 52%, and Gram-positive bacteria were suspended to a transmittance of 2028%. The suspension was then poured into a sterile container, and 10 lL was dispensed into the wells of the BIOLOG plate using an 8-tip micropipette. The plates were then incubated at 28 C for 24 h. Individual wells containing media that was purple indicated positive growth, and wells that remained colorless

2. Methods 2.1. Materials The Formosan subterranean termites, C. formosanus were provided by the United States Department of Agriculture (USDA), Formosan Subterranean Termite Research Unit, Southern Regional Research Center (SRRC), New Orleans, LA. Each termite was externally sterilized in 100% ethanol for about 1 min and then allowed to air dry for 1 min. The gut was then opened using ame sterilized ne-tip forceps and placed in an individual anaerobic culture bottle with 20-ml of tryptic soy broth (TSB) Medium. Cultures were grown anaero-

1594

L. Adams, R. Boopathy / Bioresource Technology 96 (2005) 15921598

indicated negative growth. The pattern of positive and negative results was entered into the Microlog computer program (BIOLOG), and a database search was conducted for the closest match. FAME analysis (Miller and Berger, 1985) was conducted using fresh isolate growth plated on tryptic soy agar (TSA) plates. Approximately 45 mg of each isolate was placed in the bottom of a sterile, screw top vial, and the methylation process was applied. The methylation process includes the addition of the following reagents: reagent 1 was prepared by dissolving 45 g of sodium hydroxide in a solution of 150 mL of methanol and 150 mL of deionized water, reagent 2 consisted of a mixture of 325 mL of 6.00 N hydrochloric acid and 275 mL of methanol, reagent 3 contained 200 mL of both hexane and methyl tert-butyl ether, reagent 4 was prepared by dissolving 10.8 g of sodium hydroxide in 900 mL of deionized water. One mL of reagent 1 was added to each vial. Each mixture was then vortexed for 510 s, boiled in a 100 C water bath for 5 min, vortexed for another 510 s, boiled for another 25 min at 100 C, removed from the bath, and then allowed to cool. Next, 2.0 mL of reagent 2 were added to the mixture. The mixture was then vortexed for 510 s and placed in an 80 C water bath for 10 min. The tubes were cooled rapidly by placing them in a cool water bath. Next, 1.25 mL of reagent 3 was added to each vial. The mixtures were then gently inverted for 10 min, and then the bottom phase was removed using a sterile transfer pipette. The remaining top phase was washed by adding 3.0 mL of reagent 4. The mixture was gently inverted for 5 min. Then two-thirds of the top phase was removed using sterile transfer pipettes and placed in sterile gas chromatograph (GC) vials. The vials were then capped and programmed for injection into the gas chromatograph. A gas chromatograph (Varian CP 3800: Varian Inc., Palo Alto, CA) equipped with a ame ionization detector and a DB-5MS capillary column (30 m 0.25 mm with lm thickness of 0.25 lm) and an integrator were used to analyze cells for fatty acid composition. A temperature program initiated at 150 C for 4 min was linearly increased to 250 C at 4 C/min and held for 5 min. Hydrogen was used as a carrier gas. The ow rate of carrier gas, split ratio, and injector temperature were set at 1.2 ml/min, 20:1, and 260 C, respectively. Extracts were then run through the GC individually, and the results were then compared to a database for the closest match that could be made for species identication. 2.3. SDS-PAGE analysis For SDS-PAGE analysis, the Laemmli (Laemmli, 1970) was followed. Cell extract was mixed with 7.5 lL of Laemmli dye (Bio-Rad) supplemented to contain 10% method (7.5 lL) solution (vol/vol)

mercaptoethanol and heated in a water bath at 95 C for 5 min. After cooling to 22 C, the mixture was deposited on 10% polyacrylamide gels and separated with a mini-Protean II (Bio-Rad) set at 20 mA constant current for the two gels for 15 min and then raised to 30 mA for 40 min. The gels were stained with colloidal Coomassie blue (contains 0.1% Coomassie brilliant G-250, 0.2% phosphoric acid, 10% ammonium sulfate, and 20% methanol) overnight and destained with a solution containing 1% acetic acid and 1% glycerol. A low molecular weight marker (Amersham Bioscience Co., Piscataway, NJ) was run concurrently as a size marker. 2.4. Biochemical tests Basic biochemical tests such as oxidase, urease, gelatinase, and nitrate reduction were conducted as per the method described by Benson (2002). Bacterial growth was monitored using various carbon sources as the sole source of carbon under aerobic and anaerobic conditions. For the aerobic experiment, a 100 ml of the following basic mineral salt medium was used in a 250 ml Erlenmeyer ask. The basic mineral salt medium consisted of the following compounds dissolved in deionized water: K2HPO4 (3.5 g/L), KH2PO4 (1.5 g/L), MgSO4 (0.1 g/L), NaCl (0.1 g/L), (NH4)2SO4 (0.25 g/ L). Any one of the following carbon sources served as the sole source of carbon for bacterial growth at a concentration of 0.5 g/L: glucose, benzoate, citrate, lactose, malate, sucrose, xylose, gentisate, histamine, and ketogluterate. The experiment was conducted in duplicate. The experiment was initiated with 1% inoculum (vol/ vol) from a 24 h stock culture grown on tryptic soy broth (TSB). For anaerobic experimentation, the same media mentioned above was used in an anaerobic culture bottle with a headspace of H2:CO2 (80:20) kept at 20 psi (Balch and Wolfe, 1976). The cultures were incubated at 28 C on a shaker kept at 100 rpm. To monitor cell growth, the turbidity of the culture was determined using a Spectronic 20 Spectrophotometer (Bausch and Lomb, Bualo, NY) kept at the wavelength of 600 nm. 2.5. Antibiotic prole Antibiotic susceptibility of each isolate was determined by the disk diusion method (Bauer et al., 1966). Disks of vancomycin (30 lg), erythromycin (15 lg), rifampin (5 lg), ooxacin (5 lg), tetracycline (30 lg), lincomycin (2 lg), penicillin (10 lg), novobicin (30 lg), oxacillin (1 lg), ampicillin (10 lg), ciprooxacin (5 lg), and sulfame thoxazole-trimethoprim (STX) (23.75 lg; 1.25 lg) were used. The sensitivity and resistance of each isolate were determined by the criteria of the National Committee for Clinical Laboratory Standards (1997).

L. Adams, R. Boopathy / Bioresource Technology 96 (2005) 15921598

1595

3. Results and discussion 3.1. Identication and biochemical characterization Four G facultative anaerobic bacteria were isolated from the hindgut of Formosan termite. The isolates were identied as S. marcescens, Enterobacter aerogens, Enterobacter cloacae, and Citrobacter farmeri by using BIOLOG assay and fatty acid methyl ester analysis (FAME). Both methods showed a 100% similarity of the identied isolates to their database. The isolates were G, rod, and motile. All isolates were urease and oxidase negative, positive for nitrate reduction and gas production in triple sugar iron agar (TSIA) (Table 1). S. marcescens was positive for gelatin, methyl red and VogesProskauer and produced H2S in TSIA. C. farmeri was negative for oxidase, gelatinase, and urease. All the dierent biochemical tests showed in Table 1 matched perfectly with the reported biochemical proles of the isolates and these isolates belong to the family Enterobacteriacae (Brenner, 1992; Connick et al., 2001; Frederiksen and Sgaard, 1992; Grimont and Grimont, 1992a,b; Madigan et al., 2002). 3.2. Antibiotic prole

and ampicillin. E. cloacae showed resistance to vancomycin, erythromycin, rifampin, lincomycin, penicillin, novobiocin, ampicillin, and STX. S. marcescens showed resistance to vancomycin, rifampin, tetracycline, penicillin, novobiocin, and ampicillin. C. farmeri showed resistance to vancomycin, erythromycin, rifampin, tetracycline, lincomycin, penicillin, and novobiocin (Table 2). The antibiotic resistant and susceptibility prole matched with the published prole for these organisms (National Committee for Clinical Laboratory Standard, 1997; NCTR, 2003; Nawaz et al., 2003). 3.3. FAME analysis The cellular fatty acid composition of the two Enterobacter species is shown in Fig. 2. Dierent species have dierent cellular membrane compositions. Each species has a unique fatty acid composition, making it a microbial ngerprint. FAME analysis results matched perfectly (100%) with the cellular fatty acid composition of S. marcescens, C. farmeri, E. aerogenes, and E. cloacae in the database. The use of the FAME analysis also helped to show the dierences in the closely related Enterobacter species. 3.4. SDS-PAGE analysis

Susceptibilities to the antibiotics diered among each isolate (Table 2). Enterobacter aerogenes showed resistance to rifampin, lincomycin, penicillin, novobiocin,
Table 1 Selected physical and biochemical characteristics of the isolates Characteristics and tests Physical Gram reaction Motility Shape Biochemical reactions Oxidase Nitrate reduction Gelatin Methyl red VogesProskauer H2S production in TSIA Gas production in TSIA Urease TSIAtriple sugar iron agar. S. marcescens + Rods + + + + + +

The results for the SDS-PAGE gave a good representation of protein banding in each of the G isolates.

E. aerogens + Rods + + +

E. cloacae + Rods + + +

C. farmeri + Rods + + + +

Table 2 The antibiotic prole results for various isolates Isolate E. aerogens E. cloacae S. marcescens C. farmeri Vancomycin + + + Erythromycin + + Rifampin + + + + Olfoxacin Tetracycline + Lincomycin + + + Penicillin + + + + Novobiocin + + + + Ampacillin + + + Ciprooxacin

+ = resistant, = susceptible, = inconclusive.

1596

L. Adams, R. Boopathy / Bioresource Technology 96 (2005) 15921598

Fig. 2. FAME analysis result for Enterobacter spp. The composition of fatty acids is similar for E. aerogens and E. cloacae but the intensity varies for the two species.

E. aerogenes had a distinct major band of about 25 kiloDaltons (kDa). E. cloacae had a distinct major band of about 27 kDa and showed some similarity to the taxonomically related E. aerogenes. A major band was seen at 35 kDa for S. marcescens. A distinctive major band was seen at 48 kDa for C. farmeri (Fig. 3). The banding pattern of these isolates matched with distinct protein reported earlier (National Committee for Clinical Laboratory Standard, 1997; NCTR, 2003; Nawaz et al., 2003). This further veried the identity of the four isolates. 3.5. Growth studies The growth curves for the four isolates grown with various carbon sources under aerobic condition are shown in Fig. 4. S. marcescens eectively grew on glu-

cose, benzoate, citrate, malate and sucrose as a sole carbon source. These growth substrates are reportedly used by S. marcescens (Grimont and Grimont, 1992b; Madigan et al., 2002). As reported earlier (Grimont and Grimont, 1992a; Madigan et al., 2002), E. aerogenes eectively utilized all the carbon sources with the exception of ketogluterate and E. cloacae eectively utilized glucose, citrate, lactose, sucrose, and xylose. C. farmeri eectively utilized all carbon sources except histamine and ketogluterate. Similar growth patterns and substrate preference were reported earlier for these isolates (Brenner, 1992; Frederiksen and Sgaard, 1992). Under anaerobic conditions, all the isolates used various carbon sources similar to aerobic conditions (data not shown) showing the facultative nature of these organisms.

L. Adams, R. Boopathy / Bioresource Technology 96 (2005) 15921598

1597

Fig. 3. SDS-PAGE results showing the dierentiating band for: (1) E. aerogenes, (2) E. cloacae, (3) S. marcescens, and (4) C. farmeri.

Various facultative anaerobic bacteria were isolated and identied in this study, showing the presence of this type of organism in the gut of Formosan termites. By using dierent characterization techniques, the isolates were identied as separate species. All species identied are facultative anaerobes as characterized by others (Brenner, 1992; Connick et al., 2001; Frederiksen and Sgaard, 1992; Grimont and Grimont, 1992a,b; Madigan et al., 2002). BIOLOG method is commonly used and readily available. After multiple positive identications of the organisms, it is concluded that S. marcescens, E. aerogenes, E. cloacae, and C. farmeri were successfully isolated from the Formosan subterranean termite gut in this study. FAME analysis conrmed the positive identication of all the isolates. SDS-PAGE characterization helped to show the differences in protein composition of each organism. The closely related Enterobacter species showed similarities, while S. marcescens and C. farmeri were quite dierent. The specic proteins for these organisms matched perfectly with previously published protein proles (National Committee for Clinical Laboratory Standard, 1997; NCTR, 2003; Nawaz et al., 2003). Biochemical tests showed each isolates ability to metabolize dierent carbon sources under both aerobic and anaerobic conditions. The antibiotic proling gives a comparison of the isolates, showing the dierences of each, which matched published antibiotic proling (National Committee for Clinical Laboratory Standard, 1997; NCTR, 2003; Nawaz et al., 2003). The role of facultative anaerobes in the termite gut appears to be to scavenge oxygen, which has permeated the exoskeleton into the gut (Madigan et al., 2002). The presence of large quantities of oxygen can be deadly to strictly anaerobic organisms, such as the protozoa and spirochete found in the termite gut. The facultative bacteria described in this study are found on the periphery of the termite gut while the strict anaerobes are in the center of the gut. The oxygen that permeates from the

Fig. 4. Growth curves of the isolates grown under aerobic conditions with various carbon sources. The data are average of duplicate cultures. (A) S. marcescens, (B) E. aerogenes, (C) E. cloacae, and (D) C. farmeri.

1598

L. Adams, R. Boopathy / Bioresource Technology 96 (2005) 15921598 Frederiksen, W., Sgaard, P., 1992. The genus Citrobacter. In: Balows, A., Tru per, H., Dworkin, M., Harder, W., Schleifer, K.-H. (Eds.), The Prokaryotes, second ed. Springer-Verlag, New York, pp. 27442753. Graber, J.R., Breznak, J., 2004. Physiology and nutrition of Treponema primitia, an H2/CO2-acetogenic spirochete from termite hindguts. Appl. Environ. Microbiol. 70, 13071314. Graber, J.R., Leadbetter, J.R., Breznak, A., 2004. Description of Treponema azotonutricium and Treponema primitia sp. nov., the rst spirochetes isolated from termite guts. Appl. Environ. Microbiol. 70, 13151320. Grimont, F., Grimont, P., 1992a. The genus Enterobacter. In: Balows, A., Tru per, H., Dworkin, M., Harder, W., Schleifer, K.-H. (Eds.), The Prokaryotes. Springer-Verlag, New York, pp. 27972815. Grimont, F., Grimont, P., 1992b. The genus Serratia. In: Balows, A., Tru per, H., Dworkin, M., Harder, W., Schleifer, K.-H. (Eds.), The Prokaryotes. Springer-Verlag, New York, pp. 28222848. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteiophage T4. Nature 227, 680685. Leadbetter, J., Breznak, J., 1996. Physiological ecology of Methanobrevibacter cuticularis sp. nov. and Methanobrevibacter curvatus sp. nov., isolated from the hindgut of the termite Reticulitermes avipes. Appl. Env. Microbiol. 62, 36203631. Madigan, M.T., Martinko, J.M., Parker, J., 2002. Brock Biology of Microorganisms. Prentice Hall, Upper Saddle River, NJ, pp. 477 485. Miller, L., Berger, T., 1985. Bacterial identication by gas chromatography of whole cell fatty acids. Gas Chromatography Application Note 228-41. Hewlett Packard Company, Palo Alto, CA. Nakashima, K., Watanabe, H., Azuma, J.I., 2002a. Cellulase genes from the parabasalin symbiont Pseudotrichonympha grassii in the hindgut of the wood-feeding termite Coptotermes formosanus. Cell. Mol. Life Sci. 59, 15541560. Nakashima, K., Watanabe, H., Saito, H., Tokuda, G., Azuma, J.I., 2002b. Dual cellulose-digesting systems of the wood-feeding termite, Coptotermes formosanus Shiraki. Insect Biochem. Mol. Biol. 32, 777784. National Center for Toxicological Research Laboratory Report, 2003. NCTR Report on Enteric Bacteria in Poultry. NCTR FDA Laboratory, Jeerson, AR. National Committee for Clinical Laboratory Standards, 1997. Methods for dilution antimicrobial susceptibility tests fro bacteria that grow aerobically. Approved Standards M7-A4. NCCLS, Villanova, PA. Nawaz, M., Khan, S., Khan, A., Nayak, R., Steele, R., Paine, D., Jones, R., 2003. Molecular characterization of uoroquinoloneresistant Camphylobacter spp. isolated from poultry. Poultry Sci. 82, 251258. Osbrink, W.L.A., Williams, K.S., Connick, W.J., Wright, M.S., Lax, A.R., 2001. Virulence of bacteria associated with the Formosan subterranean termite (Isotera: Rhinotemitidae) in New Orleans, LA. Environ. Entomol. 30, 443448. Schmitt, W.D., Brune, A., 1999. Hydrogen proles and localization of methanogenic activities in the highly compartmentalized hindgut of soil-feeding higher termites (Cubitermes spp.). Appl. Environ. Micrbiol. 65, 44904496. Tholen, A., Brune, A., 1999. Localization and in situ activities of homoacetogenic bacteria in highly compartmentalized hindgut of soil-feeding higher termites (Cubitermes spp.). Appl. Environ. Microbiol. 65, 44974505. Watanabe, H., Nakashima, K., Saito, H., Slaytor, M., 2002. New endo-beta-1,4-glucanases from the parabasalian symbionts, Pseudotrichonympha grassii and Holomastigotoides mirabile of Coptotermes termites. Cell. Mol. Life Sci. 59, 19831992.

exoskeleton to the gut are eectively scavenged by the facultative organisms and thus protecting the strict anaerobes in the center of the gut, which are essential for cellulose digestion and termite survival. The role of the facultative anaerobes in the Formosan termite gut is not reported before and, to our knowledge, this is the rst report on the presence of a multitude of facultative anaerobes in the Formosan termite guts. Further research is needed to better understand the ecology of these microbes. Acknowledgements This work was funded by a grant from the USDA through a subcontract from the USDAs Southern Regional Research Center in New Orleans. We would like to thank Dr. Alan Lax and Dr. Ashok Raina of the USDA Southern Regional Research Center (SRRC) in New Orleans, LA for their help and endless supply of termites. We also would like to thank Dr. Mohamed Nawaz and Mr. Donald Paine at the FDA National Center for Toxicological Research (NCTR) in Jeerson, Arkansas, for their help in the FAME analysis. References
Balch, W.E., Wolfe, R.S., 1976. New approach to the cultivation of methanogenic bacteria: 2-mercaptoethane sulfonic acid (HS-CoM) dependent growth of Methanobacterium ruminantium in a pressurized atmosphere. Appl. Environ. Microbiol. 32, 781789. Bauer, A.W., Kirby, W., Sherris, J., Turek, J., 1966. Antibiotic susceptibility testing by a standardized single disk method. Am. J. Clin. Pathol. 45, 493496. Benson, H.J., 2002. Microbiological Applications. McGraw Hill, Boston, MA, pp. 170200. Biolog, 2002. Microlog Users Manual, Release 4.2. Biolog Inc., Hayward, CA. Boga, H., Brune, A., 2003. Hydrogen-dependent oxygen reduction by homoacetogenic bacteria isolated from termite guts. Appl. Environ. Microbiol. 69, 779786. Brenner, D.J., 1992. Introduction to the family enterobacteriaceae. In: Balows, A., Tru per, H., Dworkin, M., Harder, W., Schleifer, K.-H. (Eds.), The Prokaryotes, second ed. Springer-Verlag, New York, pp. 26732695. Breznak, J., 1982. Intestinal microbiota of termites and other xylophagous insects. Annu. Rev. Microbiol. 36, 323343. Breznak, J., Brune, A., 1994. Role of microorganisms in the digestion of lignocellulose by termites. Annu. Rev. Microbiol. 39, 453487. Brune, A., Emerson, D., Breznak, J., 1995. The termite gut microora as an oxygen sink: microelectrode determination of oxygen and pH gradients in guts of lower and higher termites. Appl. Environ. Microbiol. 61, 26782681. Connick, W.J., Osbrink, W.L.A., Wright, M.S., Williams, K.S., Daigle, D.J., Boykin, D.L., 2001. Increased mortality of Coptotermes formosanus (Isoptera: Rhinotermitidae) exposed to eicosanoid biosynthesis inhibitors and Serratia marcescens (Eubacteriales: Enterobacteriaceae). Environ. Entomol. 30, 449455.