Identification of Midgut Bacteria From Fourth Instar Red Imported Fire

Identification of Midgut Bacteria From Fourth Instar Red Imported Fire Ant Larvae, Solenopsis invicta Buren (Hymenoptera: Formicidae)1 John J. Peloquin and Les Greenberg2 University of California at Riverside, Riverside, CA 92521

USA

J. Agric. Urban Entomol. 20(3): 157–164 (July 2003)

ABSTRACT We studied the normal culturable bacteria found in the gut of the fourth instar of the red imported fire ant, Solenopsis invicta Buren. Larvae were dissected under sterile conditions, and the guts were removed and homogenized in phosphate-buffered saline. After growing gram-positive bacteria from this homogenate, three of six isolates were identified with high confidence using both lipid and genetic analysis. Two of these bacterial isolates were identified as Lactococcus garviae and the third as Staphylococcus saprophyticus. The three remaining isolates were identified by lipid analysis as Enterococcus avium. However, these three isolates had a poor genetic match to that bacterium and may therefore be an unknown species. KEY WORDS: Hymenoptera, Formicidae, Solenopsis invicta, fire ant, bacteria, insect gut, enterobacteriaceae

Adult red imported fire ants (RIFAs), Solenopsis invicta Buren (Hymenoptera: Formicidae), filter particles as small as 0.88 ␮ from ingested liquid food, whereas the larvae do not filter small particles (Glancey et al. 1981). This filtering prevents bacteria from entering the adult’s gut with food. Jouvenaux et al. (1996) fed RIFA colonies food contaminated with the bacteria Serratia marcescens, Bacillus thuringiensis, and Bacillus sphaericus, and failed to find the bacteria afterwards in the adult’s gut. However, they did find that the guts of some RIFA queens contained a slow-growing gram-negative bacterium. Nevertheless, fourth instar larvae contained S. marcescens in their midguts after they were fed the contaminated food (Jouvenaux et al. 1996). The importance of the fourth larval instar to a RIFA colony has been demonstrated by Tschinkel (1988, 1995). He showed that stimulation of queen oviposition was strongly associated with the end of the fourth larval instar. He also showed that workers feed gut contents (meconium) from pupating fourth instar larvae to the queens and that this material was responsible for the increase in queen fecundity. Little is known about culturable gut bacteria in ants because the literature concentrates on as-yet-uncultured obligate symbionts (Jungen 1968, Beckham et al. 1982, Schröder et al. 1996, Boursaux-Eude & Gross 2000, Peloquin et al. 2001).

1

Accepted for publication 30 April 2004. For correspondence and reprints.

2

157

158

J. Agric. Urban Entomol. Vol. 20, No. 3 (2003)

Information that does exist is mostly a result of searches for pathogens that attack ants or from studies on ants spreading human pathogens in hospitals and similar situations (Hughes et al. 1989). Caetano & Cruz-Landim (1985) dissected the digestive tracts of workers of two species of South American ants and partially described the presence and distribution of microorganisms therein. Workers of cephalotinid ants had large numbers of nonseptate fungi and bacilli, thought to be mutualistic, in the small intestine. These bacteria exist as a relatively dense flora with a variety of bacterial species including gram-negative and gram-positive coccobacilli and anaerobes resembling Bacterioides and Clostridia (Yurman & Dominguez-Bello 1993). Cephalotinid ants consume bird feces, a behavior well documented in tephritids and other arthropods, and bird feces are believed to be a common source of gut symbionts (Prokopy et al. 1993, Epsky et al. 1997, Lauzon et al. 1998). Given the importance of the fourth larval instar in RIFA, we decided to identify the culturable bacteria that live in their midguts. This information may aid in understanding the role of the larvae in colony nutrition and could lead to genetic manipulation of the bacteria as a control strategy. We also wanted to examine the gut pH of larvae, and the antibiotic sensitivity of the identified bacteria. Materials and Methods Isolation of bacteria from ants. During April 2001, RIFA colonies were collected from Orange and Merced counties in California. A total of 20 fourth instar reproductive larvae were taken from these colonies and placed into sterile vials containing sterile mineral oil and then frozen. This procedure seals the insect gut and prevents escape of internal microorganisms or entry of external microorganisms. Vials were sealed and placed on dry ice before transportation. Subsequently, ants were washed clean of oil and surface sterilized by agitating the ants in 70% ETOH and 0.75% sodium hypochlorite and water for 2 min. They were then placed in a fresh sterile tube and washed three times in sterile water to remove ethanol and bleach. Ants were dissected in a sterile plastic dish and their midguts removed aseptically. Midguts were then homogenized in 100 mL of phosphate-buffered saline (Sambrook et al. 1989) and the homogenate from pooled samples of five larvae were each spread as a dilution series of 1.0, 0.01, 0.001, and 0.0001 mL onto Columbia nutrient agar or blood agar No. 2 plates supplemented with 5% defibrinated sheep blood. Plates were incubated at 30°C or 37°C until colonies were seen. Standard microbiological methods were used to select, culture, and propagate insect-isolated bacteria isolated from the larvae (e.g., see Fredenhagen et al. 1987, Peloquin et al. 2000). Bacterial identifications. Bacterial isolates were first identified by gramstain morphology of isolates purified on Columbia nutrient agar or blood agar plates. We chose six isolates based on colony morphology for further analysis. These were sent to Microcheck, Inc. (Northfield, Vermont), for fatty acid methyl ester (FAME) analysis. For this analysis, computers identify peaks obtained by gas chromatography and compare the fatty acid profiles of the unknown isolates to the profiles of 7000 strains in the databases. Additional bacterial samples of these same six isolates were sent to Accugenix® (Newark, Delaware) for analysis by the MicroSeq® 16S rDNA Bacterial Identification System, which compares 500 base pairs of the sample to known bacteria in a database.

PELOQUIN & GREENBERG: Identification of Midgut Bacteria

159

Antibiotic sensitivity. Antibiotic sensitivity of bacterial isolates was characterized by the Kirby-Bauer Disk Diffusion Technique for Antimicrobial Sensitivity (Madigan et al. 2000). The susceptibility of the bacteria to ampicillin, erythromycin, Chloramphenicol, Streptomycin, Kanamycin, doxycycline, and tetracycline was determined by measurement of zones of growth inhibition and comparison of the size of such zones to a reference chart. Sterile, antibioticcontaining disks were placed aseptically on lawns of cultured insect-isolated symbiotic bacteria inoculated on agar plates. Growth inhibition in the area surrounding the disks is indicative of the antibiotic sensitivity level. Determination of fourth instar gut pH. Guts were removed intact from 20 fourth instar reproductive larvae, placed in a 1.5-mL microcentrifuge tube, and homogenized using a sterile pellet pestle. Exactly 200 ␮L of glass-distilled water was added and the homogenate mixed. Thirty-microliter droplets were placed with washed disposable pipettes on pH indicator sticks (pH range 4.0 to 7.0, E. Merck, Darmstadt, Germany, and pH range 4.5 to 10, Whatman, Inc., Maidstone, England). Homogenate pH was estimated by comparison of paper color with a standard supplied by the manufacturer. Transmission electron microscopy. We brought samples of fire ant larvae to the Center for Advanced Microscopy and Microanalysis at the University of California, Riverside, where the larval midgut was fixed, sectioned, and then viewed on a Philips CM 300 transmission electron microscope (for details on methodology see Klotz et al. 2002). Results Isolation of bacteria from ants. Gram-stained preparations of the bacteria cultured from the guts of fourth instar larvae revealed large populations of grampositive coccoid bacteria (Fig. 1, top). The bacteria were free in the lumen: no evidence of attachment of these bacteria was seen in transmission electron microscopy of sections of larval midgut. There was also evidence of bacterial aggregations (Fig. 1, bottom). Although there may have been gram-negative bacteria also present in the larval ant guts, we did not culture nor observe any in microscopy of the dissected guts. The meconium that was expelled during pupation also had large populations of similar-appearing gram-positive bacteria. Bacterial identifications. Table 1 shows results of FAME and genetic analysis. The similarity index varies from 0.001 to 0.999 and expresses how closely the fatty acid composition of an unknown corresponds to the fatty acids of the strains used to generate the library entry. Values >0.300 are considered to be an excellent match (the value is not a probability ratio nor a percentage of agreement, but uses a covariance matrix, principal component analysis, and pattern recognition; Microcheck, Inc. 2003). The table also shows the closest genetic matches to the sample when aligned in a pairwise manner against the MicroSeq® Database. The percentage difference is the percent nucleotide difference between the sample and the closest matches. High confidence identification was given by both fatty acid and genetic analysis for three of the six bacterial isolates (Table 1). Two of these bacterial isolates were identified as Lactococcus garviae and the third as Staphylococcus saprophyticus. The three remaining isolates were identified by the fatty acid

160


Fig. 1. Black and white digital photomicrograph of midgut and contents dissected from the fourth instar reproductive larva of RIFA, opened, spread on a glass slide and gram-stained. Note “string of beads” morphology in top photo (400X) and the arrangement into aggregations of bacteria in the lower photograph (800X). Morphologically identical bacteria were identified as Enterococcus avium through their lipid profile.

Lactococcus garviae Staphylococcus saprophyticus Enterococcus avium Lactococcus garviae Enterococcus avium Enterococcus avium

1 2 3

4 5

6

0.646*

0.717* 0.619*

0.883* 0.548* 0.343*

Similarity indexa

a * ⳱ excellent match for genus and species; lack of * ⳱ confidence level to genus only. FAME, fatty acid methyl ester analysis of the bacterial cell wall. Base pair comparison is of 16S rDNA chains.

ID by FAME

Sample no.

Table 1. Bacterial identifications (see text for details)

Lactococcus garviae Staphylococcus saprophyticus (a) Enterococcus malodoratus (b) Enterococcus avium Lactococcus garviae (a) Enterococcus malodoratus (b) Enterococcus avium (a) Enterococcus malodoratus (b) Enterococcus avium

Closest genetic matches

0.00* 0.09* 0.46 0.83 0.0* 0.46 0.83 0.46 0.83

500 base pair % diffa

PELOQUIN & GREENBERG: Identification of Midgut Bacteria 161

162


analysis as Enterococcus avium. However, this isolate had a poor genetic match to that bacterium and may therefore be an unknown species. Determining selective antibiotics. Isolates of the gram-positive bacteria that were identified by FAME analysis as Enterococcus avium were found to be susceptible to ampicillin, erythromycin, Chloramphenicol, Streptomycin, and Kanamycin. Some isolates were susceptible to doxycycline and tetracycline, though most were at least somewhat resistant to tetracycline and derivatives. Determination of 4th instar gut pH. The E. Merck paper indicated a pH of 5.7–5.9, whereas the Whatman paper indicated a pH of 6.0. Discussion We have identified and cultured larval RIFA gut bacteria and have measured the midgut pH. At least two of the bacterial species are common and could probably be reintroduced to the ants, if necessary. Future research should look at the role of these gut bacteria in colony nutrient cycling. Furthermore, although these bacteria are not known ant pathogens, Enterococcus and similar gram-positive bacteria can be genetically manipulated (e.g., Josson et al. 1989, Trieu-Cuot et al. 1991). Potential insecticidal proteins that could be expressed by these bacteria include cholesterol oxidase (Cho et al. 1995, Shen et al. 1997), venom peptides and proteins (Diniz et al. 1993, Jiang et al. 1995), other miscellaneous peptides (Orivel et al. 2001), gene proteins for metabolic pathways that produce toxins (Waldron et al. 2001), and miscellaneous proteins that are either deleterious to ants or can act in synergy with toxic proteins (Girard & Jouanin 1999). Information on the gut pH may help in selecting proteins to be expressed by genetic manipulation. Antibiotic resistance profiles from disc diffusion assays for the Enterococcus species that we found show that the isolated bacteria are susceptible to most commonly used antibiotics, including ampicillin. Their susceptibility to a wide range of antibiotics is unexpected as antibiotic-resistant Enterococci are quite common (Jeljaszewicz et al. 2000, Kariuki & Hart 2001). Acknowledgments The authors thank the California Department of Food Agriculture for support of this work through Contract no. 99-0845.

References Cited Beckham, R. D., S. L. Bilimoria & D. P. Bartell. 1982. A survey for microorganisms associated with ants in western Texas. Southwest. Entomol. 7: 225–229. Boursaux-Eude, C. & R. Gross. 2000. New insights into symbiotic associations between ants and bacteria. Res. Microbiol. 151: 513–519. Caetano, F. H. & C. Cruz-Landim. 1985. Presence of microorganisms in the alimentary canal of ants of the tribe Cephalotini (Myrmicinae): Location and relationship with intestinal structures. Naturalia 10: 37–47. Cho, H. J., K. P. Choi, M. Yamashita, H. Morikawa & Y. Murooka. 1995. Introduction and expression of the Streptomyces cholesterol oxidase gene (ChoA), a potent insecticidal protein active against boll weevil larvae, into tobacco cells. Appl. Microbiol. Biotechnol. 44: 133–138.

PELOQUIN & GREENBERG: Identification of Midgut Bacteria

163

Diniz, M. R., M. J. Paine, C. R. Diniz, R. D. Theakston & J. M. Crampton. 1993. Sequence of the cDNA coding for the lethal neurotoxin Tx1 from the Brazilian “armed” spider Phoneutria nigriventer predicts the synthesis and processing of a preprotoxin. J. Biol. Chem. 268: 15340–15342. Epsky, N. D., B. D. Dueben, R. R. Heath, C. R. Lauzon & R. J. Prokopy. 1997. Attraction of Anastrepha suspensa (Diptera: Tephritidae) to volatiles from avian fecal material. Florida Entomol. 80: 270–277. Fredenhagen, A., P. Kenny, H. Kita, H. Komura, Y. Naya, K. Nakanishi, K. Nishimaya, M. Sugiura & S. Tamura. 1987. Role of intracellular symbiotes in planthoppers, pp. 101–103. In R. Greenhalgh and T. R. Roberts [Eds.], Pesticide science and biotechnology: Proceedings of the Sixth International Congress of Pesticide Chemistry, Blackwell Scientific Publications, Oxford, United Kingdom, 604 pp. Girard, C. & L. Jouanin. 1999. Molecular cloning of a gut-specific chitinase cDNA from the beetle Phaedon cochleariae. Insect Biochem. Mol. Biol. 29: 549–556. Glancey, B. M., R. K. Vander Meer, A. Glover, C. S. Lofgren & S. B. Vinson. 1981. Filtration of microparticles from liquids ingested by the red imported fire ant Solenopsis invicta Buren. Insectes Sociaux 28: 395–401. Hughes, D. E., O. O. Kassim, J. Gregory, M. Stupart, L. Austin & R. Duffield. 1989. Spectrum of bacterial pathogens transmitted by pharaoh’s ants. Lab. Anim. Sci. 39: 167–168. Jeljaszewicz, J., G. Mlynarczyk & A. Mlynarczyk. 2000. Antibiotic resistance in Grampositive cocci. Int. J. Antimicrob. Agents 16: 473–478. Jiang, H., Y.-X. Zhu, Y.-P. Wang, Z.-P. Wang & Z.-L. Chen. 1995. Synthesis of the spider insecticidal gene and construction of the plasmid expressing in plant. Acta Bot. Sinica 37: 321–325. Josson, K., T. Scheirlinck, F. Michiels, C. Platteeuw, P. Stanssens, H. Joos, P. Dhaese, M. Zabeau & J. Mahillon. 1989. Characterization of a gram-positive broadhost-range plasmid isolated from Lactobacillus hilgardii. Plasmid 21: 9–20. Jouvenaux, D. P., J. C. Lord & A. H. Undeen. 1996. Restricted ingestion of bacteria by fire ants. J. Invert. Path. 68: 275–277. Jungen, H. 1968. Endosymbionten bei Ameisen. Insectes Sociaux 15: 227–232. Kariuki, S. & C. A. Hart. 2001. Global aspects of antimicrobial-resistant enteric bacteria. Curr. Opinion Infectious Dis. 14: 579–586. Klotz, J. H., C. Amrhein, S. McDaniel, M. K. Rust & D. Reierson. 2002. Assimilation and toxicity of boron in the Argentine ant (Hymenoptera: Formicidae). J. Entomol. Sci. 37: 193–199. Lauzon, C. R., R. E. Sjogren, S. E. Wright & R. J. Prokopy. 1998. Attraction of Rhagoletis pomonella (Diptera: Tephritidae) flies to odor of bacteria: Apparent confinement to specialized members of enterobacteriaceae. Environ. Entomol. 27: 853–857. Madigan, M. T., J. M. Martinko & J. Parker. 2000. Brock biology of microorganisms, 9th ed., Prentice Hall, Upper Saddle River, New Jersey, 991 pp. Microcheck, Inc. 2003. Key to symbols and abbreviations in the Microcheck results table. Microcheck, Inc., Northfield, Vermont, 4 pp. Orivel, J., V. Redeker, J.-P. Le Caer, F. Krier, A.-M. Revol-Junelles, A. Longeon, A. Chaffotte, A. Dejean & J. Rossier. 2001. Ponericins, new antibacterial and insecticidal peptides from the venom of the ant Pachycondyla goeldii. J. Biol. Chem. 276: 17823–17829. Peloquin, J. J., L. Kuzina, C. Lauzon & T. Miller. 2000. Transformation of internal extracellular bacteria isolated from Rhagoletis completa Cresson gut with enhanced green fluorescent protein. Curr. Microbiol. 40: 367–371. Peloquin, J. J., S. G. Miller, S. A. Klotz, R. Stouthammer, L. R. J. Davis & J. H. Klotz. 2001. Bacterial endosymbionts from the genus Camponotus (Hymenoptera: Formicidae). Sociobiology 38: 695–708.

164


Prokopy, R. J., S. S. Cooley, L. Galarza, C. Bergweiler & C. R. Lauzon. 1993. Bird droppings compete with bait sprays for Rhagoletis pomonella (Walsh) flies (Diptera: Tephritidae). Can. Entomol. 125: 413–422. Sambrook, J., T. Maniatis & E. F. Fritsch. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1659 pp. Schröder, D., H. Deppisch, M. Obermayer, G. Krohne, E. Stackebrandt, B. Hölldobler, W. Goebel & R. Gross. 1996. Intracellular endosymbiotic bacteria of Camponotus species (carpenter ants): systematics, evolution and ultrastructural characterization. Mol. Microbiol. 21: 479–489. Shen, Z., D. R. Corbin, J. T. Greenplate, R. J. Grebenok, D. W. Galbraith & J. P. Purcell. 1997. Studies on the mode of action of cholesterol oxidase on insect midgut membranes. Arch. Insect Bioch. Physiol. 34: 429–442. Trieu-Cuot, P., C. Carlier, C. Poyart-Salmeron & P. Courvalin. 1991. Shuttle vectors containing a multiple cloning site and a lacZ.alpha. gene for conjugal transfer of DNA from Escherichia coli to Gram-positive bacteria. Gene 102: 99–104. Tschinkel, W. R. 1988. Social control of egg-laying rate in queens of the fire ant, Solenopsis invicta. Physiol. Entomol. 13: 327–350. Tschinkel, W. R. 1995. Stimulation of fire ant queen fecundity by a highly specific brood stage. Ann. Ent. Soc. Am. 88: 876–882. Waldron, C., P. Matsushima, P. R. Rosteck, M. C. Broughton, J. Turner, K. Madduri, K. P. Crawford, D. J. Merlo & R. H. Baltz. 2001. Cloning and analysis of the spinosad biosynthetic gene cluster of Saccharopolyspora spinosa. Chem. Biol. (London) 8: 487–499. Yurman, D. & M. G. Dominguez-Bello. 1993. Bacteria present in the gut of two neotropical Cephalotini ants, Cephalotes atratus and Zacryptocerus cf. pusillus. Folia Microbiol. 38: 515–518.