Introduction

The insect-bacterial association has co-evolved for more than 250 million years and has resulted in multifaceted interactions between insects and bacteria, ranging from pathogenicity to highly sophisticated mutualistic relationships (Douglas & Beard 1997; Oliver et al. 2005; Wernegreen 2002); may be extracellular or intracellular and play a role in the nutrition, physiology and reproduction of the host insect (Brune 1998; Douglas et al. 2001). On insect gut bacterial associations, the earliest report was published by Petri in 1909 in an insect species, the olive fly, Bactrocera (Dacus) oleae (Rossi).

A range of bacteria belonging to different genera, viz. Acetobacter, Agrobacterium, Arthrobacter, Bacillus, Citrobacter, Enterobacter, Erwinia, Escherichia, Hafnia, Klebsiella, Kluyvera, Lactobacillus, Listeria, Micrococcus, Pantoea, Pectobacterium, Proteus, Providencia, Pseudomonas, Raoultella, Serratia, Staphylococcus, Streptococcus and Xanthomonas, has been isolated and characterized from the gut of tephritid fruit fly species (Behar et al. 2009; Lloyd et al. 1986; Prabhakar et al. 2009a; Wang et al. 2011).

In fruit flies, symbionts act as a natural source of nitrogen, amino acids and vitamins lacking in fruit tissues to their host and are vertically transmitted to the next generation (Drew et al. 1983; Gupta & Anand 2003). Moreover, cultivable gut bacteria have several implications in pest management strategies, e.g. bacteria were found to be involved in the degradation of the toxic substances ingested by the host insect leading to insecticide resistance (Bousch & Matsumara 1967; Prabhakar et al. 2008). Certain components of bacterial odor play a vital role in fruit fly behavior as either feeding or ovipositional stimulants (Lauzon et al. 2000) and are being exploited in pest management in the form of baits or traps (Robacker 2007; Sood et al. 2010; Thaochan & Chinajariyawong 2011).

The pumpkin fly, Bactrocera tau (Walker), infests a wide range of commercially important cucurbit and solanaceous crops and has a wide distribution throughout south Asian countries (India, Sri Lanka, Bangladesh and Bhutan) to the southeast Asian countries (Thailand, Malaysia, Vietnam, Philippines and Indonesia) and the far east Asian region including Taiwan and south China (Prabhakar et al. 2009b, 2012; Thanaphum & Thaenkham 2003; White & Elson-Harris 1992). In the previous study, members of the Enterobacteriaceae (Klebsiella oxytoca and Pantoea agglomerans) were found to be the dominant microbial population in the gut of laboratory reared B. tau (Prabhakar et al. 2009a). Laboratory studies on bacterial odors (K. oxytoca and P. agglomerans) attracting B. tau adults (males and females) suggested their possible applied role in fruit fly management programmes (Sood et al. 2010). However, there is still a dearth of knowledge on the bacterial communities attracting B. tau. The cultivable gut bacteria in particular will be more useful than non-cultivable bacteria when searching for a good fruit fly attractant. Therefore, in the present study, an attempt has been made to characterize cultivable gut bacteria of B. tau, a serious pest of cucurbits in southeast Asia, India and China.

Materials and methods

Isolation of gut bacteria

The bacteria were isolated from 10-day-old adult flies reared from field-collected cucurbit fruits from vegetable fields of Himachal Pradesh (Table 1), as described by Lloyd et al. (1986). Flies were cold anaesthetized for 5 min. and surface sterilized with alcohol (70%) for 30 s. followed by sodium hypochloride (0.25%) for 1 min. and then washed three times with sterilized distilled water (SDW) to remove external contaminations. The surface sterilized flies (n = 5 adult flies from each B. tau isolate collected from the same location and host plant; N = 45 adult flies from nine B. tau isolates) were dissected open in physiological saline to remove fly gut. Gut content was streaked separately on Peptone Yeast Extract Agar (PYEA) and Brain Heart Infusion Agar (BHIA) for bacterial growth at 30 ± 1°C for 48–72 h. A single colony of each isolate was used for establishment of pure culture with respective medium and then maintained on PYEA slants and/or plates at 4–8°C.

Table 1 Isolation of gut bacteria from different populations of Bactrocera tau from Himachal Pradesh (India)
Full size table

Selection of gut bacteria for characterization

Initially, the isolated bacteria were evaluated for their attractancy to fruit fly, B. tau (choice method). Thirty gut bacterial isolates, which were frequently isolated from different fruit fly samples across the locations, were used to study their attractiveness to B. tau. Pure culture (72 h old) of different bacterial isolates was grown on PYE (peptone yeast extract) broth medium. The bacterial isolates were taken in separate petri plates and kept inside the cage (45 cm × 45 cm × 55 cm) with un-inoculated PYE broth as control. Twenty-five pairs of 5-day-old fruit flies were released in the cage and flies visiting each treatment were recorded for 30 min. The experiment was repeated six times for B. tau and data obtained were analyzed statistically by using the technique of analysis of variance for completely randomized design as described by Gomez & Gomez (1984). The treatment means were compared at 5% level of significance by least significance difference test described by Gomez & Gomez (1984). Five gut bacteria having maximum fruit fly attractiveness were selected for further characterization.

Characterization of gut bacteria

Morphological and biochemical characterization

Morphological (shape, Gram’s staining, mobility), cultural (pigment production, growth in broth medium) and biochemical (citrate, methyl red, VP, TSI, catalase, oxidase and carbohydrate fermentation tests) characterization of pure culture was done by standard techniques and isolates were identified using Bergey’s Manual of Determinative Bacteriology (Holt et al. 2000).

Molecular characterization

Extraction of genomic DNA

Total genomic DNA of each isolate was extracted following the method of Prabhakar et al. (2009a). The 48-h-old bacterial culture multiplied on Peptone Yeast Extract Broth (PYEB) was transferred to 1.5 ml microtubes and spun at 10,000 rpm for 12 min. After discarding the supernatant, the microtubes containing bacterial pellets (approx. 50 mg) were immersed in liquid nitrogen container for one min and the pellet was ground to fine powder immediately using a micro pestle. To this, 700 μl of cetyltrimethyl ammonium bromide (CTAB) extraction buffer was added and incubated at 65°C for 1 h in a water bath (York Scientific Industries, Delhi, India).

An equal volume (700 μl) of chloroform: isoamyl alcohol (24:1 v/v) was added and the contents were mixed thoroughly. Tubes were spun at 10,000 rpm for 12 min in a high speed refrigerated centrifuge (Remi Elektrotechnik, India) at 4°C. Aqueous phase was transferred to new tubes and 450 μl pre-chilled isopropanol was added and kept at −20°C for 20–30 min to precipitate the DNA. Tubes were then spun at 10,000 rpm for 12 min and supernatant was decanted. The DNA pellet was washed with 70% ethanol (three times), dried and dissolved in 100 μl of Tris EDTA buffer (10 mM Tris HCl and 1 mM EDTA, pH 8.0). RNAse @ 10 μl ml−1 (MBI Fermentas) was added and the emulsion was incubated for half an hour at 37°C. The amount of DNA was quantified by recording the absorbance at 260 nm wavelength using an UV/VIS spectrophotometer (Bio Rad, SmartSpec 3000). DNA was stored at −20°C for further use.

PCR amplification and sequencing

Polymerase chain reaction (PCR) was performed with eubacterial primers pair forward 5′AGAGTTTGATCATGGCTCAG3′ and reverse 5′ TACCTTGTTACGACTTCACC 3′ targeting the 1,400 bp rrs (16S rRNA) gene (Heddi et al. 1998) The PCR amplification was carried out in 0.2 ml PCR tubes with 25 μl reaction volume containing 10 ng of DNA template, 20 pmol of each primer in 25 mM MgCl2, 10 mM of each deoxyribonucleoside triphosphate (Fermentas), 5 units of Taq polymerase (Life Technologies India, Pvt. Ltd.) and 10X reaction buffer. Amplifications were performed using a thermal cycler (GeneAmp PCR system 9700, Applied Biosystems, Carlsbad, CA, USA) with an initial denaturation step of 5 min at 94°C followed by 35 cycles at 94°C for 45 s, 53°C for 45 s, 72°C for 30 s and a final elongation step at 72°C for 5 min. The product was separated in a 1% (w/v) agarose gel in TAE buffer (40 mM Tris-acetate, 1 mM EDTA). PCR products (1,450 bp) of 16S rRNA gene of five gut bacteria obtained through amplification with specific primers were first purified with ethanol precipitation of amplified DNA, freeze dried (Christ Alpha I-2LD), and then custom sequenced (ABI PRISM 310TM Genetic Analyzer, Applied Biosystems, USA) using the same upstream and downstream primers (Life Technologies India).

Nucleotide sequence analysis

The sequences of different bacterial isolates were blasted using online NCBI Blastn program (http://www.ncbi.nih.gov/blast). For the purpose, 42 sequences of 16S rRNA of different bacteria of high sequence similarity were selected for sequence comparison from GenBank Nucleotide Database, NCBI. The pair-wise genetic distance between five gut bacterial isolates of B. tau and other selected bacterial sequences was determined.

The evolutionary history was inferred using the UPGMA method (Sneath & Sokal 1973). The evolutionary distances were calculated using Maximum Composite Likelihood method with bootstrap test (500 replicates) (Tamura et al. 2004). All positions containing gaps and missing data were eliminated from the dataset (complete deletion option). There were a total of 298 positions in the final dataset. Phylogenetic analysis was conducted in MEGA 4.1 Software programme (Tamura et al. 2007).

Results

Isolation and identification of gut bacterial isolates of fruit fly

Gut bacteria were isolated from nine populations of fruit fly, B. tau on two culture media, viz. PYEA and BHIA (enriched culture media). It was found that bacteria were associated with all the nine populations of B. tau (Table 1). A total of 63 different bacterial colonies were observed on two culture media. Of these, 28 were isolated on PYEA and 35 on BHIA. On the basis of colony morphology (pigmentation, shape and size of colony), 16 and 14 isolates obtained on PYEA and BHIA (which were most common among different fruit fly isolates across the locations), respectively, were evaluated for the attractancy to adult fruit flies under laboratory conditions to select the most promising gut bacteria in terms of their attractancy for further characterization.

The five most promising gut bacterial isolates were selected based on their attractancy to B. tau (Table 2) for further characterization. These five gut bacterial isolates: P1B, P3A, P10A, B4A and B10B (P/B-media used for isolation, i.e., PYEA and Brain Heart Infusion Agar; 1,3,10,4.10: B tau isolate numbers giving information about the location and host; and A,B,C....: different bacterial colonies isolated from the gut of B. tau) were gram negative and rod shaped. All five isolates of gut bacteria were producing sediments when multiplied in broth medium. All the five gut bacterial isolates were non-pigment producing and showed mobility except for P10A, which produced yellow pigment on PYEA plates and was non-mobile. With regard to biochemical characteristics, all the isolates showed negative reaction for citrate, VP, TSI and gas production in glucose medium tests, whereas they were found positive for catalase and oxidase test. All isolates except PIB were negative for methyl red test (Table 3). On the basis of cultural, morphological, biochemical and 16S rRNA gene characteristics, the closest match to isolate P1B, P3A, P10A, B4A and B10B was Delftia acidovorans (91% DNA identity), Pseudomonas putida (97% DNA identity), Flavobacterium sp. (95% DNA identity), Defluvibacter sp. (98% DNA identity) and Ochrobactrum sp. (99% DNA identity), respectively (Table 3).

Table 2 Attractancy of bacterial isolates against fruit fly, Bactrocera tau
Full size table
Table 3 Morphological, biochemical and molecular characteristics of gut bacteria of Bactrocera tau
Full size table

The 16S rRNA gene nucleotide sequences of these isolates were submitted to GenBank nucleotide database under accession numbers HQ446523 to HQ446527.

Phylogenetic studies of gut bacteria

All the sequences of five bacterial isolates were compared with 42 other bacterial sequences available online in GenBank (NCBI) by multiple sequence alignment tools using ClustalW programme. The per cent pair-wise genetic distance of the five promising isolates with other bacterial sequences ranged from 0.00 to 0.70 nucleotide per site.

The dendrogram constructed by phylogenetic analysis presented in Fig. 1 showed that all the bacterial isolates, viz., P1B, P3A, B4A and B10B, clustered with Delftia, Pseudomonas, Defluvibacter and Ochrobactrum, respectively (all Proteobacteria), except P10A (HQ446525), which was clustered with Flavobacterium (a typical Bacteroidetes). Based on their affinity with known sequences in databank, the isolate P1B belongs to class β-Proteobacteria, P3A to class γ-Proteobacteria, and B4A & B10B to class α-Proteobacteria.

Fig. 1
figure 1

Phylogenetic tree based on 16S rRNA gene sequences using Unpaired Group Mean Algorithm (UPGMA) method showing the relationships between five gut bacterial isolates of Bactrocera tau. Numbers above the nodes are bootstrap values (500 replicates). Phylogenetic analyses were conducted in software MEGA 4.1. Arrows indicate gut bacteria of B. tau isolated in present study

Full size image

Discussion

Many workers suggested that nucleic acid sequence approaches, particularly 16S rRNA genes, have proved an important tool to settle the taxonomic position of the microbial community of insects (Brauman et al. 2001; Wang et al. 2011). An immense library of sequence data for 16S rRNA loci and other robust markers, allows the precise identification of many associated species, even those that resist cultivation (Stevenson et al. 2004). Over 200,000 bacterial entries exist currently for 16S rRNA, and these sequences can place most surveyed bacterial taxa securely into genera, if not species (Rupp 2004; Ueda et al. 2004).

The gut of tephritid fruit flies in general (Behar et al. 2009; Lloyd et al. 1986; Wang et al. 2011) and B. tau in particular (Prabhakar et al. 2009a; Sood & Nath 2002) is a storehouse of bacterial community. In the present study also 30 different bacterial colonies on the basis of colony morphology were screened for attractancy, but only five isolates which were found promising in terms of their attractancy to adult B. tau flies and could have a great implication on their future management programmes were further characterized. All these bacterial isolates except P. putida were not reported earlier to be associated with B. tau and having any role in fruit fly ecology. The attractiveness of these isolates to fruit flies, however, suggests their possible role in the fruit fly nutrition and physiology. Concrete and concentrated efforts across the fruit fly species are needed to elucidate the complex phenomenon to some conclusion.

Delftia was isolated from B. tau for the first time, although its association was earlier reported with cotton bollworm, Helicoverpa armigera (Hübner) (Xiang et al. 2006); wood borer, Saperda vestita (Say) (Delalibera et al. 2005); tobacco caterpillar, Manduca sexta (Brinkmann et al. 2008); and mosquito, Aedes albopictus Skuse (Zouache et al. 2009). Whereas Pseudomonas putida and members of genera Pseudomonas have been reported from many insect species including fruit flies as well as from B. tau (Brinkmann et al. 2008; Sood & Nath 2002). The presence of Flavobacterium sp., Defluvibacter sp. and Ochrobactrum sp. has also been reported in other insect species, viz. Flavobacterium sp. from ant, Tetraponera binghami Forel (van Borm et al. 2002), honey bees (Mohr & Tebbe 2007) and tobacco caterpillar, M. sexta (Brinkmann et al. 2008); however, the present findings constitute the first report of their association with fruit flies, particularly Bactrocera species.

Defluvibacter sp. is a member of the bacterial family Phyllobacteriaceae from the class α-Proteobacteria and has not been reported from the gut of any insect species. An unassigned bacterium (member Phyllobacteriaceae) has, however, been reported from the gut content of Asian longhorned beetle, Anoplophora glabripennis Motschulsky (Geib et al. 2009). The bacterial family Phyllobacteriaceae is closely related to the Bradyrhizobiaceae, Methylobacteriaceae and Rhizobiaceae. Bacteria from the families Methylobacteriaceae and Rhizobiaceae have been reported from different insect species, viz. Rhizobium and Methylobacterium from the gut of Tetraponera ants (van Borm et al. 2002) and Rhizobium from the gut content of Asian longhorned beetle, Anoplophora glabripennis (Geib et al. 2009).

Ochrobactrum sp. belongs to the α-2 subclass of the Proteobacteria (De Ley 1992). The genus was first described by Holmes et al. (1988) and the phylogenetic position of Ochrobactrum sp. was defined by De Ley (1992) and Yanagi & Yamasato (1993) on the basis of DNA ± rRNA hybridization and 16S rRNA homology studies. Its closest known relative is Brucella (De Ley 1992; Yanagi & Yamasato 1993). Ochrobactrum sp. was reported from the insect gut (Asian longhorned beetle, Anoplophora glabripennis) by Geib et al. (2009); its closest relative, Brucella sp., was isolated from the gut of wood borer Saperda vestita and identified by 16S rRNA analysis by Delalibera et al. (2005).

Earlier Klebsiella oxytoca, Pantoea agglomerans, Staphylococcus sp., Pseudomonas putida, Bacillus sp. and Enterobacter agglomerans were reported from the gut of fruit fly, B. tau (Prabhakar et al. 2009a; Sood & Nath 2002). The present finding added a few more genera to this list, but as a whole all these reports suggested that a vast range of gut bacterial diversity exists in the B. tau system. Thorough isolation and characterization with culture-dependent and -independent techniques are therefore needed to explore the gut bacterial diversity of B. tau to understand the host behavior in relation to gut bacterial community. On the other hand, vertebrate pathogenicity of these bacteria can also not be denied as some species of Ochrobactrum, Delftia, Staphylococcus and Defluvibacter were previously reported as human pathogens (Battalgia 2008; Mastroianni et al. 1999; Ogston 1984; Preiswerkz et al. 2011); this, however, requires further investigations. Detailed investigations are also needed to establish the taxonomic positions of Delftia, Flavobacterium sp., Defluvibacter sp. and Ochrobactrum sp. up to species level using chemotaxonomic and molecular approaches to understand the tri-trophic interaction among host–fruit flies–microbes.