The red palm weevil (RPW) Rhynchophorus ferrugineus Olivier (Coleoptera: Curculionidae) is widely considered the most damaging insect pest of palms in the world, even in all the countries where it has been accidentally introduced [1]. RPW larvae feed within the apical growing point of the palms, producing a wet fermenting frass inside the tunnels [2], creating extensive damage to palm tissues and weakening the structure of the palm trunk; the resulting damage is often only visible long after infestation, when palms are close to death [3–5] (Additional file 1).

Insect intestinal tracts harbour rich communities of non-pathogenic microorganisms [6, 7] and a single gut can harbour 10⁵–10⁹ prokaryotic cells [6] that have been affiliated to twenty-six phyla, at least for the insects studied to date [8]. It is increasingly evident that the microbiota of animals (humans included) plays a remarkable role in the host life. The genetic wealth of the microbiota affects all aspects of the holobiont’s (host plus all of its associated microorganisms) fitness such as adaptation, survival, development, growth, reproduction and evolution [9]. When not strictly essential for survival, the insect gut microbiota affects many aspects of host phenotype; it can increase the digestive efficiency of soluble plant polysaccharides [10, 11] and can mediate interactions between the host and potential pathogens [12]. Recent work suggests that the gut microbiota not only provide nutrients, but is also involved in the development and maintenance of the host immune system. However, the complexity, dynamics and types of interactions between the insect hosts and their gut microbiota are far from being well understood [13]. Understanding the relationship dynamics between insects and their microbiota can improve the biocontrol of insect pests, which is a focus of much insect gut microbiology studies.

Despite the economic and environmental damages caused by the RPW in all the areas where it is endemic and where it has been accidentally introduced, little is known about its gut microbiota. The bacterial community that is embedded in the frass produced inside the tunnels of the palm Phoenix canariensis Chabaud by the RPW larvae is dominated by Enterobacteriaceae with a facultative fermentative metabolism [2].

The purpose of this study was to analyse the diversity of the gut microbiota of the R. ferrugineus larvae, that represent the development stage responsible for damages to palms. Field-caught larvae were sampled from its favourite host P. canariensis in different seasons and sites in Sicily (Italy), and analysed for the diversity of their gut microbiota. The analysis of the bacterial community was carried out by culture-independent methods using temporal thermal gradient gel electrophoresis (TTGE) and FLX454 pyrosequencing of PCR-generated amplicons from the 16S rRNA gene.

In this study, the diversity of the gut microbiota of Rhynchophorus ferrugineus (RPW), collected on infested palm trees Phoenix canariensis, was first analysed by TTGE of the PCR-amplified bacterial 16S rRNA gene fragments. The TTGE profiles obtained from different lots of larvae, sampled in different seasons and geographical sites, show relatively low complexity (average of 25 OTUs) and high similarities regardless the site of sampling and season, suggesting that the composition of the RPW microbiota is stable over time and among pools of larvae from different host trees.

In order to identify the gut bacterial community of RPW larvae, the variable region 2 (V2) of the bacterial 16S rRNA gene, already successfully employed in the analysis of several microbial communities [17–19], was analysed by pyrosequencing. The analysis confirmed that the bacterial community of the RPW larvae has low diversity although, as expected, more OTUs were identified in respect to TTGE analysis. Contrasting results are reported for bacterial diversity of gut microbiota of other coleopterans; high diversity and complexity was observed among tree xylophagous beetles that rely on the microbiota for efficient lignocellulose metabolism and thus survival [8], while low diversity was recorded in the gut of the red turpentine beetle [20]. The RPW larvae are the major responsible for the palm damages because they live throughout their development inside the palm stem, feeding exclusively on palm tissues. This peculiar lifestyle may account for the low diversity detected in the gut of field sampled larvae of R. ferrugineus, regardless the investigation methods. There is strong evidence that mainly taxonomy and diet of the host can affect an organism’s gut microbial community [8, 21]. RPW larvae feed on nutrient-poor palm tissues and sap that contain mainly sucrose and glucose [22] but are poor of nitrogen [20, 23, 24]; an excess of sugars is known to reduce the complexity of the gut microbiota [25, 26]. Conversely, complex substrates, such as lignocellulose-derived materials, select complex gut bacterial communities even in highly divergent insect groups [8]. Feeding on live trees may also expose the insect to tree physiological responses, that may play a selection pressure on its microbiota. Moreover, the antimicrobial activity of larvae [27] may further shape the gut community of RPW.

The gut of RPW larvae is dominated by three phyla, Proteobacteria, Bacteroidetes and Firmicutes, that account for 98% of the assemblage. These same phyla were also found in the sugarcane weevil Sphenophorus levis Vaurie [28], which belong to the Dryophthorinae subfamily as R. ferrugineus, and that is the only weevil, to our best knowledge, that has been characterized in its microbiota. Proteobacteria and Firmicutes represent also the predominant bacterial phyla in bark beetles [20] and, in general, in all insect guts studied so far, while Bacteroidetes are more prevalent in termites, detritivorous insects and, among Coleoptera, in the root feeding Melolontha melolontha L. (Coleoptera: Melolonthidae) [8].

The genus Dysgonomonas is, unexpectedly, the most represented in the gut of RPW larvae. Dysgonomonas (phylum Bacteroidetes) are facultative anaerobes with a fermentative metabolism producing acids and no gas, that were first recovered from a human infected gall bladder [29]. Dysgonomonas is described as an opportunistic human pathogen but its habitat is unknown. Members of this genus were recently detected in microbial fuel cells (MFC) anode biofilms [30], in the gut of house flies (Musca domestica L.) [31] and in eight separate Drosophila populations where its presence is not restricted to any one locality, species, or diet type [21]. Its presence in such a high number in the insect gut, and in RPW gut in particular, deserves to be further investigated because it might play an important role in the insect biology.

Salmonella, Enterobacter, Budvicia and other Enterobacteriaceae are highly represented in the 454 assemblage; as in other insects, they could play a beneficial role in nutrition, in the degradation of plant polymers and fermentation of sap sugars. Members of Enterobacteriaceae were also identified as intracellular symbionts of grain weevils Sitophilus spp. (Curculionidae) [32] and some isolates are able to fix nitrogen, thus contributing to a supplementary nitrogen source [20, 23].

Entomoplasma is the sixth genus to be represented in terms of abundance in the RPW gut (3%). Entomoplasma is a glucose fermenting non-helical mollicutes and its presence in the RPW gut is consistent with what is presently known of its habitat. This genus could be considered a marker of the Coleopteran microbiota. All five currently described Entomoplasma species, in fact, were isolated from the gut or haemolymph of various firefly beetles (Coleoptera: Lampyridae) and green tiger beetles (Coleoptera: Cicindelidae) [33].

In spite of being affiliated to three different phyla, all the first six dominating bacterial genera of the RPW gut are facultative or obligate anaerobes with a fermentative metabolism. Dysgonomonas is a facultative anaerobe that produces acids and no gas; Lactococcus is a omofermenting lactic acid bacterium and some species can produce acetoin. The other three dominating genera belong to the Enterobacteriaceae characterized by mixed acid fermentation with production of lactic, acetic, succinic acid and ethanol (Salmonella), or 2,3-butanediol fermentation, producing butanediol, ethanol, CO₂ and H₂ (Enterobacter and Budvicia). Entomoplasma is also a glucose fermenting bacterium. These results suggest that the peculiar life-style of RPW larva and its gut exert a strong selective pressure towards those microbial species that are specialised to grow in a high sugar environment and that these species probably have a competitive advantage on those that cannot tolerate organic acids.

Interestingly, two genera of Enterobacteriaceae, Pantoea and Rahnella, which had previously been isolated from frass, were not detected in the gut. Rahnella isolates from frass have their closest relatives in components of the microbiota of the red turpentine beetle Dendroctonus valens LeConte (Coleoptera: Scolytidae) [20] and of the larvae of the lepidopteran Hepialus gonggaensis Fu & Huang (Lepidoptera: Hepialidae) [34]; Pantoea from frass are close to bacteria of the fungus garden of the leaf-cutter ant Atta colombica Guérin-Méneville (Hymenoptera: Formicidae), where they contribute to external plant biomass degradation and nitrogen fixation [35] (Additional file 5). High identities of RPW gut isolates with frass isolates and with other beneficial insect-associated bacteria suggest that the RPW gut microbiota cooperates, in a continuum with the frass microbiota, to the fitness of the larva inside the palm. Thus, while a unique midgut-associated microbiota can be distinguished from the environmental bacterial community in some insects [36], the peculiar lifestyle of RPW larvae makes such discrimination difficult or probably meaningless. In fact, RPW larvae feed in a very confined environment, consisting of tunnels burrowed in the palm trunk, where they continuously ingest both fresh palm tissues and frass, composed of chewed and/or digested plant tissue, so that re-acquisition by ingestion of bacteria from the environment is highly probable to occur.

Beyond nutritional aspects, the gut and frass fermentation products, such as acetoin and organic acid derivatives, ethyl esters, act as insect aggregation pheromones playing a role of attraction to other insects and promoting new oviposition events on the same tree [37].

Acidification caused by bacterial fermentation could also confer other advantages to the insect host, as some microbial toxins of Lepidoptera, such as Bacillus thuringiensis toxins, are activated by alkaline conditions. Thus, the RPW microbiota might help protect this insect from B. thuringiensis toxin by decreasing the midgut pH [38]. Moreover, together with that of fermenting yeasts, the bacterial metabolic activity increases the temperature inside the palm tissues, helping weevil overwintering [39].

It is well known that culture-based methods fail in describing the total diversity of natural bacterial communities so that, as expected, culture-independent techniques yielded a higher diversity of bacteria. Culturing under aerobic conditions led to the detection of nine bacterial genera in the RPW larval gut. Both pyrosequencing and culturing revealed that Enterobacteriaceae is the most represented bacterial family in the gut of RPW larvae.

In this work, the culture-based approach helped in obtaining a better description of some members of Enterobacteriaceae as the complete sequence of the 16S rRNA gene could be obtained from the isolated bacteria. The pyrosequencing approach, that relies upon a short 16S rRNA gene fragment, did not detect sequences of the genus Klebsiella, that was instead abundantly isolated by culturing. Failing of its detection could be due to low variability of the V2 region between Klebsiella and Enterobacter[12, 40] and the sequences of Klebsiella might have been included in the genus Enterobacter by the RDP Classifier software.

Another genus detected by cultivation but absent in the 454 assemblage was Bacillus that might be present at very low levels in the RPW gut, so that its detection might be impaired by PCR biases. Bacilli isolated from the gut are close to B. muralis and B. simplex, and cluster separately from palm endophyte bacilli and frass bacilli previously isolated, that are related to the B. cereus/thuringiensis group.

Cuticle Bacillus isolates, that survived sterilization procedures, form a separate cluster from gut bacilli and are closer to the Bacillus isolates previously obtained from frass and from healthy palms as endophytes [2] (Additional file 5). This suggest that they belong to a bacterial community external to the larvae, that might contribute to the fitness of larvae inside the plant tissues. The cuticle aerobic spore-forming bacteria might produce antimicrobial molecules that could negatively affect the sensitivity of the larvae to entomopathogenic fungi and bacteria [41].

A low bacterial diversity and the presence of a prevailing sugar-fermenting microbiota suggest that the digestion of plant polymers (cellulose, hemicellulose) is not a primary function of the RPW larvae. However, cellulolytic and hemicellulolytic bacteria were previously isolated by enrichment cultures from the gut of RPW larvae and were mainly affiliated to the Gamma and Alphaproteobacteria of the genera Pseudomonas, Enterobacter Microbacterium and Paenibacillus [2]. The presence of these genera in the RPW gut was confirmed by pyrosequencing (Additional file 6). Matching the 454-reads with the 16S rRNA gene sequences of the gut cellulolytic isolates, we obtained up to 99% identity of cluster_3902 (3 sequences) with the cellulolytic isolate Pseudomonas sp. R-8 (Genbank accession JN167546) and 98% identity of five different clusters (for a total of 159 sequences) with the cellulolytic RPW gut isolate Enterobacter sp. R-10 (Genbank accession JN167548) (Additional file 6). Both isolates were positive to the Congo Red test and able to grow on xylan in pure culture [2] but their hydrolytic activity on plant polymers in situ has to be demonstrated (as, for example, it might be inhibited by sap sugars).

The gut of insects that rely on sugar-based diets, particularly those belonging to the orders Diptera, Hymenoptera and Hemiptera, are often dominated by acetic acid bacteria (AAB), [16]. Although the larval RPW diet is almost exclusively based on sugars, we were unable to detect AAB using a consolidated method based on the enrichment culture technique [42]. Moreover, the absence of AAB in the RPW gut was confirmed by deep sequencing, where only two sequences were affiliated to the genus Acidisoma (Acetobacteriaceae) (Additional file 2). AAB are common in sugary acidic and alcoholic habitats, but are usually limited by nutrients other that their primary carbon source. AAB are common in fruit-feeding Drosophila species but are absent in flower-feeding flies [21]. Their absence in the RPW larvae could be explained by microbial interactions occurring inside the gut. The enrichment cultures set to specifically isolate AAB led, instead, to the isolation of Klebsiella strains that could outcompete AABs and that could fulfil also the nitrogen fixation function [20, 43], allowing the insect to live on a substrate with a high C/N ratio.

The RPW microbiota is composed mainly of facultative and obligate anaerobic bacteria with a fermentative metabolism. These bacteria might have a key role in the insect nutrition, and other functions that need to be investigated. Further research, focusing on the functional traits of the bacteria inhabiting the gut of R. ferrugineus, is critically important to establish if some bacteria may exert an essential role for the insect or might represent an obstacle for the optimization and promotion of the use of entomopathogenic fungi and bacilli in an integrated pest management approach.