Microbial Ecology

, Volume 64, Issue 1, pp 268–278

Gut-Associated Bacteria Throughout the Life Cycle of the Bark Beetle Dendroctonus rhizophagus Thomas and Bright (Curculionidae: Scolytinae) and Their Cellulolytic Activities

Authors

  • Jesús Morales-Jiménez
    • Departamento de Microbiología, Escuela Nacional de Ciencias BiológicasInstituto Politécnico Nacional
    • Laboratorio de Ecología Molecular, Departamento de Sistemas BiológicosUniversidad Autónoma Metropolitana—Xochimilco
  • Gerardo Zúñiga
    • Departamento de Zoología, Escuela Nacional de Ciencias BiológicasInstituto Politécnico Nacional
  • Hugo C. Ramírez-Saad
    • Laboratorio de Ecología Molecular, Departamento de Sistemas BiológicosUniversidad Autónoma Metropolitana—Xochimilco
    • Departamento de Microbiología, Escuela Nacional de Ciencias BiológicasInstituto Politécnico Nacional
INVERTEBRATE MICROBIOLOGY

DOI: 10.1007/s00248-011-9999-0

Cite this article as:
Morales-Jiménez, J., Zúñiga, G., Ramírez-Saad, H.C. et al. Microb Ecol (2012) 64: 268. doi:10.1007/s00248-011-9999-0

Abstract

Dendroctonus rhizophagus Thomas and Bright (Curculionidae: Scolytinae) is an endemic economically important insect of the Sierra Madre Occidental in Mexico. This bark beetle has an atypical behavior within the genus because just one beetle couple colonizes and kills seedlings and young trees of 11 pine species. In this work, the bacteria associated with the Dendroctonus rhizophagus gut were analyzed by culture-dependent and culture-independent methods. Analysis of 16S rRNA sequences amplified directly from isolates of gut bacteria suggests that the bacterial community associated with Dendroctonus rhizophagus, like that of other Dendroctonus spp. and Ips pini, is limited in number. Nine bacterial genera of γ-Proteobacteria and Actinobacteria classes were detected in the gut of Dendroctonus rhizophagus. Stenotrophomonas and Rahnella genera were the most frequently found bacteria from Dendroctonus rhizophagus gut throughout their life cycle. Stenotrophomonas maltophilia, Ponticoccus gilvus, and Kocuria marina showed cellulolytic activity in vitro. Stenotrophomonas maltophilia, Rahnella aquatilis, Raoultella terrigena, Ponticoccus gilvus, and Kocuria marina associated with larvae or adults of Dendroctonus rhizophagus could be implicated in nitrogen fixation and cellulose breakdown, important roles associated to insect development and fitness, especially under the particularly difficult life conditions of this beetle.

Introduction

Microbial communities of many groups of insects have been widely studied [7, 11, 13, 36, 40]. In particular, spectacular examples of species-level bacterial diversity have been found in the gut, and complex associations have been recognized between gut bacteria and insects [7, 27]. These interactions are diverse, ranging from antagonism and commensalism to mutualism and from obligate to facultative [17, 30]. In addition gut bacteria can contribute to insect development and survival through synthesis of essential nutrients, nitrogen fixing, uric acid recycling, food digestion, pheromone production, and metabolism of toxins [6, 20, 31, 38, 43]. They also come to have important implications in fitness, niche diversification, and species diversification [30].

Bark beetles carry on their life cycle on nutritionally poor and unbalanced substrates as phloem, bark and wood [58]. These substrates are rich in complex polysaccharides (e.g., cellulose and hemicelluloses), but scarce in other nutrients (e.g., assimilable nitrogen). While associations between bark beetles and their gut bacteria have barely been studied, their environmental characteristics strongly suggest that both bacteria and yeasts may have important functional roles in cellulose breakdown as well as provision of B vitamins, sterols, and/or essential amino acids [32, 49]. Another possible function of gut bacteria is to facilitate the detoxification process [2] as bark beetles must tolerate defensive compounds present in the host tissues during the host colonization and insect development. Perhaps, this toxic environment exerts a selective pressure that could explain why bacterial communities that have been studied in bark beetles are less diverse compared to those found in other insects.

Dendroctonus rhizophagus Thomas and Bright (Coleoptera: Curculionidae: Scolytinae) is endemic of the Sierra Madre Occidental in the northwest from Mexico, where these parasites kill trees of 11 pine species [34]. It has an atypical behavior as this species does not carry out mass attacks as do other species of the genus. Usually, only one couple colonizes the bottom of the stem of seedlings and young pine trees (< 3 m, 10 cm diameter) [12, 51].

The life cycle of this species is annual and synchronous and is regulated largely by conditions of temperature and humidity. The emergence, dispersal, and colonization of Dendroctonus rhizophagus occur in late July and early August, while oviposition and larval development take place in July–April, and finally, pupation and imago maturation happen in May–July [19].Dendroctonus rhizophagus always kills their hosts, unlike other stem-colonizing bark beetle species such as the black turpentine beetle Dendroctonus terebrans and the red turpentine beetle Dendroctonus valens, which, in their native distribution range, develop within large pine trees (> 5 m) without killing them. This condition probably strongly limits the presence of microorganisms in the gut of this species because they must survive in an environment poor in nutrients and toxic due to the effect of tree monoterpenes. However, those gut-associated microorganisms that are able to survive probably harbor very effective and efficient metabolic functions key for the survival of their hosts. That is, both insects and microbiota must overcome a quickly changing environment due to the rapid degradation experienced by trees in a short period immediately after the onset of colonization. For this reason, the aim of this work was to describe the gut-associated bacterial community of this bark beetle in different developmental stages by culture and culture-independent methods and to test the cellulolytic capacities of cultured bacteria.

Material and Methods

Bark Beetle Collection

Larvae, pupae, and feed emerged adults from Dendroctonus rhizophagus were collected from two geographical locations in the Sierra Madre Occidental, Mexico (Table 1). All samples were obtained manually, directly from galleries of infested pine trees using fine forceps. They were then transported to the laboratory in sterile vials containing sterile moist paper. Larva, pupa, and adult insects, after being disinfected superficially with 70% ethanol and submerged repeatedly in a phosphate buffer solution (PBS) to avoid external contamination, were dissected under sterile conditions. In the case of adults, the gut was extracted after removal of elytra, wings, and tergites to expose the insect abdomen. The individual guts were transferred to a 1.5-ml microcentrifuge tube with 0.2 ml of PBS or culture medium.
Table 1

Dendroctonus rhizophagus samples used in this study

Location in Mexico

Location code

Latitude/longitude

Host tree

Insects (No.)

San Juanito, Bocoyna, Chihuahua

BCH2

27°92' N/107°60' W

Pinus engelmannii

Adults (20)

Pupae (20)

Eggs (300)

El Salto, Durango

SD

23°50' N/105°22' W

Pinus arizonica var cooperi

Larvae (40)

Microbiology and Molecular Biology Techniques

Bacterial isolation and culturing were assessed with techniques previously described [35]. Bacterial viable counts were performed on single guts of at least five insects for each life stage and sex. After scoring CFU values, single colonies were selected, and pure cultures were stored at -70°C for further analysis.

Bacterial and gut metagenomic DNAs from five larva, pupa, and adult beetles were extracted following protocols described by Hoffman et al. [25] and Morales-Jiménez et al. [35]. The pure DNAs were stored at -70°C until they were used in molecular techniques. RAPD fingerprints were generated from bacterial isolates in order to recognize bacterial related species. For this purpose, a single primer was used, and PCR conditions were those described by Williams et al. [57]. 16S rRNA genes of bacterial isolates and the gut bacterial community were amplified by using the primers and PCR conditions described in Relman [44]. PCR products were purified using the QIAquick PCR purification kit (Qiagen, Valencia, CA) and sequenced in an ABI PRISM 310 genetic analyzer (Applied Biosystems, Foster City, CA) using the same primers. 16S rRNA libraries of gut metagenomic DNA were generated by employing purified PCR products cloned in Escherichia coli Top-10 cells (Invitrogen-Life Technologies, Carlsbad, CA) with pJET1.2/blunt (CloneJETTM PCR Cloning Kit; Fermentas, Glen Burnie, MD) according to the manufacturer's instructions. Transformants were subjected to plasmid extraction by standard methods [45], and a restriction analysis with EcoRI was performed to detect insertions. Plasmidic DNA of each clone of the 16S rRNA library was digested with HpaII and HhaI endonucleases to display RFLP patterns in electrophoresis at 3% high-performance agarose 1000 (GIBCO Laboratories, Grand Island, NY). The plasmids of each RFLP pattern were extracted using a High Pure Plasmid Isolation Kit (Roche) and sequenced with the ABI PRISM 310 genetic analyzer (Applied Biosystems, Foster City, CA) using pJET 1.2 forward and pJET 1.2 reverse sequencing primers.

To know whether gut bacterial communities are different among sexes and insect life stages, a DGGE analysis was performed. DNA obtained from pools of five larva, pupa, and adult guts was used for PCR amplification of V3–V5 of the 16S rRNA gene. DGGE primers, PCR conditions, and the general methodology of this technique were performed following Muyzer et al. [37], while the silver stain procedure was that of Sanguinetti et al. [46]. Selected DGGE bands were excised from gels, reamplified with the same primers, and then sequenced as described previously. Maximum likelihood analysis of DGGE band sequences was performed using the K80 + I + G model (α = 0.434 for the gamma distribution; p-inv = 0.15; transition/transversion ratio = 0.93) with 1,000 bootstrap replicates.

Phylogenetic Analyses

Clone sequences obtained were tested for chimera structures using Bellerophon (http://comp-bio.anu.edu.au/bellerophon/bellerophon.pl) [28], and chimeras were excluded from further analysis. Sequences from clones and isolated bacteria were compared with the non-redundant GenBank library using BLAST search [3]. A collection of taxonomically related sequences was obtained from the National Center for Biotechnology Information (NCBI) Taxonomy Homepage (http://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome.html/). DNA sequences were aligned using CLUSTAL X [52], edited and confirmed visually in BIOEDIT [23].

Maximum likelihood analyses were performed using PhyML [21] (http://atgc.lirmm.fr/phyml/). MODELTEST 3.06 [41] was used to select appropriate models of sequence evolution by the AIC model [42]. The GTR + I + G model (α = 0.459 for the gamma distribution; A = 0.237, C = 0.233, G = 0.317, T = 0.212; p-inv = 0.204) was selected for the tree search. The confidence at each node was assessed by 1,000 bootstrap replicates. Anabaena affinis was used as outgroup. The similitude percentages among sequences were calculated using MatGAT v. 2.01 software [9]. The limits for genus and species were set at 95% and 97%, respectively [47]. Due to the high similarity among 16S rRNA gene sequences of Pantoea agglomerans, Klebsiella spp., Enterobacter spp., and Raoultella terrigena carbohydrate fermentation tests (d-melezitose and l-Sorbose ), an API-20E bacterial identification test strip (bioMérieux, Marcy I'Etoile, France) was employed to confirm the phylogenetic approach [24]. All sequences generated in this study were deposited in the GenBank database, under the accession numbers JN12146 through JN12175.

Isolation of Cellulolytic Microorganisms

Individual larva, pupa, and adult guts were placed in sterile Eppendorf tubes containing 200 μl of PBS and processed as described in previous sections. Serial ten-fold dilutions were spread on duplicate plates of Congo red agar (0.5-g l-1 K2HPO4, 0.25-g l-1 MgSO4, 1.88-g l-1 carboxymethyl cellulose, 0.2-g l-1 Congo red, 2-g l-1 gelatin, 100-ml l-1 soil extract, and 15-g l-1 agar). Plates were incubated in a growth chamber at 28°C for 3 to 5 days. The cellulolytic activity of microorganisms was detected by a clear zone around the colonies [50]. Pure cultures were obtained by multiple subsequent subculturing on Congo red agar. All strains with cellulolytic activities were grown in a mineral medium with carboxymethyl cellulose as the sole carbon source (2.5-g l-1 NaCl, 7.0-g l-1 K2HPO4, 3.0-g l-1 KH2PO4, 0.1-g l-1 MgSO4, and 2.5-g l-1 Carboxymethyl cellulose). Colony cellulolytic activity was indexed as the diameter of the cellulolytic halo divided by the diameter of the colony. At least two measurements were taken for each colony type.

Results

Bacterial Species in Dendroctonus rhizophagus Larvae

Ten RFLP patterns were identified in all clones of 16S rRNA gene libraries using HpaII and HhaI. A clone representing each RFLP pattern was sequenced and identified by nucleotide similitude and the phylogenetic approach, and from these analyses, two to three bacterial genera were identified in the larval gut by culture and culture-independent analyses. Phylogenetic analysis of 16S rRNA revealed that the relative abundance of Rahnella aquatilis and Pseudomonas fluorescens clones was 81% and 9%, respectively (Table 2). Meanwhile, the cultured populations of Rahnella aquatilis, Stenotrophomonas maltophilia, and Pseudomonas fluorescens were around 2.85 × 106 ± 4.29 105 (98%), 3.75 × 104 ± 6.61 103 (1%), and 2.8 × 103 ± 5.65 × 102 (0.1%) UFC/gut (% of total culturable bacterial population), respectively (Table 2 and Fig. 1).
Table 2

Bacterial taxa associated with guts of larvae, pupae, and adults of Dendroctonus rhizophagus in culture-dependent and culture- independent analyses

Identified bacteriaa

Detection strategiesb, c

Codes

Insect stage

Larva

Pupa

Adult

Rahnella aquatilis

Isolation

4-DR, 6-DR, 12-DR, DR-2A, PDR-D, PDR-4, PDR-1, PDR-H

2.85 × 106 ± 4.29 105

1.95 × 103 ± 1.02 × 103

2.2 × 106 ± 5.6 × 106

Library clones

DR-12A, DRL-D6, DR-A3, DR-E12, DRL-F6, DRL-B3, DR-D9, DR-A4

0.8

NP

0.96

DGGE

B-B, B-C

+

+

+

Raoultella terrigena

Isolation

DR-E5

ND

ND

+

Library clones

 

ND

NP

ND

DGGE

 

ND

ND

ND

Stenotrophomonas maltophilia

Isolation

1-DR, 2-DR, 3-DR

3.75 × 104 ± 6.61 × 103

ND

5.5 × 104 ± 3.13 × 104

Library clones

 

ND

NP

ND

DGGE

 

ND

ND

ND

Pseudomonas fluorescens

Isolation

DR-E10

2.8 × 103 ± 5.65 × 102

ND

1.6 × 103 ± 0

Library clones

DRL-1E, DRL-C11

0.1

NP

ND

DGGE

 

ND

ND

ND

Acinetobacter lowffii

Isolation

DR-A6

NP

NP

4.94 × 103 ± 1.8 × 102

Library clones

 

ND

NP

ND

DGGE

 

ND

ND

ND

Ponticoccus gilvus

Isolation

19-DR

ND

ND

2 × 103 ± 6.9 × 102

Library clones

 

ND

NP

ND

DGGE

 

ND

ND

ND

Kocuria marina

Isolation

DR-E1

ND

ND

+

Library clones

 

ND

NP

ND

DGGE

 

ND

ND

ND

Klebsiella sp.

Isolation

 

ND

ND

ND

Library clones

DRL-1C

0.1

NP

0.04

DGGE

 

ND

ND

ND

Propionibacterium sp.

Isolation

 

ND

ND

ND

Library clones

 

ND

NP

ND

DGGE

B-D

+

+

+

NP not performed, ND not detected

aThe limits for genus and species were 95% and 97%, respectively (Schloss and Handelsman, 2005)

bThe relative abundance was determined by viable count of cultured bacteria and clone proportions in 16S rRNA libraries

cAll attempts to construct a 16S rRNA library of pupa gut were unsuccessful. A possible explanation of this result could be the low bacterial densities detected in the pupa stage

https://static-content.springer.com/image/art%3A10.1007%2Fs00248-011-9999-0/MediaObjects/248_2011_9999_Fig1_HTML.gif
Figure 1

Maximum likelihood tree (-lnL = 11,257.67698) of bacterial community associated with Dendroctonus rhizophagus gut. The 16S rRNA sequence of Anabaena affinis was used as outgroup. Scale bar indicates 10% estimated sequence divergence. Bootstrap support values are indicated for major nodes having values ≥50%

Bacterial Species in Dendroctonus rhizophagus Pupae

The culture fraction of the microbial community associated with the pupa gut was dominated exclusively by Rahnella aquatilis, with a density around 1.95 × 103 ± 1.02 × 103 UFC/gut, and in the unculturable fraction by Propionibacterium sp. detected using DGGE.

Bacterial Species in Dendroctonus rhizophagus Adults

Culture-dependent and culture-independent analysis of adult guts revealed organisms affiliated with γ-Proteobacteria and Actinobacteria (Table 2 and Fig. 1). Likewise, in larvae, Rahnella aquatilis was the most abundant species (n = 25, 96%) in the 16S rRNA library, but only one clone was identified as Klebsiella sp. The γ-Proteobacteria Rahnella aquatilis, Raoultella terrigena, Stenotrophomonas maltophilia, and Pseudomonas fluorescens as well as Actinobacteria Ponticoccus gilvus and Kocuria marina were cultured (Table 2 and Fig. 1). The sequence of the 16S rRNA gene of the strain identified as Raoultella terrigena was clustered with sequences of Raoultella terrigena, Pantoea agglomerans, and Klebsiella spp., with similarities above 99%. On the other hand, metabolic tests of this strain, such as d-melezitose and l-sorbose fermentation and the API-20E test, confirm that this strain belongs to the Raoultella terrigena species. The populations of Rahnella aquatilis, Stenotrophomonas maltophilia, Pseudomonas fluorescens, and Acinetobacter lowffii were around 2.2 × 106 ± 5.6 × 106 (97%), 5.5 × 104 ± 3.13 × 104 (2%), 1.6 × 103 ± 0 (0.07%), and 4.94 × 103 ± 1.8 × 102 (0.2%) UFC/gut (% of total culturable bacterial population), respectively. The densities of Raoultella terrigena and Kocuria marina were not determined because these strains were isolated from enrichment cultures. The Actinobacteria Ponticoccus gilvus was not found in culture-independent analysis, but a population of approximately 2 × 103 ± 6.9 × 102 UFC/gut was estimated.

Diversity of Bacterial Species by DGGE Analysis

DGGE analysis of bacterial communities associated with larvae, pupae, and adults of Dendroctonus rhizophagus guts showed a low number of bands in both sexes and all three life stages, an observation consistent with the number of species detected by culture and culture-independent methods. The comparison of band patterns among larvae, pupae, adults, and sexes in Dendroctonus rhizophagus showed that the communities are highly homogeneous (Figs. 2 and 3a, b). A total of ten bands in males, females, and larvae were recognized, while only nine bands were present in pupae. The comparison of migration bands of DGGE analysis revealed that both Stenotrophomonas maltophilia and Rahnella aquatilis were broadly distributed in the alimentary canal of Dendroctonus rhizophagus during all stages of the life cycle. Band intensity of DGGE patterns suggests that Rahnella aquatilis was the most abundant bacteria in the Dendroctonus rhizophagus gut in all life stages. Sequences from some DGGE bands clustered with sequences of Rahnella aquatilis and the genus Propionibacterium (Fig. 4) support the idea that this member of the family Enterobacteriaceae is a major bacterium in the Dendroctonus rhizophagus gut. Furthermore, an intense DGGE band was associated with the 18S rRNA gene of insects in a BLAST analysis (Fig. 3b).
https://static-content.springer.com/image/art%3A10.1007%2Fs00248-011-9999-0/MediaObjects/248_2011_9999_Fig2_HTML.gif
Figure 2

Denaturing gradient gel electrophoresis of 16S rDNA PCR products obtained from field-collected guts of Dendroctonus valens and Dendroctonus rhizophagus. Adult female guts of Dendroctonus rhizophagus (lane 1 anterior midgut, lane 2 posterior midgut, lane 3 hindgut), adult female guts of Dendroctonus valens (lane 4 anterior midgut, lane 5 posterior midgut, lane 6 hindgut), adult male guts of Dendroctonus rhizophagus (lane 7 anterior midgut, lane 8 posterior midgut, lane 9 hindgut), adult male guts of Dendroctonus valens (lane 10 anterior midgut, lane 11 posterior midgut, lane 12 hindgut), larva guts of Dendroctonus rhizophagus (lane 13), and larva guts of Dendroctonus valens (lane 14 anterior midgut, lane 14 posterior midgut, lane 16 hindgut)

https://static-content.springer.com/image/art%3A10.1007%2Fs00248-011-9999-0/MediaObjects/248_2011_9999_Fig3_HTML.gif
Figure 3

Denaturing gradient gel electrophoresis of 16S rDNA PCR products obtained from field-collected guts of Dendroctonus rhizophagus.a Comigration comparison among adult male guts (lanes 1 and 3), adult female guts (lane 2), pupa guts (lane 4), larva guts (lane 5), larva guts of Dendroctonus valens (lane 6), mix of DNA from bacteria isolates of Dendroctonus rhizophagus (lane 7), and bacterial isolates (lanes 8 and 9, Stenotrophomonas maltophilia; lanes 10 and 11, Rahnella aquatilis). Full black arrow, Stenotrophomonas maltophilia. Black arrowhead, Rahnella aquatilis.b DGGE for band sequencing. Larva guts (lane 1), pupa guts (lane 2), adult male guts (lanes 3 and 5), adult female guts (lane 4), and larva guts of Dendroctonus valens (lane 6). The sequences of band A was clustered with 18S rRNA genes of insects, B and C with Rahnella aquatilis, and C with Propionibacterium sp.

https://static-content.springer.com/image/art%3A10.1007%2Fs00248-011-9999-0/MediaObjects/248_2011_9999_Fig4_HTML.gif
Figure 4

Maximum likelihood tree (-lnL = 1,166.06911) of some DGGE bands of the bacterial community associated with Dendroctonus rhizophagus gut. The 16S rRNA sequence of Anabaena affinis was used as outgroup. Scale bar indicates 10% estimated sequence divergence. Bootstrap support values are indicated for major nodes having values ≥50%

Cellulolytic Bacteria

Stenotrophomonas maltophilia and Ponticoccus gilvus strains with densities around 1,300 ± 81.64 and 1,000 ± 346.41 CFU/gut isolated from male guts showed cellulolytic activity on plates of Congo red–cellulose medium (Fig. 5, Table 3). Populations of cellulolytic bacteria were much lower than the total number of cultured bacteria (0.16–1.16 × 106 CFU/gut). Also, Kocuria marina showed cellulolytic activity, but this strain was isolated from an enrichment culture in a mineral liquid media with CMC as the sole carbon source. The extracellular enzyme activity indexes of Ponticoccus gilvus, Stenotrophomonas maltophilia, and Kocuria marina were approximately 8.2 ± 2.45, 4.18 ± 0.28, and 3.8 ± 0.65, respectively.
https://static-content.springer.com/image/art%3A10.1007%2Fs00248-011-9999-0/MediaObjects/248_2011_9999_Fig5_HTML.gif
Figure 5

Cellulolytic activity of bacterial isolates from the gut of Dendroctonus rhizophagus in Congo red-CMC. aPonticoccus gilvus 19 DR. bStenotrophomonas maltophilia 2 DR. cKocuria marina DRE1

Table 3

Identification and cellulolytic activity of bacteria isolated from D. rhizophagus gut

Isolate

Enzyme activity (indexa)

Stenotrophomonas maltophilia (2-DR)

8 ± 0.65

Kocuria marina (DR-E1)

4.18 ± 0.28

Ponticoccus gilvus (19-DR)

8.2 ± 2.45

aEnzyme activity was indexed as the diameter of the colony plus the clear zone around it divided by the diameter of the colony in Congo red–cellulose medium

Discussion

A total of nine bacterial taxa were found in the larva, pupa, and adult gut of Dendroctonus rhizophagus by culture-dependent and culture-independent methods. This bacterial diversity is slightly lower than that observed in other bark beetles [35, 54, 59]. The gut-associated bacterial community in pupae was lower in number and diversity than in larvae and adults (Table 2). In the pupa stage, only Rahnella aquatilis was cultured, and Propionibacterium sp. was detected by DGGE. Similar results have been reported in the pine engraver (Ips pini), where low bacteria densities were recorded in the pupa gut [16]. We think that such low bacterial densities in the pupa gut may be due to decreased metabolic activity during this developmental stage, although morphological changes that the gut undergoes during the insect metamorphosis could be an additional factor.

DGGE patterns of 16S RNA amplified from metagenomic DNA from larvae, pupae, and adults showed few differences in bands and no significant differences between the number of bands displayed at different life stages. The results show that at all stages, Rahnella aquatilis is a widespread bacterium in the Dendroctonus rhizophagus gut, whereas the other bacteria were present only in some particular stages. This situation was also observed in the bacterial community associated with the Dendroctonus valens gut by DGGE [35]. However, DGGE profile comparisons of gut-bacterial communities among the different development stages and between the sexes of these species showed that their total bacterial communities are similar (Figs. 2 and 3).

Kocuria marina, Ponticoccus gilvus, and Raoultella terrigena were exclusively found in the adult gut by culturing methods. Meanwhile, the genus Klebsiella was recognized only by 16S rRNA libraries in larvae and adults. The culturable fraction of gut microbiota of Dendroctonus rhizophagus was integrated by seven bacterial species. Similar results have been reported in the Dendroctonus frontalis, Dendroctonus valens, and Dendroctonus micans gut, where 13, 14, and 7 culturable bacteria species were recognized, respectively [35, 55, 59]. The scarce gut bacterial community is a common feature in all bark beetles studied, probably due mainly to the antibacterial activity of some monoterpenes present in pine resin and/or limitation of nitrogen sources [2, 53]. This tendency in bark beetles contrasts with the high bacterial diversity recorded in other insects, such as wood-boring beetles. For example, the Asian longhorned beetle (Anoplophora glabripennis) and the emerald ash borer (Agrilus planipennis) harbor a total of 23 and 42 bacterial genera in their gut, respectively. Curiously, both beetles develop in sapwood. The Asian longhorned beetle feeds on the cambium and phloem of maples, horsechestnuts, poplars, willows, elms, mulberries, and black locusts. On the other hand, the emerald ash borer feeds on the phloem of ash trees [48, 55]. Compared with other insects that feed on woody tissues, such as Reticulitermes speratus (Isoptera: Rhinotermitidae) that harbor around 268 phylotypes in 11 bacterial divisions [26, 39], the gut bacterial community of Dendroctonus rhizophagus appears to be extremely simple.

Full evidence from viable counts, DGGE, and relative abundance of clones in 16S rDNA libraries suggest that Rahnella aquatilis is the dominant species in the Dendroctonus rhizophagus gut. The abundance of this nitrogen-fixing bacterium suggests that the nitrogen-fixing process must be a very important dietary supplement of assimilable nitrogen for the insect. Rahnella aquatilis seems to be widespread in the gut of Dendroctonus species, although quantitative data are not available. It has also been isolated from other studied species, including Dendroctonus valens and Dendroctonus frontalis. It has frequently been isolated or detected in 16S rRNA libraries from larvae and adults of Dendroctonus valens [35], and it reached a frequency of detection of 12.8 % by DGGE [1]. Also, this bacterial species was the most common one detected in 16S rRNA libraries from larval and adult Dendroctonus frontalis gut, although it could be isolated from larva gut only once [54]. The evidence that this work and other studies cited suggests that Rahnella aquatilis might be recognized as a resident gut bacterium of Dendroctonus. Additionally, neither Rahnella spp. has been detected in other insects feeding on wood and phloem, such as the Asian longhorned beetle and the pine engraver [15, 16, 48, 55]. However, Rahnella aquatilis has been recognized as the dominant bacterium in Hepialus gonggaensis (a moth) and Decticus verrucivorus (wart-biter cricket) [14, 60], and it has also been isolated from seeds, ectomycorrhizas, and sapwood sawdust of conifers [10, 18, 29], suggesting that it could be a conifer endophytic bacterium. In this sense, in our laboratory, an attempt was made to detect Rahnella species in healthy pine phloem, but no strains were isolated (data not shown). Evidently, beyond the clear Rahnella aquatilisDendroctonus relationship, more studies are necessary to determine the ecological status of this nitrogen-fixing bacteria in the tree-bark beetle environment.

Rahnella aquatilis, Pseudomonas fluorescens, and Stenotrophomonas maltophilia γ-Proteobacteria were commonly found in larva and adult guts of Dendroctonus rhizophagus, suggesting that these bacteria are maintained during the metamorphosis from larva to adult. The genus Pseudomonas has been reported in adult guts of Dendroctonus frontalis and in the cerambycids Anoplophora glabripennis, Saperda vestita, Rhagium inquisitor, and Leptura rubra [22, 48, 55]. Although the role of this bacterium in the gut of wood- and bark-inhabiting beetles is unknown, we hypothesize that it could be involved in terpene transformation of plant resin compounds, due to its capabilities to metabolize monoterpenes and phenolics compounds [5]. Stenotrophomonas maltophilia has been found in adults of Dendroctonus valens and Dendroctonus frontalis and at all life stages of Ips pini [16, 35, 54]. This ubiquitous nitrogen-fixing bacterium, in association with other diazotrophs found in Dendroctonus rhizophagus, such as Klebsiella sp., Rahnella aquatilis, and Raoultella terrigena, may fix and concentrate assimilable nitrogen for insect nutrition, as has been demonstrated in other insects [4, 40]. Both insect larvae and adults exhibit acetylene reduction (data not shown), but the particular contribution of each nitrogen-fixing bacterial species in the gut must be determined.

On the other hand, Stenotrophomonas maltophilia (γ-Proteobacteria), Ponticoccus gilvus, and Kocuria marina (Actinobacteria) isolated from the Dendroctonus rhizophagus gut were the bacterial isolates capable of degrading carboxymethylcellulose in vitro. Neither Actinobacteria has ever been reported in other bark beetles or tree tissues. Although isolates of the genus Kocuria with cellulolytic activity have been isolated from the gut of the termite Zootermopsis angusticollis [56], to our knowledge, no cellulolytic capacities have been reported in members of the Ponticoccus or Stenotrophomonas genera. The presence of cellulose-degrading bacteria has been demonstrated in the gut of insects that feed on woody tree tissues, such as wood-boring beetles, including Saperda vestita and Agrilus planipennis [15, 55]. The cellulolytic bacteria obtained from the Dendroctonus rhizophagus gut could be involved in the degradation of cellulosic substrates such as pine bark and phloem, enabling them to serve as a carbon source. On the other hand, other bacteria, including Stenotrophomonas maltophilia, Pseudomonas, and Acinetobacter, could participate in the oxidation, fermentation, and hydrolysis of the cellulose and lignin derived aromatic products [33].

Cellulose and hemicellulose hydrolysis, anaerobic respiration, and nitrogen fixation take place in specialized dilated hindguts of termites, but no equivalent morphological adaptations in the bark beetle gut are observed [8]. Notwithstanding, we observed partial digested woody material in adults that recently colonized the pine trees and larva galleries of Dendroctonus rhizophagus. The gut of Dendroctonus spp. is histologically and morphologically compartmentalized, probably with their particular physiochemical microenvironments for cellulose degradation or nitrogen fixation. Evidently, more studies must be performed in order to recognize the role of gut cellulolytic bacteria in the Dendroctonus spp. life cycle.

In this study, a characterization was made on the bacterial community associated with the gut of larvae and adults of Dendroctonus rhizophagus, found by using culture and culture-independent techniques. This is the first report concerning the bacterial community within the gut of Dendroctonus rhizophagus. Several important bacteria were recognized, including nitrogen fixing and cellulolytic guilds, and their roles in the context of the insect–microbe relationship were discussed.

Acknowledgments

We would like to thank Félix Aguirre Garrido for the technical assistance with DGGE. This work was supported by grants SIP 20080688, 20090738, 20100430, and 20111068; IPN; and CONAFOR-CONACyT 69539. Jesús Morales-Jiménez would like to thank CONACyT, and PIFI-IPN for the scholarships.

Copyright information

© Springer Science+Business Media, LLC 2012