Molecular microbial ecology

Understanding symbiosis requires identifying and characterizing the bacterial partners This has classically been done with microscopic observations and by isolating bacteria associated with the host, further studying the isolated microorganisms, manipulating them, and reintroducing them into their host when possible . Although this approach yields thrilling results (e g , the legume-rhizobia symbiosis), it is not always possible to characterize the molecular events occurring during the interaction of the microbial partner with its host, particularly when the microbial partner cannot be isolated Analysis is further restricted when community structure and its diversity are the subject of the investigation The development of methods enabling the detection, identification, quantification, tracking, and more recently whole genomes sequence analysis of unculturable bacteria has revolutionized the field of microbial ecology in general, and that of symbiosis and mutual-istic interactions in particular This quantum leap is mostly attributable to the availability of ever more sophisticated techniques that can be rapidly adopted by research laboratories . Most prominent among those are the rRNA-based technologies that use ribosomal genes (mainly the 16S rRNA gene) as phylogenetic markers Tools based on the rRNA approach can be used, among other possibilities, to analyze population structure and to compare community patterns, for in situ quantification, and for studying spatial distribution at the microbial scale In recent years the use of the rRNA approach has enabled the further characterization of the enteric microbiota of many insects (Egert et al , 2003; Reeson et al , 2003; Dillon and Dillon, 2004; Mohr and Tebbe, 2006) Excellent books and reviews covering these methods have been published (for further details see Van Elsas et al , 2007; Akkermans et al , 2004), and therefore a description of their principles is not within the scope of this chapter

Understanding the evolution of symbiotic relationships is now within our reach: lineages can be traced back, the specificity and depth of the interaction can be determined with great precision, and impacts on genome evolution in both partners measured A now classical protocol applied for the discovery of microorganisms associated with a host is to extract DNA from an organ or from the whole host body and to construct a 16S rDNA library based on the amplification of the target gene or parts thereof The resulting clones can then be grouped according to the cloned sequences' restriction patterns (e .g., amplified rDNA restriction analysis, ARDRA) and representative inserts of each group sequenced or random clones sequenced The size of each ARDRA group can be used as an indication of the distribution of the various sequences within the community As large-scale sequencing becomes cheaper, sequencing large numbers of random clones is becoming common, yielding better information on diversity and richness (Huber et al , 2007) Based on the acquired data, specific oligonucleotides can be designed for fluorescent in-situ hydridiza-tion (FISH), and used for localizing and quantifying target organisms in the host's tissues . These tools have proven extremely successful, revealing hitherto unknown associations between bacteria and arthropods (Favia et al , 2007; Fukatsu and Nikoh, 2000; Kikuchi et al , 2005), deciphering modes of transmission between parents and progeny (Dobson, 2003; Kikuchi et al , 2007; Wang et al , 2004), uncovering patterns of genome evolution (Hosokawa et al , 2006), and pointing to multiple partners' symbioses (Ikeda-Ohtsubo et al , 2007), including the occurrence of intracellular bacterial symbionts of bacteria (von Dohlen et al , 2001)

The analysis of the microbiota's community structure can also be pursued with denaturing gradient gel electrophoresis (DGGE), single-strand conformation polymorphism (SSCP), or terminal fragment length polymorphism (T-RFLP) These techniques enable population profiling of large numbers of samples in parallel, and with the proper statistical tools, community patterns can be compared (Lacava et al ., 2007; Mohr and Tebbe, 2006; Donovan et al ., 2004; Reeson et al ., 2003) . In the two former techniques, bands may be further extracted from the gels and sequenced, enabling identification of specific populations

These tools by which uncultured microorganisms are discovered and identified have become instrumental in microbial ecology Yet, biochemical characterization and genetic manipulations are greatly improved when the target organism is isolated Analysis of sequences originating from clone libraries or from DGGE/SSCP extracted bands can greatly facilitate the isolation of target bacteria Identification directs the researcher toward specific bacterial groups and therefore toward adequate isolation protocols, thereby greatly increasing the odds of isolating or enriching for the target organism Recent examples of this efficient strategy are the identification and isolation of Asaia sp that form dominant populations in the Asian malarial mosquito vector Anopheles stephensi (Favia et al , 2007) and the enrichment for an aerobic phototrophic acidobacterium from a Yellowstone National Park spring (Bryant et al , 2007)

In our work on the medfly gut community, we make large use of a polyphasic approach that includes the isolation of culturable bacteria on growth media and molecular, culture independent analyses . The enormous majority of the medfly gut community is composed of various Enterobacteriaceae that were found using both approaches (see below) Further, direct and culture-based identification lead us to hypothesize that diazotrophic and pecti-nolytic functions are performed by many members of the gut community Demonstration of these hypotheses was achieved using specific media and chemical analyses (Behar et al , 2005, and below) We further studied the relationship between gut bacteria, the insect's developmental stage and its fruit host, identified seasonal and geographical fluctuations in community structure, and isolated a minor but important component of the community (Behar et al , 2005, 2008a, 2008b, and see below)

Metagenomics can provide unequaled amounts of data on microbial community structure and function, especially when combined with large scale 16S rRNA gene library analyses As an example, the hindgut microbiota of a wood termite was recently described using metagenomics (Warnecke et al , 2007) The diversity and richness of this bacterial community was revealed along with metabolic and enzymatic functions linked to it, such as CO2 reductive acetogenesis, N2 fixation, cellulose and xylan degradation genes, and lig-nocellulose degradation

Novel sequencing equipment based on pyrosequencing, Illumina (Solexa), or SOLID technologies (Margulies et al , 2005; Metzker, 2005) add enormous power and very high throughput capacities to the researcher's tool box . First applications to the field of microbe-arthropod interactions have already led to tangible results: for example, the possible cause of colony collapse disorder (CCD) in Apis mellifera, the European honey bee, may be a dicis-trovirus This virus was rapidly identified by very large scale sequence analysis using pyrosequencing (Cox-Foster et al , 2007) Although this study was initiated to help identify the cause of the disease, it also provided much data on the composition of bacterial, fungal, and viral communities of the bee It should be mentioned, however, that at present, the phylogenetic resolution of these high throughput approaches is rather limited due to the short sequence reads . The greater precision achieved in this study was due to complementary analysis of 16S rRNA gene libraries .

New techniques have been proposed that expand the sensitivity of the PCR-based rRNA approach . The use of inosine at the 3' end of 16S rRNA-targeted primers instead of a specific base was shown to substantially increase the proportion of phyla that are poorly amplified, or not amplified at all, when universal primers are used (Ben-Dov et al ., 2006) . To date, this technique has only been applied to study a disease in corals (Barneah et al., 2007). We have experimented with suicide polymerase endonuclease restriction (SuPER) PCR, a novel rRNA-based approach (Green and Minz, 2005) . Whereas inosine-based primers enable effective amplification of sequences missed by standard primers, in SuPER PCR, sequences yielding dominant amplicons can be selectively digested . This "frees" the reaction to amplify sequences originating in minor populations (Green and Minz, 2005) . In our work, the application of SuPER PCR in a DGGE format yielded new banding patterns . Band analysis showed they all originated from various species of pseudomonads (more below)

0 0

Post a comment