UW-Madison Dept of Bacteriology

Research Description

Research in the Handelsman lab focuses on understanding the structure and function of microbial communities, addressing questions that have challenged plant pathologists and microbiologists since the inception of our discipline. We are finding ways to better describe the organization of microbial communities through decoding the language by which the members communicate. This understanding will lead to development of new technologies to enhance plant health and discovery of new medicinal and agricultural chemistry.

The emerging image of microbial communication networks is one of numerous bacterial signals that are often scrambled or amplified by other members of the community. At this early stage in our understanding it is difficult to imagine how bacteria make sense of their chemical environment as they are bombarded with signals from all sides. But the study of carefully chosen systems will reveal the networks that provide order to the communication, and understanding communication, in turn, will help elucidate the structure and function of microbial communities. We have selected three discrete but complementary systems for this phase of our work, one of which follows the activities of a specific bacterium in a community context, one that explores the challenges of defining community, and one that represents a contained ecosystem that can be manipulated in the laboratory.

1. We are examining the interaction of Bacillus cereus with its host plant and the microbial community in which it functions, allowing for a detailed modeling of the chemical signaling in the plant rhizosphere that determines bacterial survival, gene expression, and effects on the host plant and on protists that affect plant health.
We have chosen biocontrol of plant disease by Bacillus cereus as a means of modeling a specific microbial communication systems. The portrait of biocontrol that has emerged over the last decade indicates that successful disease suppression depends on carefully calibrated signal exchange among the many partners of the system. We have determined that B. cereus strains are ubiquitous on plant roots and in soil and that many of these strains enhance plant health when inoculated onto seeds at planting. The bacterium suppresses disease through diverse mechanisms, including the release of small, diffusible molecules. We discovered a new antibiotic, zwittermicin A, that is central to disease suppression by preventing normal development of plant pathogenic protists. Zwittermicin also inhibits growth of certain bacteria in the microbial community of the soil. The mechanism of inhibition remains unsolved, although a mutant analysis in E. coli points to an effect on transcription, although studies of transcription in vivo and in vitro fail to support this mechanism. Our working hypothesis is that zwittermicin inhibits growth of bacteria by inhibiting transcription initiation in a promoter-specific fashion.

To define the chemical signals exchanged between B. cereus and its prokaryotic and eukaryotic associates, we developed a promoter trap system to identify genes that are regulated by biotic signals. We found a gene encoding a putative lipoprotein that is induced by a consortium of amino acids released from seeds during germination. The gene affects the ability of the bacterium to grow competitively on germinating seeds. The promoter trap system has also identified a suite of genes regulated by bacteria found associated with B. cereus in the rhizosphere and most of these genes appear to be novel. The current thrust of this work is to determine the function of genes regulated by chemical signals issued by other members of the microbial communities in which B. cereus must function.

Our basic research on microbial communication moves to application in the enhancement of crop health by B. cereus. One strain of B. cereus, UW85, consistently suppressed disease and increased soybean yields in five consecutive growing seasons in Wisconsin while another, AS4-12, suppressed alfalfa diseases as effectively as commonly used synthetic chemicals. However, despite is reliability and efficacy, B. cereus has not yet been registered for agricultural use because it contains human enterotoxins that are associated with food poisoning. We have attempted to remove the genes for enterotoxin production to make a safer bacterium for agricultural use. In our first attempt to remove the gene for one of the three subunits of the human toxin from the B. cereus genome, we demonstrated functional redundancy of one component of the toxin and we are now in the final stages of constructing a triple mutant that is deficient in all three enterotoxins. We intend to complete the genetics soon so that the mutant can be tested for biocontrol on soybeans in the ’02 field season. In addition to bringing to fruition our longstanding goal of providing effective biologically-based methods for disease control for agriculture, this work will also provide the basis for studying the role of the enterotoxins in the ecology of B. cereus in the soil microbial community.

2. We are using a new methodology, termed 'metagenomics,' to examine the genomes of uncultured microorganisms in soil communities, offering a fantastic and largely unexplored complement to our current cultu re-based understanding of microbial life that will bring new definition to the concept of microbial community.

This project is designed to understand the phylogeny, function, and modes of communication of the microbial life in soil that cannot be cultured by standard techniques. The soil likely contains the most complex and rapidly changing microbial community on Earth. The challenge in studying this community is in defining it in space and time and then describing its changing structure and accompanying functions. Although soil microbiology is one of the oldest branches of microbiology, the definition of the community has eluded us because we have looked at the soil through the petri plate, a prism that refracts, distorts, and limits the vision of its subject.

For a long time, microbiologists have known that fewer than 1% (and perhaps as few as 0.1%) of the viable bacteria in soil can be cultured on known media. Collaborative work between my lab and and others, as well as numerous studies by other groups have demonstrated through an analysis of the 16S rRNA genes of soil organisms that the uncultured Bacteria and Archaea diverged deeply from the cultured organisms. This work tantalized scientists around the world because of the predicted vast metabolic diversity that would likely be associated with such phylogenetic diversity. However, for years it was not clear how to tap into this genetic potential without attempting to culture every organism, which seemed unwieldy given the size of the soil community. To address this challenge, in a multi-lab collaboration involving the labs of Bob Goodman (http://www.plantpath.wisc.edu/fac/rmg.htm) and Jon Clardy (http://www.chem.cornell.edu/department/Faculty/Clardy/clardy.html), we developed the approach that we designated ‘metagenomics’, which is the analysis of the collective genomes of the microorganisms in the soil community. Our approach is to clone DNA in large fragments directly from soil into a culturable host and conduct a sequence-based and functional genomic analysis on it. The intended outcomes of this project are diverse and ambitious, including the isolation of new chemical signals, new secondary metabolites that might have utility to humans, and the reconstruction of an entire genome of an uncultured organism.

Although the metagenomic analysis of the soil is in its infancy, a number of exciting discoveries have been made that indicate that a cornucopia of chemistry and biology is waiting to be unearthed. In a collaborative effort, we constructed a number of libraries that collectively contain more than 1 Gigabase of anonymous soil DNA in E. coli. We have found two structurally related, novel antibiotics, designated turbomycin A and turbomycin B, a multigene pathway for a previously described antibiotic, and numerous 16S rRNA genes that indicate that a vast diversity of organisms, including many that diverge deeply from cultured microbes, contributed DNA to the library. The 16S rRNA genes or other phylogenetic indicators, sequence analysis, and functional screening provide the opportunity to link phylogeny and function. Ultimately, this information may provide the basis for strategies to culture the organisms from which the DNA was derived or may provide the basis for in situ probing to determine when, where, and with whom the genes are expressed.

While elucidating a comprehensive picture of the soil microbial community may not be feasible, we believe that we will be able to capture a full view of one microbial function in soil that confers a selectable phenotype using traditional and molecular approaches. To test this possibility, the Handelsman lab has recently initiated a new project to understand the mechanism by which bacteria acquire phosphorous from P-limited soils. We will use traditional culture-based methods as well as metagenomics to identify genes responsible for acquisition of phosphorous from highly reduced forms, such as phosphonite, and mineral forms such as apatite and the highly insoluble lanthanide phosphates. The site of interest is the soil in a boreal forest in Alaska, including soils that are under permafrost and others that undergo rapid and frequent freeze-thaw cycles. In collaboration with a geologist, a microbial physiologist, and an ecosystem ecologist, we are hoping to determine the microbial contribution to phosphorous availability in a system that is dependent on P for productivity.

3. We are using the gypsy moth midgut as a wholly contained ecosystem, asking questions about the role of each of the microorganisms within the microbial community and as factors affecting the physiology of the host.

The third area of research in the Handelsman lab is the gypsy moth midgut, which serves as a model for studying a contained community, with the goal of attaining a comprehensive understanding of its function at the genomic, chemical, and ecological levels. We initially started studying the gypsy moth community after we and our collaborators demonstrated that zwittermicin A, the antibiotic produced by B. cereus strains that are active in biocontrol, is a powerful potentiator of Bacillus thuringiensis toxicity to gypsy moth larvae. This finding led us to study the molecular ecology of the microbiota of the insect midgut because one mechanism by which zwittermicin A might potentiate is by altering the composition of the midgut community. To test this hypothesis, we studied the community by culture-dependent and culture-independent methods. We found that the gypsy moth midgut contains a rich and diverse community; that culture-independent methods reveal a greater diversity of bacteria than culturing; and that zwittermicin A and foliage species in the diet alter the composition of the community. Perhaps the most intriguing finding is that all gypsy moths examined, independent of the source of the larvae or their diet, contained a culturable Enterococcus faecalis and a 16S rRNA sequence of a g-Proteobacterium that deeply diverged from all known sequences in this Division of bacteria.

We are keen to study the interaction between the readily culturable E. faecalis and the as yet unculturable g-Proteobacterium. But there are few methods in microbial ecology that facilitate the isolation of single microbial variables in an otherwise undisturbed ecosystem. Therefore, our initial studies will involve developing methods to subtract certain species from the midgut community opening up the possibility of defining the function of each member of the community. To accomplish this, in collaboration with entomologist Ken Raffa and experts in bacterial plasmids and gene expression , we have begun building microbial assassins, which we call “The Trojan Horse,” that will enter a community and selectively destroy certain members. The initial application of this technology will be to test the hypothesis that E. faecalis provides the gypsy moth with developmental signals or nutrients, or reduces its vulnerability to invading pathogens or parasites.

The gypsy moth system presents the opportunity to conduct a comprehensive analysis of a metagenome since there are only about a dozen species in the gut of the caterpillar. It is feasible, with today’s technology, to fully sequence the genomes of these organisms and reassemble them. Moreover, the community is sufficiently simple and defined, that it is possible to assign functions to most of the organisms.