Final Report Summary - PBDNH (Learning from the locals: A metagenomic investigation into the plant biomass degrading capacity in Norwegian native herbivores)
Lignocellulosic (plant) biomass provides a giant potential source of energy and valuable products and is thought to play a major role in a future “bio-based economy”. Unfortunately, this enormous resource comprising of various grasses, wood, forest product residues and agricultural residues, so far remains largely untapped. This is predominantly due to the recalcitrance of lignocellulosic biomass, coinciding with a lack of understanding how this recalcitrance is overcome in nature and can be overcome industrially. Lignocellulosic biomass consists of three main components; cellulose, heteroxylans and lignin, with the relative proportions of the three dependent on the material source. There exist a number of current chemical and mechanical means of degrading lignocelluloses though they are expensive, slow and relatively inefficient. However in nature, lignocellulose can be readily deconstructed by enzyme-driven hydrolysis, and a variety of free-living organisms have subsequently evolved to take advantage of this material as a nutrient source. In particular, obligate herbivores have evolved to maintain a symbiotic relationship with a specialized consortium of gut microbes (microbiomes) that underpins lignocellulose deconstruction (and anaerobic fermentation).
Microbial community analysis consistently shows the gut microbiome of herbivores (i.e. the rumen of a cow) is commonly dominated by a limited number of frequently observed bacterial phyla, namely the Firmicutes and the Bacteroidetes. Degradation of the most abundant plant polysaccharide (cellulose) within the rumen has for the greater part been attributed to the metabolic capabilities of species affiliated to the bacterial phyla Firmicutes and Fibrobacteres. While the Bacteroidetes represent a numerically dominating phylum in the rumen their functional role is poorly understood due to their difficulty to cultivate in pure culture and thus examine in the laboratory. Therefore, in this project we have used the latest advancements in DNA sequencing, computational biology and enzyme technologies to target the uncultured Bacteroidetes species that dominate these natural environments and perform the processes we seek to understand.
In this project objectives we have focused on the microbiome of the Svalbard rumen, In particular, we have studied a dominant bacterium identified in this microbiome, referred to as SRM-1 ,whose genome was partially reconstructed from metagenomic data. In the final phase of the project we have characterized a polysaccharide utilization locus (PUL) from the SRM-1 genome, which revealed novel enzymes and binding proteins involved in biomass turnover. The PUL in question is encoded within a 30 kb gene cluster that encodes 13 genes, including two consecutive pairs of SusC/D homologues and seven glycoside hydrolases that implied activity against β-(1,4)-glucans (3 x GH5, GH94) and mannans (GH26, GH130, GH2). We have performed biochemical characterization on the outer membrane GH26 and GH5 (n=2) enzymes. All purified enzymes were screened by a newly developed technique for high-throughput endo-GH-substrate determination using chromogenic hydrogel substrates (via collaboration with Prof. William Willats at University of Copenhagen). GH5 and GH26 enzymes demonstrated versatile activity against mixed-linkage glucans, xylan, xyloglucan, galactomannan and amorphous cellulose; all substrates that are highly accessible to the bacterium in nature.
As anticipated, this functional characterization of the enzymatic machinery of an uncultures bacterium has revealed novel enzymes for biomass processing. Furthermore, the results reveal unprecedented substrate versatility within a single PUL, in contrast to previous publications which have demonstrated narrow specificity regarding the target glycans of PULs i.e. one cognate PUL for each substrate.
All Objectives for the project have been reached. More specifically for the four key sub-goals:
1. Creating a metagenomic dataset of the Svalbard reindeer rumen microbiome using 454 sequencing. Done and published (Pope et al., 2012, PLoS One). Here the microbial community has been identified.
2. Annotation of the metagenome to reconstruct a genetic and metabolic blueprint of the rumen microbiome of Svalbard reindeer including the use of newly developed bioinformatic software. Done and published (Pope et al., 2012, PLoS One). Here, the genetic potential of the microbial community has been mapped.
3. Discovery of new enzymes and/or accessory proteins for biomass conversion using functional screening of the fosmid library and sequence-based screening of the metagenomic dataset. Done and published (Pope et al., 2012, PLoS One); follow up work to be published in a paper that has been submitted for publication (Naas, Pope et al., 2013). Here, potentially interesting enzymes for biomass conversion have been identified by combining bioinformatics and functional analyses.
4. Over-expression and characterization of potentially interesting enzymes and accessory proteins and evaluation of their potential in applied bioenergy production projects in the host laboratory. Done; some results published in 2012 (MacKenzie, Pope et al., 2012) and follow up work to be published in a paper that has been submitted for publication (Naas, Pope et al., 2013). Here, interesting enzymes have been overexpressed, purified and characterized in detail. Work on the evaluation of the suitability of the newly discovered enzymes for bioenergy (and biorefining) purposes is in progress in the host laboratory. Interestingly, some of the key enzymes discovered are mannanases, which is of particular interest since one of the dominating biomasses in Norway is (mannan-rich) spruce.
Importantly, the project has created massive spin-off:
• Phil Pope has used his top-competence in the field to contribute to a variety of other studies and scientific papers, both nationally and internationally.
• Phil Pope has contributed to education of researchers at the host department and has been important for setting up novel and fruitful collaborations within Europe.
• Phil Pope has attracted additional funding, including an ERC starting grant to continue and expand his work in the field.
The results of this project have been presented in several (invited) lectures, mainly for a scientific audience. In addition, the project has been presented as part of the host institution’s much wider effort in developing enzyme technology for biomass processing and biorefining. Such dissemination has been addressing scientists, biorefining and enzyme companies, and policy makers. During the project period, the host group has seen a large increase in collaborative biorefining projects with companies, including Borregaard (www.borregaard.com) and Cambi (www.cambi.com). Also, the group is heavily involved in a recently started FP7 collaborative project called Waste2Go (http://www.waste2go.eu/index).
The attached figure illustrates a key result of the project. The figure legend is as follows:
Figure 1. Overview over interdisciplinary technologies and experimental data used to describe microbial carbohydrate metabolism, with specific focus on a dominant, deeply-branched uncultured Bacteroidales lineage identified in the Svalbard reindeer microbiome. The picture the Svalbard reindeer rumen and the dicots plants and monocot grasses that constitute the reindeers diet (upper right panel). Specific measurements using carbohydrate microarrays with specificity for cell wall components detected high levels of xylan, beta-glucan, xyloglucan and mannan in a reindeer rumen sample (right panel). SRM-1 (left) is a Bacteroidales-affiliated bacterium detected using 16S rRNA pyrotag sequencing and estimated to constitute ~11% of the Svalbard rumen microbiome. The central picture shows a hypothetical model indicating predicted gene function and protein localization of a PUL-like system encoded within the reconstructed genome of the SRM-1 phylotype. In the second half of the project, the functionality of many of these proteins has been investigated in detail by biochemical analysis.
Microbial community analysis consistently shows the gut microbiome of herbivores (i.e. the rumen of a cow) is commonly dominated by a limited number of frequently observed bacterial phyla, namely the Firmicutes and the Bacteroidetes. Degradation of the most abundant plant polysaccharide (cellulose) within the rumen has for the greater part been attributed to the metabolic capabilities of species affiliated to the bacterial phyla Firmicutes and Fibrobacteres. While the Bacteroidetes represent a numerically dominating phylum in the rumen their functional role is poorly understood due to their difficulty to cultivate in pure culture and thus examine in the laboratory. Therefore, in this project we have used the latest advancements in DNA sequencing, computational biology and enzyme technologies to target the uncultured Bacteroidetes species that dominate these natural environments and perform the processes we seek to understand.
In this project objectives we have focused on the microbiome of the Svalbard rumen, In particular, we have studied a dominant bacterium identified in this microbiome, referred to as SRM-1 ,whose genome was partially reconstructed from metagenomic data. In the final phase of the project we have characterized a polysaccharide utilization locus (PUL) from the SRM-1 genome, which revealed novel enzymes and binding proteins involved in biomass turnover. The PUL in question is encoded within a 30 kb gene cluster that encodes 13 genes, including two consecutive pairs of SusC/D homologues and seven glycoside hydrolases that implied activity against β-(1,4)-glucans (3 x GH5, GH94) and mannans (GH26, GH130, GH2). We have performed biochemical characterization on the outer membrane GH26 and GH5 (n=2) enzymes. All purified enzymes were screened by a newly developed technique for high-throughput endo-GH-substrate determination using chromogenic hydrogel substrates (via collaboration with Prof. William Willats at University of Copenhagen). GH5 and GH26 enzymes demonstrated versatile activity against mixed-linkage glucans, xylan, xyloglucan, galactomannan and amorphous cellulose; all substrates that are highly accessible to the bacterium in nature.
As anticipated, this functional characterization of the enzymatic machinery of an uncultures bacterium has revealed novel enzymes for biomass processing. Furthermore, the results reveal unprecedented substrate versatility within a single PUL, in contrast to previous publications which have demonstrated narrow specificity regarding the target glycans of PULs i.e. one cognate PUL for each substrate.
All Objectives for the project have been reached. More specifically for the four key sub-goals:
1. Creating a metagenomic dataset of the Svalbard reindeer rumen microbiome using 454 sequencing. Done and published (Pope et al., 2012, PLoS One). Here the microbial community has been identified.
2. Annotation of the metagenome to reconstruct a genetic and metabolic blueprint of the rumen microbiome of Svalbard reindeer including the use of newly developed bioinformatic software. Done and published (Pope et al., 2012, PLoS One). Here, the genetic potential of the microbial community has been mapped.
3. Discovery of new enzymes and/or accessory proteins for biomass conversion using functional screening of the fosmid library and sequence-based screening of the metagenomic dataset. Done and published (Pope et al., 2012, PLoS One); follow up work to be published in a paper that has been submitted for publication (Naas, Pope et al., 2013). Here, potentially interesting enzymes for biomass conversion have been identified by combining bioinformatics and functional analyses.
4. Over-expression and characterization of potentially interesting enzymes and accessory proteins and evaluation of their potential in applied bioenergy production projects in the host laboratory. Done; some results published in 2012 (MacKenzie, Pope et al., 2012) and follow up work to be published in a paper that has been submitted for publication (Naas, Pope et al., 2013). Here, interesting enzymes have been overexpressed, purified and characterized in detail. Work on the evaluation of the suitability of the newly discovered enzymes for bioenergy (and biorefining) purposes is in progress in the host laboratory. Interestingly, some of the key enzymes discovered are mannanases, which is of particular interest since one of the dominating biomasses in Norway is (mannan-rich) spruce.
Importantly, the project has created massive spin-off:
• Phil Pope has used his top-competence in the field to contribute to a variety of other studies and scientific papers, both nationally and internationally.
• Phil Pope has contributed to education of researchers at the host department and has been important for setting up novel and fruitful collaborations within Europe.
• Phil Pope has attracted additional funding, including an ERC starting grant to continue and expand his work in the field.
The results of this project have been presented in several (invited) lectures, mainly for a scientific audience. In addition, the project has been presented as part of the host institution’s much wider effort in developing enzyme technology for biomass processing and biorefining. Such dissemination has been addressing scientists, biorefining and enzyme companies, and policy makers. During the project period, the host group has seen a large increase in collaborative biorefining projects with companies, including Borregaard (www.borregaard.com) and Cambi (www.cambi.com). Also, the group is heavily involved in a recently started FP7 collaborative project called Waste2Go (http://www.waste2go.eu/index).
The attached figure illustrates a key result of the project. The figure legend is as follows:
Figure 1. Overview over interdisciplinary technologies and experimental data used to describe microbial carbohydrate metabolism, with specific focus on a dominant, deeply-branched uncultured Bacteroidales lineage identified in the Svalbard reindeer microbiome. The picture the Svalbard reindeer rumen and the dicots plants and monocot grasses that constitute the reindeers diet (upper right panel). Specific measurements using carbohydrate microarrays with specificity for cell wall components detected high levels of xylan, beta-glucan, xyloglucan and mannan in a reindeer rumen sample (right panel). SRM-1 (left) is a Bacteroidales-affiliated bacterium detected using 16S rRNA pyrotag sequencing and estimated to constitute ~11% of the Svalbard rumen microbiome. The central picture shows a hypothetical model indicating predicted gene function and protein localization of a PUL-like system encoded within the reconstructed genome of the SRM-1 phylotype. In the second half of the project, the functionality of many of these proteins has been investigated in detail by biochemical analysis.