Skip to main content

AmylolEnzymes Captured by Targeted Metagenomics

Final Report Summary - AMYLOMICS (AmylolEnzymes Captured by Targeted Metagenomics)

Executive Summary:
Europe is traditionally strong in application of enzymes for industrial processing of bulk carbohydrate biomass. Biocatalytic tools and technologies are integrated parts of many diverse product pathways such as in production of sweeteners, fermentable sugars, thickeners and prebiotic oligosaccharides. Polysaccharide derivatives can be used in various fields; in cosmetics, in food and feed applications and in the chemical and pharmaceutical industries. The carbohydrate industry is of central importance for European industrial biotechnology, promoting research, creating revenues for SMEs and shaping research directions. There is a need for a range of robust enzymes for synthesis, fractionation and/or modification of different kinds of polysaccharides. The carbohydrate industry is receptive to test novel enzymes, new enzymatic processing methodologies and is constantly searching for new or more economical alternatives to existing tools and techniques, both for the production of novel higher value products and for optimization and improvement of older processes. In order to provide enzymes suitable to a greater range of industrial conditions the effort and scope of enzyme discovery needs to be expanded; from traditional easily cultivated organisms to include extremophilic robust organisms and to access the order of magnitudes greater diversity of environmental microbes. A major goal of AMYLOMICS was to develop efficient metagenomic methodology for accessing genetic resources in geothermal areas for enzymes of interest for the carbohydrate industry. Enrichments for fermentative thermophiles were carried out in a number sites by cultivation on different carbohydrate substrates and under different conditions. Species composition was monitored by molecular methods and the data evaluated with regard to species novelty and diversity. An enormous diversity was detected in these habitats, the majority of which is still to be characterized and exploited. More than 4500 of novel genes encoding carbohydrate active enzymes were identified in project using the bioinformatic tools developed. The quality of the sequenced data was improved by innovative approaches including, sequence capture and paired end sequencing. 303 complete genes (221 genes from metagenomes and 82 from genomes) encoding novel starch and other carbohydrate processing enzymes were extracted for further study. 204 genes were cloned and 157 expressed. 138 enzymes were screened for industrial relevant properties and more than fifty candidate enzymes were selected for application studies and detailed product analysis including high yield production in GRAS. Organisms. Commercial potential was evaluated. Patents were applied for 5 enzymes and 15 new enzymes entered the demonstration and marketing phase. A number of scientific papers were published. One particular enzyme was developed and licensed to an external SME for use in a recently developed biorefinery process. The underlying aim of AMYLOMICS is to help increase economic growth and sustainability of the sugar industry, to improve efficiency of bioconversion processes; to increase product diversity and to decrease waste. The enzymes obtained can be used in different processing platforms i.e. for production of fermentable sugars or specialized, novel or altered oligosaccharides. The starch industry is the most developed sector of the carbohydrate industry and European companies and plays a leading role on the global market. The AMYLOMICS project placed large emphasis on developing starch processing enzymes and more than 800 novel starch processing genes were retrieved in the project. The technology developed in the project enables rapid retrieval of novel genes encoding enzymes from extreme metagenomic resources that have a variety of uses. In the larger context of the enzyme technology, this will lead to a more environmentally friendly industry and improved economics for European industries and greater product diversification.
Project Context and Objectives:
The aim of the AMYLOMICS project is to develop new, robust enzymes for the starch and carbohydrate industries. The use of enzymes should enable the formation of new primary products, such as oligosaccharides of defined sizes, composition and degree of branching, new types of linkages, cyclic or more complex polysaccharides, increased digestive resistance, and secondary sugar derivatives such as substituted starches and polyols. Fundamental to the success of the project will be the development and use of an efficient platform technology for enzyme screening based on Targeted Metagenomics, which comprises: microbial enrichment techniques, massive parallel sequencing and microarray based sequence capture.

The starch and carbohydrate industries depend greatly on biocatalytic processes in modifying and fractionating polysaccharide and its derived components for the production of industrial and consumer products. There is a constant need for a range of robust enzymes that can be used for the synthesis, fractionation and/or modification of carbohydrates. It is particularly receptive to test novel enzymes and new enzymatic processing methodologies and it is constantly searching for new or more economical alternatives to the existing tools and techniques, both for the production of novel higher value products and for the optimization and improvement of older processes.

Currently, there is a need for new and efficient technologies for metagenome mining. Exploitation of high-throughput sequence data of metagenomes has been hampered by the high yield of short length DNA reads with limited informative content. AMYLOMICS addresses this problem by developing an efficient gene retrieval platform based on massive parallel sequencing and microarray sequence capture techniques. These methodologies combined in the project enable “genome walking” of complex environmental DNA. Consequently, this greatly facilitates the access to the largely unexplored wealth of genes in the environment and greatly improves upon current technologies. The Metagenomic mining platform developed in this project essentially sets the stage for the next level of metagenome exploitation as it efficiently circumvents the need for labour intensive cloning, construction of large insert gene-libraries and subsequent functional screening. It also evades the limitation of sequence based PCR methods.

Microbial enrichment and flow cytometry (FCM) cell sorting techniques will be developed and employed for obtaining metagenome libraries of optimal diversity as well as targeting them towards a particular metabolic activity. This ensures focused and more economical gene mining efforts. We have termed this methodological platform, which consists of microbial enrichments, high throughput sequencing and sequence capture: Targeted Metagenomics.

A special focus is on harvesting new genes/enzymes from thermophilic biospheres, especially from environments extreme in more than one aspect, i.e. thermo-acidophilic and thermo-alkalophilic biotopes. Due to the nature of carbohydrate polymers, there is usually an advantage of using moderate to high temperatures (60-100°C) in industrial catalytic processes for lowering the viscosity of polysaccharides and increasing enzymatic access. Furthermore, enzymes adapted to these environments are not only thermostable but also often resistant to proteases, organic solvents and extremes of pH. Consequently, they are tolerant to many of the adverse conditions encountered in industrial processes.

Of major importance for the partners in the project are a fast route to production and application of the enzymes retrieved. In order to substantially increase the expression success range of host/vector systems will be designed and/or optimized and used in the project for fast and efficient expression of the target genes/enzymes. Optimum codon composition for maximum expression will be analysed and genes accordingly re-synthesized prior to cloning.

As some important leads are expected to be suboptimal in some critical properties important for their application and economical returns, selected target enzymes will be improved by evolutionary techniques in order to meet criteria for successful industrial application.

Scientific and technological objectives
AMYLOMICS has the following specific and measurable scientific and technological objectives:

Scientific objectives:
• Better understanding of microbial ecosystems, their structure and how they evolve
• Better understanding of how a specific microbial biodiversity can be targeted in terms of metabolic activity by using specific enrichment techniques and flow cytometry
• Better perspective of microbial communities and their corresponding metagenomes e.g. in terms of targeted metabolic activities
• Better perspective of the metagenome gene reservoir and its related enzymatic reactions with reference to eventual exploitation
• Discovering novel genes and robust enzymes for specific applications and understanding their properties
• Discovering novel enzymatically modified products from starch or other carbohydrates
• In-depth studies on the structure-function relationships of enzymes from extremophiles (bioinformatics, protein structure determination)
• Acquisition of scientific data for processing carbohydrates using thermophilic enzymes of diverse specificities for product development from carbohydrates
• Acquisition and transfer of knowledge and expertise for the biotechnological exploitation of complex extremophilic microbial communities

Technological objectives:
• Development of strategies to analyze and manage metagenomes
• Developing and advancing sequence based platform technologies by combining massive parallel sequencing with microarray based sequence capture
• Application of this sequence based platform technology to identify new genes and pathways of biotechnological interest
• Development of novel enzymes for replacing and optimizing existing biocatalytic processes in the carbohydrate industry with a special focus on the processing of starch
• Developing new biocatalytic processes for novel high value products from polysaccharides
• Development of a viable production platform for biocatalysts from extremophiles enabling implementation of economical and sustainable biotechnological processes
• Development of novel biocatalysts to ensure optimal properties, stability and performance under industrial conditions
• Innovative process engineering and downstream-processing for the efficient production of enzymes and products
• Development of a pipeline for providing access to interesting valuable biocatalysts, which have not been selected for production process development for potential eventual industrial application
• New bioinformatic tools/software for analyzing and managing metagenomes

The objectives listed above are further reflected in the deliverables and milestones described in corresponding sections.
Project Results:
The project has been highly successful according to plan at the outset, and has reached nearly all of its scientific and technical goals and in fact substantially more in many of the tasks. Cooperation was ensured by active communication through emails and at the meetings resulting in substantial collaboration between partners. Integration in the project was exemplary and a number of enzymes were taken from one partner/WP to another - from discovery to full process development, demonstration and marketing. The project followed the timeline relatively accurately as visualized in the Gant chart of the DOW. Carbohydrate processing enzymes were targeted in the AMYLOMICS project. The enzymes should enable the formation of new products from various polysaccharides, such as oligosaccharides of defined sizes, composition and degree of branching, new types of linkages, cyclic or more complex polysaccharides. The timeframe, three years, is relatively short for taking enzymes from gene discovery to industrial process, and only a fraction of discovered enzymes could be investigated. Still, the project can be considered a highly successful in this aspect.
The project was divided into two interconnected pipelines:
Pipeline 1: Targeted Metagenomics - for bioprospecting
This part entailed technological innovation and development of a metagenomic bioprospecting platform and comprises: Microbial enrichment techniques, massive parallel shot gun and paired end sequencing and solution based sequence capture.
Pipeline2: Product and process development
This part comprises a complete pipeline for enzyme discovery and development of new carbohydrate and starch processing and modifying enzymes. It encompasses Pipeline 1 as a source of genes/enzymes.
Metagenome samples were provided in WP1; sequence information was extracted and analysed and a database constructed in WP2; the data was improved in WP3 and targeted biocatalysts made available for downstream analysis and development in WP4-9. Following below is a concise summary of the main results and progress made in the project. For clarity it is arranged according to the main tasks or work packages with the main significant results specifically emphasized.

Pipeline 1: Metagenomics (WP1, WP2 & WP3)

WP1 Metagenome libraries
Anaerobic and microaerophilic enrichments were carried out by MAT from more than 10 habitats. Several polysaccharide substrates and culture conditions were deployed and diversity was monitored by analysing 16S rRNA clone libraries from the enrichments. Difficulties of obtaining reproducible results were experienced at the start and approaches and procedures were developed for obtaining high enough amounts of high quality DNA as work proceeded. A great number of metagenome samples were produced and seven were selected for pyrosequencing in WP2 based on
1. Novelty of species detected on the basis of the 16S rRNA diversity in clone libraries
2. Metabolic types of species – the likelihood of containing the target enzymes
3. Having the option of adjusting diversity and evenness of the final metagenomic libraries in subsequent bioinformatic in composite metagenome sequence libraries. This involves combining information from samples from the same site enriched for under different conditions at later steps (WP2 and 3).

Diversity analysis
Enriched metagenomes: Diversity of enrichment samples was monitored by ABI sequencing of corresponding 16S rRNA clone libraries. A total of 76 of samples enriched under different condition were analysed. Massive data was generated on species composition and species abundance in relation to environmental variables. The data is accessible and managed in a special phylogeographic database developed in a parallel project by MAT

Direct metagenomes: Biomass samples were also taken for diversity analysis in 32 circumneutral and alkaline hot springs from different parts of Iceland (MAT) varying in pH, temperature and chemical composition. Three samples were taken from each hot spring; microbial mat, water and sediment. Tagged 16S rRNA amplicon pyro-sequencing data was obtained using two pairs of universal primers targeting two different regions in the 16S rRNA sequence. More than 500 000 16S rRNA sequences were analysed from 32 circumneutral hot springs and distribution of taxa was related to environmental factors. An enormous diversity was detected in these habitats, the majority of which is still to be characterized and exploited

Normalization with flow cytometry: Preliminary investigations showed that flow cytometry had little cell resolving power on the basis of bacterial size (and/or shape) and it was concluded that this task would not add significantly to eventual sequence diversity per effort and cost. However, cytometry was deemed useful for comparing microbial density between in samples from different geothermal habitats.

Significant results
1. Selective enrichment techniques were developed and applied for obtaining metagenome libraries of thermophiles of optimal diversity for sequence based bioprospecting.
2. A number of geothermal areas of different characteristics in Iceland were explored and a total of 177 enrichment cultures were analysed.
3. Ten different sites were selected for metagenomic bioprospecting in situ. Microbial traps and selective substrates were deployed for enrichment in situ using a special porous matrix containing diverse complex carbohydrates. A rough estimate of the species composition in 95 cultures was performed by analysing 16S rRNA gene sequences from 12-30 clones from each of the different cultures. In general, the diversity within enrichment cultures from enrichment in situ trap samples was lower than in cultures from direct samples, most probably explained by pre-selection by the substrate in the trap. Selection of enrichment cultures for subsequent metagenomic sequencing was based on the above estimates of species composition followed by inspection of species diversity and taxa ratios within the cultures as well as the quality and quantity of the DNA extracted.
4. Lab enrichments of direct samples as well as trap samples were performed using anaerobic and microaerophilic culture conditions. These were enriched - using diverse recalcitrant carbohydrates or special starch derivatives in the medium as well as variable culture conditions in temperatures, pH and media - to enhance growth of diverse microbial species. Lower diversity was generally observed I these samples compared with in situ enrichments.
5. Enrichments enhances growth of microorganisms with the desired characteristics. They can also eliminate the diversity of microorganisms obtained from the original sample, as the enrichment substrate is a selective force. Thus, both novel species and diversity can be eliminated.
6. Another, conclusion from the WP is that samples having low diversity, but high novelty for the bioprospecting effort are important for obtaining genes/enzyme and information that otherwise would be lost. Therefore, it is important not to exclude metagenomes of low diversity, but high novelty and having relatives previously shown to harbour a variety of carbohydrate processing enzymes (e.g. Thermotogae). Sequencing costs of such samples are instead less and sequence coverage individual species in the sample more extensive
7. An important sample collection and an associated metagenomic database was created on microbial distribution in geothermal biotopes in Iceland. It was based on retrieval and analysis of 500 000 16s rRNA genes directly retrieved from 32 geothermal biotopes varying in physicochemical characteristics and 177 anaerobic and microaerophilic enrichments. The massive information generated will serve as the basis for future targeted biotechnological exploitation of thermophiles.

WP2 Metagenomics
FLX sequencing of metagenomes: Seven metagenome libraries were selected from WP1 for bioprospecting based on novelty of species detected on the basis of the 16S rRNA analysis of clone libraries. DNA was isolated and primary shot gun sequencing was carried out for all the samples. The libraries varied, some showed little diversity, but high species novelty whereas others showed high to moderate diversity, but little evenness. Two composite metagenomes (CM1 and CM2) were produced (to adjust diversity and evenness) from samples taken at different locations at the same site which had some overlap in species composition, but different fractional representation. These composite metagenomes were designed specifically for the genome walking approach in WP3 and a special care was taken to obtain high amounts of high quality DNA from the three constituent samples. Primary sequencing of each separate metagenome was carried out and the coverage improved by re-sequencing as needed. This allowed separate and combined assembly and different bioinformatic evaluation and comparison and modulation of ensuing processing steps, sequence capture and paired end sequencing in WP3.

Software development, data management and annotation: A custom metagenome database for result storing was developed by SMG as well as a web interface for data accession. Format for annotation results was established. Two approaches were used to annotate the sequences with KEGG functional categories: Conservative assembly where the assembly took place before annotation in conventional way and Functional Assembly, a pipelined specially designed by SMG were raw reads were trimmed, filtered for low quality regions and classified according to phylogeny. Subsequently, codon regions were identified and then searched with HMM profiles corresponding to KO system from KEGG. All reads belonging to same KO number were grouped and gene assembly was carried out on them. The latter method yielded significantly higher number of annotation hits compared to conventional assembly and annotation.The metagenomes, were annotated for gene retrieval and expression in WP3 and the first enzymes selected for primary characterization in WP4 and WP5.

The metagenome analysis platform, was released as stable version in October 2013, implementing as a pipeline of tools for managing, analysis and reporting of (meta) genomic data running on a server for this purpose and found at URL

WP3: Sequence Capture and gene retrieval
Metagenome improvement: The primary objective of WP3 was to develop strategies for improving metagenome libraries. The improvement included extension of sequence reads (singletons) and assemblages of longer continuous sequences (contigs) obtained in WP2. A new protocol for sequence capture, adapted for metagenome sequencing, was developed. Additionally, one metagenome was improved further by paired end sequencing. By alternating of shot gun sequencing, sequence capture and paired end sequencing, sequence reads are assembled into longer contigs and scaffolds of linked contigs linked in a specific order and orientation.
Three metagenome libraries were selected for improvement, the two composite metagenomes (CM1 and CM2) obtained by enrichment, and for comparison one metagenome (YL1) obtained by direct DNA extraction from a marine geothermal habitat. All three were re-sequenced for higher sequence coverage and sequence capture carried out for all. Improvement by sequence capture was achieved in all samples but the extent depends on the complexity of the metagenome. The effect was less pronounced for the enriched metagenomes (CM1 and CM2) than for the direct metagenome (YL1). The enriched metagenomes are composed of few major representatives, which were already relatively well covered from the primary shot gun sequencing, whereas the diversity was by far higher in the unenriched metagenome and consequently less well covered. The singletons in the enriched libraries may be derived from the rare species which are not well represented by the metagenome. Further improvement by liking contigs was then obtained on one metagenome by additional paired end sequencing.

Sequence retrieval: By combined effort of MAT and SMG in WP2 and WP3, carbohydrate processing genes were specifically targeted and a total of more than 4500 genes were detected. Of those approximately 800 encoded starch processing and modifying enzymes. The presence of genes encoding carbohydrate active enzymes was five times greater in the enriched metagenomes despite much greater diversity in the direct metagenomes. The number of complete genes was substantially increased (approx. 37%) by sequence capture.
221 novel carbohydrate active genes from metagenomes C1 and C2 and additionally 82 from genomes were selected for bioinformatic analysis and retrieval and preliminary characterization in WP4 and subsequent studies of the gene products in downstream workpackages. Selected complete genes with used affinity tags were cloned into E. coli vectors developed by USTUTT allowing positive regulation of gene expression and fast purification. A total of 201 genes were cloned and 157 expressed.

Pipeline 2: Product development pipelines (WP4, WP5, WP6, WP7, WP8 and WP9)

WP4. Enzyme characterization
138 starch-converting and carbohydrate modifying enzymes were subjected to primary screening for confirmation of activity predicted by bioinformatic analysis using appropriate assays. This included spectrophotometric assays, e.g. conventional DNS, thin layer chromatograph (TLC) and high pressure liquid chromatography (HPLC). Temperature and pH activity range was also broadly determined. 50 enzymes were the selected for secondary characterization involving more thorough biochemical research where parameters such as specific activity, temperature and pH optima, substrate specificities and end-product formation, were studied. A panel of carbohydrate substrates was used for primary analysis of substrate range and product formation, such as polysaccharides of different types and oligosaccharides. Factors determining the enzyme reaction specificities, such as the hydrolysis/transglycosylation ratio were examined, when appropriate. Structural modelling was done on selected proteins and genetic variants were obtained by site directed mutagenesis and analysed. Selected enzymes for in depth analysis included starch processing enzymes such as alpha- and beta amylases, pullulanases, glucoamylases; polysaccharide lyases such as alginate and chondroitin lyases; chitinases; beta-glucanases, xylanases and both alpha- and beta transglycosidases. The biochemical results along with results of detailed product analysis in WP5 were used for evaluating scientific value and industrial potential of the enzymes and served as the basis for the selection of enzymes for downstream product development in WP7-9 or dissemination in WP10.

Significant results
1. A great variety of starch and carbohydrate processing enzymes were made available for study and development.
2. An active recombinant chondroitin lyase was fully characterized (manuscript submitted)
3. Three chitinases with different end product patterns were characterized and selected for downstream development and production. A commercial interest for a specific applications were identified during the project and the enzymes fulfilled set criteria. One of the chitinases was subsequently licensed to an external company.
4. Four different alginate lyases were characterized and selected for downstream analysis and development. These are first thermophilic alginate lyases to be studied in depth. Crystal was generated for one of them for structural determination. (Patent applications pending and manuscript is in preparation).
5. Six novel thermostable GH3 beta-glucosidases were modelled and studied in depth They appear to be structurally quite different and they showed a wide range of substrate specificities and optimum temperature up to 80°C, (Manuscripts in preparation)
6. A thermostable cyclodextrin transferase was studied in depth and compared to two benchmark enzymes Amano and Toruzyme, respectively for their ability to make primary coupling product by using cyclodextrins as a donor and alkyl glucoside as an acceptor. When using alpha and beta CD as donors. Interestingly, the enzyme could utilize gamma CD as a donor to make coupling product which was not seen in the two commercial enzymes (manuscript in preparation). Initial trial has been done to produce flavonoid glycoside using the enzyme. (Manuscript submitted to Glycobiology. (Ara et al).
7. A transglycosylating GH57 enzyme was characterized that is capable of making cyclic amylose products and can be used in alkylglycoside production from starch (not CDs) and yields products in a wide range (Manuscript submitted to Appl Microbiol & Biotechnol).
8. 3D-structure and substrate docking of a thermostable GH1 glucosdidase completed, explaining the specificity difference between 3 and 4´-glycosylated flavonoids. (Manuscript in preparation, Kulkarni et al)

WP5 Product analysis
The objective of this WP was to gain information on the detailed structure of oligo- and polysaccharide products of the selected enzymes isolated in this project comprising a number of activities such as branching enzymes, beta-transglucosidases, glucansucrase, alginate lyse, and oligosaccharide producing amylases, chitinases as well as transfer activity of glucosidases and alpha-galactosidase and genetic variants obtained by site directed mutagenesis This involved detailed structural analyses of products, including chemical, enzymatic, chromatographic, mass spectrometric and NMR spectroscopic approaches. The composition/linkages or the type(s) of glycosidic bonds and the degree of branching and branching points were determined when appropriate.

Significant results
The following enzymes were studied in depth and shown to have properties high potential for indnsustral applications and/or of appreciable scientific value.
1. Beta transglucosidase GH17: Analysis of the glucosyl transfer products of a beta transglucosidase enzyme with laminari-oligosaccharides and laminarin (trans--glucosylation). The enzyme was shown to have true branching activity and efficient for making branched beta-glucan oligosaccharides. Patent pending (PV_CT) and; publication in preparation.
2. Alginate lyases: Detailed product formation and comparison of four novel thermostable alginate lyases on alginic acid, G-blocks, and M-blocks. The enzymes were shown to form very different degradation products. These enzyme are suitable for making alginate oligo-saccharides of different kinds or can be used to degrade alginate to potentially mono-uronates sugars for ethanol fermentation or for utilization in as substrates for synthesis of platform chemicals. These are the only thermostable alginate lyases described to date and are highly suitable for diverse biorefinery applications. (Patent pending and publication in preparation).
3. Glucansucrase: Study performed on product formation of recombinant glucansucrase GTFA from Lactobacillus reuteri strain sucrose/malto-oligosaccharides substrate (trans--glucosylation). Can be used for directed synthesis of branched mixed linkage alpha-glucan. Of high interest in the food industry. (Two papers: in Published in Journal of Biotechnology and in Glycobiology )
4. Alpha galactosidase (GH36): Structural studies of products obtained by incubation of an -galactosidase and genetic variants obtained by site directed mutagenesis, with a mixture of melibiose and sucrose (trans--galactosylation). Greater understanding of transfer vs. hydrolysis activity in this family of enzymes. Application potential in the sugar refining industry. (Publicaton in preparation).
5. Starch processing enzyme of family GH13. Analysis of the glucosyl transfer products, obtained by incubation of the GH13 enzymes Amo9 and Amo30, out of the stock of partner 1 (MAT) and in collaboration with partner 6 (USTUTT), with pnp-malto-oligosaccharides (trans--glucosylation). Highly interesting results on branching activity in these enzymes. Capable of making branched cyclodextrins. (Publication in preparation).
6. Beta-Glucosidases (GH1 and GH3). Structural studies of products obtained using beta-glucosidases and genetic variants obtained by site directed mutagenesis, with different flavonoid acceptors (trans-beta-glucosylation). Important understanding was obtained on structure/function relationship regarding these enzymes in utilizing these enzymes for glucosylation. Application potential in synthesis of bioactive glycosides. (Published in Applied and Environmental Microbiology)
7. Chitinases: Analysis and comparison of degradation products of three thermostable chitinases. Different degree of hydrolysis was observed making them suitable for different applications. One enzyme was licenced to an external company for use in biorefinery application

WP6 Enzyme evolution
The objective was to optimize specific characteristics of selected enzymes in order to increase their performances and thereby the industrial application potentials. Enzymes of a particular interest with regard to industrial applications, yet suboptimal due to deficiency of features such as an appropriate efficiency, stability and solubility, were candidates for the work. Both site directed random mutagenesis were applied. Random mutagenesis and selection for improved variants proved unsuccessful despite intensive effort. The aim was to increase the degree of a branching enzyme (GH13) and generate acidophilic variants of two enzymes, a pullulanase and a glucoamylase. Assays were cumbersome or insensitive. Two glucosidases belonging to families GH1 and GH3 enzymes, and an alpha galactosidase (GH36) were judged as having transferases activity of industrial interest. Their characterization was therefore expanded further in order to elucidate/generate activity on more industrially interesting substrates and/or for obtaining alternative products. This involved elucidating more clearly the 3D-structure of the enzymes in an attempt to determine molecular determinants of specificity and the potential influence of specific residues. Site-directed mutations were introduced, in order to further elucidate the possibilities or assess limitations for substrates/ or reactions based on their structural properties. In this deliverable, only mutants that were extensively characterized are described. Valuable information was gained regarding these aspects. (Manuscripts in preparation (D4.4))

WP7. Expression optimization
The objective of WP 7 was to develop an expression system for producing selected novel starch and carbohydrate modifying enzymes. The lead beneficiary was partner 6 (USTUTT) which has a long experience in developing expression system. The “expression optimization” work in WP7 encompassed the following objectives:
• Improving expression of genes which turned out to be poorly expressed following primary cloning into the standard expression vector and hosts
• Expression of genes which were not expressed following primary cloning into standard expression vector, judged by lack of activity and recombinant proteins on SDS
• High yield expression of genes in appropriate hosts, i.e. for production of enzymes intended for applications in the food and pharmaceutical industries
To achieve the goals a set of molecular tools were applied, e.g. different fusion proteins for improving gene expression; increasing solubility and reducing inclusion body formation; facilitating purification of the produced enzymes by affinity chromatography; also chaperon proteins to improve folding and solubility; rare tRNA genes to improve translation, cloned into the expression vectors to avoid instability associated with two plasmids in one cell.
Although E. coli was the main production host other hosts were studied and developed as well. Special emphasis was on GRAS organism (Generally Regarded As Safe) for production of enzymes intended for applications in the food and pharmaceutical industries. In this context, an extensive work was done to develop expression vectors for Bacillus subtilis and Corynebacterium glutamicum.

A total of 32 Amylomics genes were subjected to expression optimization in WP7. The expression of almost all of the genes was improved by applying specific combinations of the molecular tools in different vectors and hosts.

Significant results
1. Improved expression of the majority of the genes fed into WP7, enabling production of interesting proteins for further studies in downstream workpackages
2. Great variety of E. coli rhamnose inducible expression vectors which allow highly regulated expression, especially suitable for high scale fermentation and production of proteins known to bring a metabolic load to the production host.
3. Successful strategy in avoiding inclusion body formation by fusing the target protein to E. coli maltose binding protein, the NusA protein or the S-loop at the gene level. (Published in Biocatalysis and Biotransformation)
4. Expression systems developed for Bacillus subtilis based on promoters and regulators of the genes for mannitol and glucitol degradation
5. Two expression systems developed for gene expression in C. glutamicum, one based on the classical E. coli tac/lacIq regulatory system and a completely new one using the regulatory system for ferulic and vanillic acid degradation of C. glutamicum.
6. Four different thermostable alginate lyases expressed in high yields in E. coli and GRAS organisms.
7. Number of starch processing enzymes expressed in high yields for giotransformation testing in Wp9 and production for the internet store of PROK.
8. Production of chitinase enzyme in GRAS enabling immediate industrial exploitation
9. Despite high expression, no activity could be detected in some of the protein, e.g. most of the GH family 57, which were specially targeted in the Amylomics, were not active on the starch substrates tested

WP8 Fermentation
This work package focused on the production of selected enzymes by the use of the newly developed host vector system in WP7. The production strains that were established in WP7 and exhibited good expression, were grown in fermentations of different scales. The workpackage was divided into two tasks: Fermentation optimization for establishing conditions for large scale production of enzymes intended for bioconversion processes and batch production for analysis within Enzyme were produced in 1-10 L batch or fed-batch fermentations, yielding enough material for the development of bioprocesses in WP9, and to secure appropriate quantities for dissemination in the internet store of PROK. Expression systems included E. coli and various GRAS organisms from USTUTT developed in WP7. The work included expression optimization at the genetic level along with media selection, inducer optimization and development of feeding strategies. Initial experiments were done in low-scale fermentation experiments for analysing optimization at the genetic level. Work in this WP was highly successful and is not clearly separated from WP4 and WP7, a number of enzymes were produced for the diverse purposes of the project. No significant problems were encountered. More detailed description is given in corresponding deliverable reports and in the progress and periodic reports.

WP9 Bioprocess development
This work involved establishing and investigating the potential of selected enzymes in bioconversion processes. Initially enzymes were tested on a small scale with substrates and purified biocatalysts in different proportions. Parallel reactions were carried out with biocatalysts present as whole cell suspensions or crude cell protein extracts as appropriate for specific enzymes. Optimal reaction parameters, including biocatalyst formulation and loading, temperature, pH buffer requirements and maximum substrate loading were determined at this scale. Performance was then evaluated under high substrate loading, using product yields and biocatalyst processivity as the key parameter.

Starch converting enzymes
Of special interest were more thermoacidophilic counterparts to enzymes already used in industry but also novel activities were expected to be found where new application possibilities would be explored. Roquette expressed a special interest in more efficient and more acidophilic debranching enzymes and a number of pullulanses and neopullanaseses were delivered to ROQ for bioconversion assessment. They were on par with enzymes already used in Roquette’s process, but not better and therefore did not warrant further development.

Other carbohydrate active enzymes were also tested in semi-industrial trials, including three chitinases for production of chitin oligosaccharides of discreet sizes, a glucansucrase, GFTA, for production of branched glucan (reuteran) and an alginate lyase for processing of macroalgae for extraction of flavour substances.

Chitinase for production of chito-oligosaccharides
The chitooligosaccharides produced by three different chitinase were differed in size range. One of those was of particular interest to an external company and was analysed on a Pilot scale. The enzyme yielded 5 kg oligosaccharide powder from 40 kg chitosan flakes. Analysis of the size distribution of the oligosaccharides revealed favourable size distribution than the previously used enzyme mixture (benchmark). The enzyme expressed in high yield in a particular GRAS production host was subsequently licensed to this company.

Chitinase for production of chitobiose
Production of chitobiose was relatively efficient using two other chitinases compared to enzymes in previous attempts in the project. The yield of products was estimated to be close to 50%. Taking the enzyme activity and stability into account and the conversion efficiency it is estimated that by optimizing and increasing production by a factor of 10, e.g. in a Bacillus secretion system, the amount of enzyme from a 40 litre fermentor is sufficient for degradation of 1 ton of chitosan substrate.

Alginate lyase as an aid in extraction of flavour components from Brown algae.
Prior processing of seaweed biomass with one of the four thermophilic alginate lyases obtained in the project significantly improved subsequent peptide and amino acid extractions following treatment with protease.

WP10 Demonstration and exploitation WP12 Dissemination
PROK worked with partners on characterization of enzymes, identifying commercial potential. In the first reporting period , six Amylomics enzymes were introduced into the Prokazyme webstore and have been promoted & marketed by PROK in the internet store: Glucan elongation enzyme - elongates linear α-glucan chains in starch and amylose; (Amo31); Naringinase - cleaves the aglycone off L(+)-rhamnose glycosides – i.e. naringin; (Amo65); Hesperidinase - cleaves the aglycone off L(+)-rhamnose glycosides – i.e. hesperidin – (Amo66); α-Galactosidase Hydrolyzes α-1.6 linked α-galactose from oligosaccharides such as raffinose, melibiose, and stachyose or polymers such as galactomannans (Agal104); Highly thermostable β-Galactosidase from Thermus brockianus - hydrolyses terminal, non-reducing α-D-galactose residues. (Bgal112); Fucosidase, cleaves various α-fucosylated oligosaccharides (Fuc123).Product sheets and details can be found on
In the second reporting period seven additional enzymes were introduced in the webstore. One additional enzyme developed in the Amylomics project was licensed for in-house production and use by an external company. Therefore a total of 14 new enzymes, resulting from the Amylomics project have been commercially demonstrated and thereof 13 were introduced in the Prokazyme webstore. After introduction of the new Amylomics enzymes into the Prokazyme webpage, two contacting campaign have been carried out through direct contacts with email and to previous Prokazyme customer base. The speciality companies that had been identified previously (D12.3) were contacted by email. The overall response was about 15% and 17 active new customers who have ordered one or more enzyme have resulted so far from this effort. These include the following: Tate & Lyle, Roche, Amyris, MQD, Biothera, Univision, Jamjoom, Univ. Chile, Covance lab, Arachem, Univ. California, Sheff Inst Cancer, Consulab USA, Pfizer, Naturolendo, Stratos. In reporting period 2, two more enzymes from the Amylomics project were introduced in the Prokazyme webstore under the general trade name of ThermoActiveTM enzymes: Glucantransferase (Gtfa163) and a thermostable β-glucosidase (Bgl162). Also two more enzymes are ready for launch, i.e. α-glucansucrase and a thermostable xylanase. Also five more enzymesd are now available upon inquiry or under MTA´s, since they are also in patenting process, i.e β-Branching Glycosyltransferase and four thermostable alginate lyases. In addition to the above contacts, which are composed of both speciality enzymes/reagents providers and users as well as some companies engaged in providing technical enzymes for various industrial sectors have been contacted and several parties have responded and the following have become active customers in addition to those listed previously. Lexogen GmbH, International RINP Inc, Brunschwig Chemie B.V Jamjoom Pharmaceuticals Co Ltd, Kim & Friends Inc. Fraunhofer IME, Sensient Flavors Inc., Chrom Matrix Inc, Illumina Inc, Theranos Inc Sistemas Genómicos, Eurofins Food Testing Netherlands B.V Ski MSKCC NY, Chalmers and TH Mittelhessen.

PROK mediated contact and negotiated successfully with a company that specializes in production of oligosaccharides for the medical industry. Enzymes developed and produced by MAT and USTUTT were tested on proprietary substrates. Subsequently PROK mediated an licence agreement from RTD partner Univ. Stuttgart. This agreement was executed on 15. February 2014. This will further the dissemination and market visibility of the AMYLOMICS project.
One of the roles of PROK was to negotiate and assist partners in marketing of discovered enzymes. PROK has assisted Matis and University of Groningen in writing patent applications and reaching agreement IP rights and costs of patent application for a novel β-glucantransferase and alginate lyases.
The agreements were as follows:
1. Among Matis and Rijksuniversiteit Groningen on rights concerning a US provisional patent application on Glt20 that was filed on Date: March 26. 2013. No: US 61/805,360. The title: „Glucan-branching Enzymes and Their Methods of Use“. Applicants: Matis ohf and Rijksuniversiteit Groningen. Inventors: Gudmundur O. Hreggvidsson, Jon O. Jonsson, Olafur H. Fridjonsson, Justyna M. Dobruchowska, and Johannis P. Kamerling.

2. The same patent was then filed again as PCT patent application on March 26. 2014 based on the priority date obtained by the US provisional application above.

3. Among Matis, Universität Stuttgart and Rijksuniversiteit Groningen on rights concerning a US provisional patent application on Thermostable Alginases. According to this agreement, a US provisional patent application, US 61/926,009 was filed on January 10. 2014. The title: „THERMOSTABLE ALGINATE DEGRADING ENZYMES AND THEIR METHODS OF USE “. Applicants: Matis ohf, Universität Stuttgart and Rijksuniversiteit Groningen. Inventors: Gudmundur O. Hreggvidsson, Jon O. Jonsson, Bryndis Bjornsdottir, Olafur H. Fridjonsson, Josef Altenbuchner, Hilde Watzlawick, Justyna M. Dobruchowska, and Johannis P. Kamerling.

4. Prokazyme also mediated an agreement that was signed and executed on February 15. 2014, between Universität Stuttgart, as an Amylomics partner, and the external company. The agreement relates to the licensing of a certain enzyme developed in the AMYLOMICS project to the company for in-house production, manufacturing and use of the enzyme. The company owned patents covering the use of such enzymes in their process but in the Amylomics project the enzyme was made available in a GRAS host among other expression improvements that were useful for the company.

5. In addition to the above Prokazyme made and executed agreements with three Amylomics partners, i.e. Rijksuniversiteit Groningen, Lund University and Universität Stuttgart for marketing of certain AMYLOMICS enzymes developed by those partners in the Prokazyme webstore. Press releases were already issued for two of those agreements, on 24. Sept. 2013 and 13. Feb. 2014, respectively.

Scientific publications
Nine papers were published and two more have been submitted. Number of manuscripts are also in preparation.

List of publications
1. Hans Leemhuis, Tjaard Pijning, Justyna M. Dobruchowska, Sander S. van Leeuwen, Slavko Kralj, Bauke W. Dijkstra, Lubbert Dijkhuizen. 2013. Glucansucrases: Three-dimensional structures, reactions, mechanism, α-glucan analysis and their implications in biotechnology and food applications. Journal of Biotechnology. Vol. 163/Issue 2: 250-272

2. P. Lundemo, P. Adlercreutz , E. N. Karlsson. 2013. Improved Transferase/Hydrolase Ratio through Rational Design of a Family 1 -Glucosidase from Thermotoga neapolitana. Applied and Environmental Microbiology. Vol. 79/Issue 11:3400-3405

3. Jurica Zucko , Olafur H Fridjonsson , Solveig K Petursdottir , Ranko Gacesa , Janko Diminic , Paul F Long , John Cullum , Daslav Hranueli , Gudmundur O Hreggvidsson , Antonio Starcevic. 2013. Browsing metagenomes for novel starch and carbohydrate industry enzymes—AMYLOMICS case study. Current Opinion in Biotechnology, Vol. 24: S21

4. J. M. Dobruchowska , X. Meng , H. Leemhuis , G. J. Gerwig , L. Dijkhuizen , J. P. Kamerling. 2013. Gluco-oligomers initially formed by the reuteransucrase enzyme of Lactobacillus reuteri 121 incubated with sucrose and malto-oligosaccharides. Glycobiology. Vol. 23/Issue 9:1084-1096

5. Anna Ekman, Monica Campos , Sofia Lindahl , Michelle Co , Pål Börjesson , Eva Nordberg Karlsson , Charlotta Turner. 2013. Bioresource utilisation by sustainable technologies in new value-added biorefinery concepts – two case studies from food and forest industry. Journal of Cleaner Production. Vol. 57: 46-58

6. Lei Wang, Hildegard Watzlawick , Olafur Fridjonsson , Gudmundur Hreggvidsson , Josef Altenbuchner. 2013. Improved soluble expression of the gene encoding amylolytic enzyme Amo45 by fusion with the mobile-loop-region of co-chaperonin GroES. Biocatalysis and Biotransformation. Vol. 31/Issue 6:335-342

7. Javier A. Linares-Pasten, Maria Andersson and Eva N. Karlsson. 2014. Thermostable Glycoside Hydrolases in Biorefinery Technologies. Volume: 3. Issue Number: 1:26-42
DOI: 10.2174/22115501113026660041 (could not be uploaded to portal)

Submitted papers
1. C J. Paul, H Leemhuis, J M. Dobruchowska, C Grey, S S. van Leeuwen, L Dijkhuizen, and E Nordberg Karlsson Characterization of the 4-α-glucanotransferase from a thermophilic Archaeon: production of cyclic amylose and alkyl glycosides. (submitted to Glycobiology Appl Microbiol & Biotechnol)

2. Ara KZG, Lundemo P, Adlercreutz P, Fridjonsson O, Hreggvidsson GO, Nordberg Karlsson E. A novel CGTase from Thermoanaerobacterium thermosulphurigenes. (submitted to Glycobiology)

3. Ranko Gacesa, Jurica Zucko, Solveig K. Petursdottir, Olafur H. Fridjonsson, Janko Diminic, Paul F. Long, John Cullum, Daslav Hranueli, Gudmundur O. Hreggvidsson and Antonio Starcevic. MEGGASENSE - the metagenome/genome annotated sequence natural language search engine: a platform for the construction of sequence data warehouses. Submitted to Genome Biology

Book chapters
1. Gashaw Mamo, Reza Faryar, Eva Nordberg Karlsson. Microbial Glycoside Hydrolases for Biomass Utilization in Biofuels Applications. 2013. In: Biofuel Technologies. Springer Berlin Heidelberg, Berlin, Heidelberg. p171.

2. Kazi Zubaida, Gulshan Ara, Samiullah Khan, Tejas S. Kulkarni, Tania Pozzo , Eva Nordberg Karlsson. 2013. Glycoside Hydrolases for Extraction and Modification of Polyphenolic Antioxidant. In: Advances in Enzyme Biotechnology. Springer India. New Dehli.

In both Pipelines of the project, the Metagenomics and Product development pipelines, MAT (coordinator) has played a central role in providing genes and enzymes and metagenome sequence data to relevant partners. Cooperation on specific tasks was ensured by active communication through emails and at the meetings resulting in substantial collaboration between partners. Collaboration and integration of work in the project was exemplary and a number of enzymes were taken from one partner/WP to another - from discovery to full process development, demonstration and marketing. The project followed the timeline relatively accurately as visualized in the Gant chart of the DOW. The timeframe, three years, is relatively short for taking enzymes from gene discovery to industrial process, and only a fraction of discovered enzymes were taken through the pipeline. Still, the project can be considered a highly successful in this aspect.
Accomplishments of the project can be evaluated according various measures
1. Progress indicators set out in the corresponding the table from the DOW (Table 3.2.2_2. It shows that that set aims as regards the bioprospecting features were successfully met.
2. Publications: 9 published papers and book chapters
3. Patent application: 2 applications for 5 enzymes, a beta-branching enzyme and 4 different alginate lyases for potential use in marine biorefineries utilizing macroalgal polysaccharides
4. Process development: A total of 5 enzymes (reported in D9.2 and .D9.3)
a. One Enzyme (chitinase) (in GRAS producing production host) evaluated and licenced to external company for immediate application in a biorefinery for production chitooligosaccharides of discreet sizes.
b. Four enzymes evaluated and tested with success in an industrial context: Two chitinases (Amo161 and Amo162) suitable for production of chitobiose, one alginate lyase (Amo69) applicable for use in extraction of flavour components from Brown algae (tested in the parallel SME EU-project TASTE). Glucan sucrose applicable for producing mixed linkage branched beta glucans.
5. Demonstration and marketing: A total of 15 new enzymes. Market entry has been started with creation of trademarks and public product sheets and they have entered the demonstration and marketing phase through the internet store of PROK.
Potential Impact:
AMYLOMICs placed a strong emphasis on demonstration and dissemination activities, for increasing visibility of potentially commercially exploitable results from the project to end-users, building on past experience, established channels, outside industrial contacts and proven strategies in prior projects. The partners used their extensive connections and venues in their respective research fields to actively introduce enzymes and potential processes developed in the project to potential users and other biotechnology agents. One partner, Prokazyme, was specifically designated to oversee demonstration and dissemination activities. Prokazyme mediated negotiations for commercial exploitations and IP rights of selected intellectual property and industry property produced within the project and acted as the marketing body to outside industrial interest by providing an internet site where enzymes of potential interest are offered and can be bought sufficient amounts for testing.
The project was introduced on the homepages of partners and an Amylomics homepage ( was made and kept updated with results from the project. One of the major goal of AMYLOMICS was to extend visibility of results and products beyond the lifetime of the project, a problem common in EU-projects. Bilateral agreements were executed between university partners within the project on marketing of their enzymes through the internet store of PROKAZYME. Promotional material was made as well as public actions taken, such as press releases and introductions to potential customers. After introduction of the new Amylomics enzymes into the Prokazyme webpage, and campaigns were carried out through direct contacts with email and to previous Prokazyme customer base. Speciality companies that had been identified previously were contacted by email. From this effort and a number of new customers were acquired.

The project was introduced in conferences specially aimed at industry such as Carex Forum: ‘Biotechnological exploitation of extremophiles’ held at Novozyme (Copenhagen, 2011), the MIcroB3 – MetaExplore: Harvesting Environmental Genomes for Development of Biocatalysts (Groningen, 2013) and Innovative SMEs under Horizon 2020: Towards market’s breakthrough on bio-based industries (Brussels 2014). The project was also introduced in a number of local press releases, in scientific fairs and local conferences aimed at the public, in Sweden, Iceland and Croatia. The project was introduced on Bio-Net youtube (

15 new enzymes entered the demonstration and marketing phase during the project. Market entries were started with creation of trademarks and public product sheets. A number of enzymes retrieved and developed in the projects are expected to follow suit in the coming months and years. Three patent applications were made in the project for specific enzymes in biorefinery processes. One particular carbohydrate active enzyme was discovered and developed for production in a GRAS organisms and licenced to an external SME for use in a recently developed biorefinery process in production of oligosaccharides for use in the medical industry.

The results of the project were presented in a number of scientific conferences mostly in Europe, but also in Israel, Cuba, Indonesia and India. These included specialized conferences and workshops on carbohydrates and carbohydrate enzymes, the 26th International Carbohydrate Symposium (Madrid, 2012), Eurocarb 16 and 17 (Sorrento, 2011 and Tel Aviv, 2013) and the 9th & 10th Bioengineering Symposia CBM9 and CBM10 (Lisbon, 2011 and Prague 2013) and more specifically on starch processing and amylolytic enzymes, the 6th European symposium on enzymes in Grain processing (Copenhagen 2011), and 5th Symposium on the alpha-amylase family (Slovakia, 2013) Two book chapters and nine papers were published during the project and three more have been submitted. A number of manuscripts are in preparation.
The underlying aim of AMYLOMICS is to help increase economic growth and sustainability of the carbohydrate industry, to improve efficiency of bioconversion processes; to increase product diversity and to decrease waste. Enzymes that were obtained in the project can be used in different processing platforms i.e. for production of fermentable sugars or specialized, novel or altered oligosaccharides. Such products can replace traditional sugars as sweeteners in various food merchandises. They can also be used as food supplements 6 with health promoting benefits and as agents for improving texture, palatability, stability and storage time of food products.
Industry is constantly in search of improved and more environmentally benign processes, novel products and derivatives with health beneficial effect. In the AMYLOMICS project an enormous number of novel genes were retrieved from a number of metagenomes. Carbohydrate processing genes were specifically targeted and a total of more than 4500 genes were identified. Of these, approximately 800 encoded starch processing and modifying enzymes. 303 complete genes (from metagenomes genomes) encoding novel starch and other carbohydrate processing enzymes were extracted for further study. 204 genes were cloned, 157 expressed and 138 enzymes were screened for industrial relevant properties.
The massive information generated in Amylomics on microbial diversity of geothermal areas will serve as the basis for future targeted biotechnological exploitation of thermophiles. It was demonstrated that by selective enrichment, while decreasing diversity resulted in greater number of complete target genes and further significant improvement was obtained by alternating shot gun sequencing, sequence capture and paired end sequencing. These results are of importance for designing efficient mining strategies of environmental biotopes for enzymes of industrial interest. The carbohydrate industry and especially the starch processing sector is highly developed in Europe and European companies and play a leading role on the global market. The large number genes retrieved in the project will serve as an important future source for carbohydrate active enzymes for use in European industry.

Main contact information

Coordinator: Prof Gudmundur Hreggvidsson, MATIS, Vinlandsleid 12, 113, Reykjavik, Iceland,

Prof Daslav Hranueli, SEMGEN, Laniste 5D, 10000, Zagreb, Croatia, HR

Dr. Jakob K. Kristjansson, PROKAZYME, Vinlandsleid 14, 113 Reykjavik Iceland,

Prof. Lubbert Dijkhuizen, RIJKSUNIVERSITEIT GRONINGEN, Broerstraat 5, 9712CP Groningen, Netherlands,

Dr. Josef Altenbuchner, UNIVERSITAET STUTTGART, Keplerstrasse 7 , 70174 , STUTTGART , Germany,

Prof. Eva Nordberg-Karlsson, LUNDS UNIVERSITET, Paradisgatan 5c, 117 Lund, Sweden

Dr. Sophie Defretin, ROQUETTE FRERES SA, Rue De LA Haute Loge, 62136 Lestrem, France,

Dr. Ian Fotheringham, INGENZA, Wallace Building, Roslin BioCentre, EH25 9PP Roslin, Scotland,