Fermentation Organism Development

Investigate Pentose Sugar Transport in Zymomonas

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Ohio State University, Columbus, OH 43210

Principal Investigator: Tyrrell Conway, 614.292.2301

Contract Number: XCG-8-18124-01

Contract Period: 08/98–04/99

Contract Funding:

FY 1998: $75,830

Objective: To characterize and improve pentose transport in Zymomonas mobilis.

Approach/Background: There are no naturally occurring industrial biocatalysts that efficiently convert hemicellulose-derived pentose sugars, and this lack severely limits ethanol production from lignocellulosic feedstocks. Recently, an NREL research team genetically engineered the efficient ethanol-producing bacterium Zymomo-nas mobilis with the ability to ferment xylose and arabinose to ethanol. Although this recombinant biocatalyst represents an enormous improvement, in that for the first time we have strains with the metabolic pathways necessary for converting pentoses to ethanol, no provision was made to improve the entry of pentoses into the cell.

Until recently, very little was known about sugar transport by Z. mobilis. A single report, published in 1985, demonstrated that glucose transport in Z. mobilis is a nonenergy-dependent (nonconcentrative), low affinity, high velocity process. The very nature of this facilitated-transport system makes it so difficult to study experimentally that the details of the transport process were not pursued in any laboratory until the 1990s. We then cloned the gene that encodes the Z. mobilis glucose facilitator (glf) and expressed it in E. coli. During the previous FY 1996 and FY 1997 funding period, we characterized sugar transport by Glf and established that not only is this the only sugar transporter in Z. mobilis, but that it also has relatively poor kinetics for xylose transport by comparison to glucose.

The basic approach for improving xylose uptake was to modify either the expression levels or the kinetic properties of the native glucose transporter, Glf. Alternatively, it was thought that it might be possible to express a foreign xylose transporter in Z. mobilis. Lastly, the transport kinetics for arabinose would be measured and the shortcomings of this process would be addressed in a fashion similar to that used for xylose. First we overexpressed the native Glf in Zymomonas; the resulting strain did not show any significant improvement in uptake xylose. This result indicated that overexpression of the native Glf was insufficient to improve xylose uptake and that rather the relative affinity for xylose must be addressed. Next, the native E. coli xylose transporter, encoded by the xylE gene, was cloned and expressed in recombinant Z. mobilis. This strain was able to transport xylose at a two-fold higher rate than did the isogenic strain lacking the E. coli xylose transporter. However, the overproduction of the E. coli XylE transporter profoundly reduced the growth rate and growth yield of the recombinant Z. mobilis. Apparently the energy-dependent transport mechanism of E. coli XylE caused an energy drain that severely hindered growth of the recombinant Z. mobilis strain. This result indicated that it would be necessary to use only energy-independent transport systems for improving pentose sugar uptake. It would be necessary to develop a system for screening for mutant transporters with improved pentose transport properties. This idea became the focus of the FY 1999 research activities. The approach that was decided upon was to use ‘laboratory directed-evolution’ to create a transporter with good kinetics for xylose or arabinose (or both).

Status/Accomplishments: This is a daunting task because directed-evolution schemes require a method for screening very large numbers of clones in order to select the few that show ideal properties. Unfortunately, the genetic systems available for use with Z. mobilis are not efficient enough to be used for "high-throughput" screening.

Thus it was decided that E. coli, which is arguably the most studied and best understood of all creatures, would be used as a model to screen the thousands of clones generated by directed evolution. The high efficiency of E. coli transformation protocols makes it ideally suited for high-throughput screening of large libraries of clones. We think that a strong selection system in E. coli can be created with mutant strains completely deficient in xylose uptake while retaining the capacity for xylose catabolism. No such mutant is currently available, but we have designed a suitable strategy.

E. coli possesses two systems for xylose uptake. The first is the proton/xylose symporter encoded by xylE. The second is an ATP dependent (ABC) transporter encoded by xylFGH. An E. coli xylE, xylFGH double mutant strain should be unable to transport, and therefore unable to grow on xylose, making it a suitable host for selection of cloned transporter genes that rescue the ability to grow on xylose. An analogous approach, suitable for screening for improved arabinose transporters, was also designed.

The E. coli strain to be used for screening of improved xylose transporters must contain mutations in xylE and xylFGH, and the E. coli strain to be used for screening of improved arabinose transporters must contain mutations in araE and araFGH. The mutations must be created by consecutive genetic manipulation of the wildtype. The primary mutations, xylE or araE, respectively, have been easy to make and both are completed. Likewise, mutations of xylFGH and araFGH have been constructed. However, we have had great difficulty in combining both mutations in a single strain. Thus we ended the funding period only partially finished with the mutant constructions that are foundational to the approach we wish to use for selecting improved pentose transporters. New techniques for mutant construction in E. coli are on the horizon and these may be of value in future research.

Publications and Presentations: None

Summary Date: March 2000

Pathway Engineering to Improve Ethanol Production by Thermophilic Bacteria

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Dartmouth College, Hanover, NH 03755-1404

Principal Investigator: Lee Lynd, 603.646-2231, Mary Lou Guerinot, 603.646-2231

Contract Number: XCG-7-17015-01

Contract Period: 07/98-09/00

Contract Funding:

FY 1998: $99,603

Approach/Background:In one process step, consolidated bioprocessing (CBP) produces cellulase enzymes, causes cellulose hydrolysis, and ferments the resulting sugars to ethanol. The CBP strategy has the greatest potential of any process improvement ever analyzed to lower the cost of producing ethanol from cellulosic biomass. CBP may be pursued either by improving product production properties of organisms that naturally utilize cellulose, or by heterologously expressing cellulase enzymes in an organism that naturally has good production production properties.

During the period of performance we used the former approach. The project objective is to to improve ethanol yields by using gene knockout in two thermophiles: the cellulolytic Clostridium thermocellum and the xylose-using Thermoanaerobacter thermosaccharolyticum. Carrying out such gene knockout requires cloned catabolic genes and a gene-transfer system. At the start of the project we had neither.

Status/Accomplishments: During FY 1998, our main accomplishments were the successful cloning and sequencing genes from C. thermocellum for two of the four gene knockout targets of interest to us: acetate kinase and phosphtransacetylase. These genes were found to exist as part of a putative operon, and showed high sequence homology and similarity relative to the same genes from other organisms. In addition, we worked toward development of a gene transfer system, although this objective was not achieved during the period of performance.

During FY99, we successfully cloned and sequenced a hydrogenase from C. thermocellum, which is one of our target enzymes for gene knockout work. With genes for several target enzymes in hand, our primary focus shifted to developing a gene transfer system. Motivated by studies in the literature that found trying many strains useful in developing genetic systems, we isolated and substantially characterized 12 new strains of thermophilic, anaerobic, cellulose-degrading bacteria. As well, we obtained preliminary results (differential growth on antibiotic-selective media, plasmid recovery, and retransformation) that indicated transformation by use of the broad-host range plasmid pHV33 with a chloramphenicol-resistance marker. Because spontaneous DNA uptake as well as electroporation gave positive results, both approaches were pursued. By the end of the period we still had not achieved the goal of a reliable, high-frequency transformation system, although we are quite hopeful that our new approaches will ultimately bear fruit. Development of a gene transfer system is the sole objective of continuing NREL-supported work involving thermophilic bacteria at Dartmouth.

Publications and Presentations:

1. Stevens, D., D. Stevenson, M.L. Guerinot, and L.R. Lynd. 1998. Cloning and sequencing of the genes encoding phosphotransacetylase and acetate kinase from C. thermocellum. Clostridium V. Presented at Fifth International Workshop on Regulation of Metabolism, Genetics, and Development of the Solvent- and Acid-Forming Clostridia. Toulouse, France. June 27 [1998].
2. Stevens, D., M.L. Guerinot, and L.R. Lynd. 1997. Sequence of acetate kinase and phosphotransacetylase from C. thermocellum ATCC 27405. Genbank Accession # AF041841.
3. Desai, S.G., D. Stevenson, M. Ozkan, J. Beane, M.L. Guerinot, and L.R. Lynd. 2000 (in preparation). Characterization and development of thermophilic, saccharolytic bacteria for consolidated bioprocessing. To be presented at the Sixth International Workshop on Regulation of Metabolism, Genetics, and Development of the Solvent- and Acid-Forming Clostridia. Champaign-Urbana, IL. June [2000].
4. Ozkan, M., S.G. Desai, Y. Zhang, D. Stevenson, J. Beane, M.L. Guerinot, L.R. Lynd. Characterization and development of thermophilic cellulolytic bacteria for biomass processing. To be presented at the Twenty-second Symposium on Biotechnology for Fuels and Chemicals, Gatlinburg, TN. May [2000].
5. Desai, S.G., D. Stevens, D.M. Stevenson, H.L. Prince, M.L. Guerinot, and L.R. Lynd. Cloning and sequencing of hydrogenase from Clostridium thermocellum 27405. Genbank accession #AF148212.

Summary Date: March 2000

Enhancement of Acid Tolerance in Zymomonas mobilis

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: University of Wisconsin, Madison, WI53706-1187

Principal Investigator: Charles W. Kaspar, 608.263.6936

Contract Number: XCG-7-17037-01

Contract Period: 06/98–08/99

Contract Funding:

FY 1999: $50,000

Objective: To generate fundamental information on pH homeostasis in Zymomonas and its general stress system and design strategies to produce a more acid-tolerant and robust Zymomonas production strain.

Approach/Background:In an attempt to augment the acid-tolerance properties of the engineered strain, two genes were introduced into Z. mobilis to enhance the acid tolerance or promote survival. The mechanisms of pH homeostasis or stress-protection systems in Z. mobilis have not been fully elucidated. Stress proteins have been identified in Z. mobilis that are produced in response to ethanol-shock, heat-shock, or entry into stationary phase, and these proteins are homologous to heat shock proteins from other bacteria (i.e., DnaJ, DnaK, GroES, and GroEL). In this study, the dps gene from E. coli which protects DNA from various assaults, including low pH, and a portion of cbpA from encoding for a small basic (33%) peptide (sbp) were cloned into Z. mobilis CP4 to supplement the existing stress-protection system(s). In addition to enhancing the acid tolerance of genetically engineered strains of Z. mobilis, results from this study provide a better understanding of the stress-protection systems and will be valuable in future attempts to expand the industrial fermentation potential of this organism.

Status/Accomplishments: Investigations of the survival of log- and stationary-phase cells of Z. mobilis in glycine-HCl buffer (pH 3.0) and in sodium acetate-acetic acid buffer (pH 3.5) found no statistical differences. The absence of a stationary-phase protection system was also supported by the absence of hybridization in southern blot analyses with probes to rpoS and dps from E. coli. These results indicate that Z. mobilis is different from other Gram-negative bacteria that contain stationary-phase protection systems. In addition, Z. mobilis strains CP4 and ATCC 39676 tested negative for arginine and lysine decarboxylase, two key pH homeostasis enzymes. These results suggest that the pH and ethanol tolerance exhibited by Z. mobilis is probably due to a passive protective system such as an H+ and ethanol-impermeable membrane.

The plasmid pZB186 was modified by removing tetA and inserting bla (GenBank accession #JO1749) at the same location (pZBJB99-1). The promoter and chloram-phenicol-resistance gene were also removed. At this site, the p-tac promoter from pKK-223-3 and dps (pZBJB99-2) or sbp gene (pZBJB99-3) was inserted. CP4 strains harboring pZBJB99-1, pZBJB99-2, or pZBJB99-3 exhibited significantly increased acid tolerance in comparison with the wildtype strain when challenged at pH3 (Trypticase soy broth adjusted to pH 3.0 with HCl). The greatest survival (percentage of survivors) occurred in CP4 strains carrying the engineered plasmids pZBJB99-2 (Ptac-dps) and pZBJB99-3 (Ptac-sbp). The enhanced survival of CP4 with pZBJB99-1 is consistent with the results for CP4 (w) grown in the presence of 100 µg/ml ampicillin. It has been previously reported that an ampicillin-resistant determinant is chromo-somally located in Z. mobilis, but it was not reported if the gene encoded for b -lactamase or an alternative mechanism of resistance (i.e., altered membrane or membrane-binding protein). These results indicate that growth in the presence of ampicillin enhances the survival at low pH and growth in the presence of higher concentrations of ampicillin (200 mg/ml) increases acid tolerance. CP4 with pZBJB99-1 constructs were grown in 200 FACE="Symbol" mg/ml because they could tolerate the higher amount owing to the plasmid-encoded b -lactamase gene. It is unclear why growth in the presence of ampicillin increases acid tolerance.

Ampicillin-induced acid tolerance also played a significant role in the survival of CP4 strains in sodium acetate-acetic acid buffer (pH 3.5). CP4 strains with pZBJB99-1, pZBJB99-2, and pZBJB99-3 had significantly greater survival than both the parent strain and the parent strain grown in the presence of 100 mg/ml ampicillin. However, there was no significant difference in the survival of CP4 with pZBJB99-1 or pZBJB99-2, which indicates that growth in the presence of 200 mg/ml ampicillin was responsible for the noted acid tolerance in sodium acetate buffer and not Dps production. These results differ from the results noted for the survival of CP4 with pZBJB99-2 in HCl-acidified TSB. Thus, different mechanisms of acid-tolerance may be involved with different acids and in the presence or absence of nutrients. Some additional protection to acid was afforded by pZBJB99-3, which enhanced survival throughout the course of the acetic acid challenge. Previous studies with E. coli genes under the regulation of the Ptac promoter have shown that the level of expression is only twice that of a promoter-less gene in Z. mobilis. This result indicates that a promoter with increased expression compared to Ptac should be evaluated.

Results from this study have demonstrated that dps and gene encoding for a portion of the cbpA gene afford some acid tolerance to Z. mobilis although protection by Dps was demonstrated only when HCl was used as the acidulent. Additionally, growth in the presence of ampicillin induced or selected for acid tolerance in Z. mobilis. Studies are needed to determine if increased acid tolerance is associated with beneficial fermentation properties and ethanol production.

Publications and Presentations: None

Summary Date: March 2000

Genomics in Zymomonas mobilis

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Integrated Genomics Inc., 2201 W. Campbell Drive, Chicago, IL 60612

Principal Investigator: Yuri Nikolsky, (312) 491-0846

Contract Number: ACG-8-18159-01

Contract Period: 09/98–03/99

Contract Funding:

FY 1998: $35,000

Objective: Our objective is to obtain genomics research services so that Z. mobilis can be better understand at the genomic level.

Approach/Background: Integrated Genomics Inc. (IGI) focuses on bacterial genome sequencing and computerized annotation. IGI also provides a full line of genomics research services and products for the pharmaceutical, agricultural, environmental, and chemical industries. Recently, IGI sequenced the Zymomonas mobilis genome. IGI also reconstructed metabolic pathways for Z. mobilis.

Status/Accomplishments: The focus of this work is to identify open reading frames, pro-moter regions, and pathways that can be used to maximize ethanol yield and enhance robustness of Z. mobilis. An overview of the carbohydrate metabolism is provided. Several open reading frames and flanking sequences were searched, including those for the enzymes aldose reductase, NADH oxydases/NADH hydrogen-ases, and pentose transporters. The subcontractor will produce metabolic modeling of the pathways. We are particularly interested in energy metabolic pathways in Z. mobilis. Metabolic modeling of the Embden-Meyerhof-Parnas pathway can be also performed in Zymomonas.

Publications and Presentations: None

Summary Date: March 2000

Improve Zymomonas for Xylose and Arabinose Fermentation

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory, Corn Refiners Association, and National Corn Growers Association

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO 80401-3393, http://www.nrel.gov

Principal Investigator:Min Zhang, 303.384.7753, Min_zhang@nrel.gov

Contract Number: N/A

Contract Period: 10/98–09/99

Contract Funding:

FY 1999: $459,829; CRA/NCGA: $80,000

Objective: Our objectives are to improve the stability of the recombinant xylose-and-arabinose–fermenting Zymomonas without using antibiotics, and to improve the fermentation performance of xylose-and-arabinose-fermenting Zymomonas.

Approach/Background: The ethanologenic bacterium Z. mobilis has been metabolically engineered to ferment xylose and arabinose. Sveral key enzymes have in the xylose assimilation, arabinose assimilation, and pentose phosphate pathways have been introduced into a single microorganism that efficiently ferments hexose and pentose sugars (derived from lignocellulosic feedstocks) to ethanol. The recombinant Zymomonas strains were constructed by cloning the genes necessary for pentose metabolism into a shuttle vector, followed by introducing them into the microorganism using tetracycline (Tc) as a selection maker. The plasmid-bearing xylose-and-arabinose–fermenting Zymomonas strain is not stable in the absence of antibiotic (Tc) owing to loss of the plasmid that results in low fermentation yield.

In addition, using antibiotics for selection of recombinant microorganisms at commercial scale is not only a costly process but it also imposes environmental safety issues. It is preferable not to use antibiotic-resistance genes in genetically engineered microorganisms.

To enhance the genetic stability of the recombinant Zymomonas in the absence of antibiotic selection, a more desirable approach is to integrate genes of interest into the host genome. Although a number of methods are established for integrating genes into E. coli and other well-known microbes, no method has been described for Z. mobilis. In FY 1997, a gene integration system in Z. mobilis using mini-Tn5 was developed. In FY 1998, four genes necessary for xylose fermentation were inserted into the Z. mobilis genome using the mini-Tn5. In addition, another transposon transposition system using mini-transposon Tn10 was developed. Also, targeted gene integration and gene inactivation based upon homologous recombination were demonstrated for the first time.

We combine the gene integration systems developed in the past two years to insert all seven genes needed for xylose and arabinose fermentation into the Zymomonas genome.

Status/Accomplishments: Using transposon integration and homologous recombination, the seven key genes encoding enzymes in the xylose assimilation (xylA/xylB), arabinose assimilation (araBAD), and pentose phosphate (talB/tktA) pathways were introduced into the Z. mobilis genome. The genetic stability of the xylose-and-arabinose–fermenting Zymomonas in the absence of antibiotic selection was significantly improved.

Two operons containing Pgap-xylA/xylB and Peno-talB/tktA were assembled in mini-transposon Tn5 and inserted into the Zymomonas genome through random integration to generate chromosomal integrated strains capable of fermenting xylose. These strains were evaluated for stability in nonselective medium (containing glucose only) and for fermentation performance in a mixture of glucose and xylose. Two of the strains demonstrated stability greater than 90 generations in nonselective medium and one of them, strain C25, demonstrated superior performance on glucose and xylose. This strain was chosen for use as a host for integrating the arabinose-assimilation genes. The operon that contained genes required for assimilating arabinose (Pgap-araBAD) was subsequently integrated into the chromosome of xylose-utilizing integrant C25 either by homologous recombination (site specific) or by transposon integration (random integration).

In homologous recombination, a lactate dehydrogenase gene (ldh) in the chromosome of C25 was used as a target for the integration of Pgap-araBAD. An integrative plasmid (replicative in Zymomonas ) containing ldh region inserted with Pgap-araBAD was used to transform Z. mobilis C25. After plasmid curing and enrichment in a rich medium containing arabinose, potential integrants were isolated. Southern hybridization analysis confirmed that Pgap-araBAD was integrated in the ldh in the chromosome of C25.

For integration using Tn10 transposon, a cassette was constructed in which Pgap-araBAD was inserted between two short inverted repeats in a nonreplicative plasmid. The cognate transposase gene, however, was located outside the cassette to increase the stability of the integrated genes. This plasmid was transferred from an E. coli donor to C25 through conjugation. The transconjugates (integrants) were selected directly on mating plates with arabinose as the sole carbon source. Southern hybridization analysis confirmed that Pgap-araBAD was integrated at different loci in the genome of C25.

All the integrants from both homologous recombination and transposon integration were able to grow on xylose and arabinose as the sole carbon source and co-fermented glucose, xylose, arabinose to ethanol. Enzymatic assays confirmed that all the integrated genes were expressed. Two chromosomal integrated strains demonstrated stability for 160 generations in nonselective medium (containing glucose only). The chromosomal integrated strains demonstrated an ethanol process yield (83% of the theoretical) similar to that of the plasmid-bearing strain. This yield was derived from a mixture of glucose, xylose, and arabinose, although the integrated strains had only a single copy of the seven arabinose-and-xylose–fermenting genes.

Additionally, we found a better XI from a Lactobacillus strain. The XI showed high activity over a wide pH range, pH 5–7.5. Additionally it is more tolerant to ethanol. The XI appears to be a good source for XI in Zymomonas.

Future directions for improving Zymomonas as robust microbial biocatalysts for efficient conversion of a variety of sugar feedstreams to ethanol include (1) screening, selecting, and expressing better pentose transporters, xylose-assimilating enzymes, and potential energy-generating enzymes; (2) cloning the xylitol-producing genes and eliminating xylitol formation through gene inactivation in Zymomonas; (3) adaptation to corn fiber hydrolysate or sawdust hydrolysate using chemostat and chemical mutagenesis; (4) constructing chromosomal integrated strains in hosts that possess preferred characteristics; and (5) using flux analysis and functional genomics as tools to facilitate our metabolic engineering efforts.

Publications and Presentations:

1. Zhang, M., Y.C. Chou, X.K. Lai, S. Milstrey, N. Danielson, K. Evans, A. Mohagheghi, and M. Finkelstein. 1999. Recent advances in development of Zymomonas mobilis for pentose utilization. Presented at the Twenty-first Symposium on Biotechnology for Fuel and Chemicals, Fort Collins, CO. May 3–7 [1999].
2. Chou, Y. C., S.P. Milstrey, K.W. Evans, and M. Zhang. 1999. Inactivation of the D-lactate dehydrogenase gene in Zymomonas mobilis through homologous recombination. Presented at the general meeting of the American Society for Microbiology, Chicago [1999].
3. Zhang, M., Y.C. Chou, A. Mohagheghi, K. Evans, S. Milstrey, X.K. Lai, and M. Finkelstein. 2000. Genetic improvement of Zymomonas mobilis for ethanol production: Chromosomal integration of the xylose and arabinose-fermenting genes. Presented at the Twenty-second Symposium on Biotechnology for Fuels and Chemicals, Fort Collins, CO. May 1–4 [2000].

Summary Date: March 2000

Develop Advanced Lactobacillus Biocatalyst

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO 80401-3393, http://www.nrel.gov

Principal Investigator: Min Zhang, 303.384.7753, Min_zhang@nrel.gov

Contract Number: N/A

Contract Period: 10/97–09/98

Contract Funding:

FY 1998: $115,780

Objective: Our objective is to introduce the ethanol production pathway into Lactobacillus MONT 4.

Approach/Background: Lactobacillus is a promising microorganism for producing ethanol because it can ferment many of the sugars commonly found in biomass. It offers potential advantages in biomass fermentations including high ethanol tolerance, resistance to the inhibitors present in hydrolyzates, fermentation at low pH, thermotolerance, and high lactate yield (wildtype obligate homofermentative Lactobacilli do not produce ethanol). Lactobacillus is used commercially to prepare a variety of food and feed products. Because it is thermotolerant, Lactobacillus could permit simultaneous saccharification and fermentation (SSF) processes using commercially available Trichoderma cellulase preparations with optimal enzyme activities around 50^oC. Metabolic engineering of homofermentative Lactobacillus for ethanol production from lignocellulosic feedstocks would require the introduction of genes encoding enzymes for both pentose metabolism and ethanol production. In FY 1996 we introduced a xylose metabolism pathway into a homofermentative Lactobacillus strain, MONT4, to produce lactate from xylose at a yield of 0.95 g lactate/g xylose. The focus of this work is introducing the ethanol-production pathway into L. MONT 4.

Status/Accomplishments: We have introduced the Zymomonas mobilis pyruvate decarboxylase (PDC) and alcohol dehydrogenase (ADH) genes, designated as pdc and adh, respectively, into L. MONT4 using the lactococcal expression vector. The Zymomonas adh and pdc genes were expressed in L. MONT4 as demonstrated by enzymatic analysis. However, the L. MONT4 recombinant strain produced only traces of ethanol and as much lactic acid as the wildtype strain, suggesting that the carbon flow in the recombinant strain is still directed toward lactic acid. Consequently, inactivation of the lactate dehydrogenases in L. MONT4 became necessary for redirecting the carbon flow to ethanol production.

We identified that L. MONT4 possesses both L(+)-lactate dehydrogenase (LLDH) and D-(-)-lactate dehydrogenase (DLDH) activities. The structural genes encoding LLDH and DLDH from L. MONT4, designated as lldh and dldh, respectively, were cloned in Escherichia coli using degenerated primers. Deleted versions of both the lldh and dldh genes were constructed and cloned into a temperature-sensitive integration vector for the inactivation of the lldh and dldh genes. A chromosomal deletion in the dldh gene was obtained in L. MONT4 by a two-step homologous recombination process, resulting in a DLDH-negative derivative of L. MONT4. Research is in progress to create a stable chromosomal deletion in the lldh gene of L. MONT4 to obtain a LLDH-negative deriv-ative of L. MONT4 . Future research efforts will include introducing the Z. mobilis pdc and adh genes into lactate dehydrogenase-negative derivatives L. MONT4 and testing for ethanol production and also to create a double mutant which is deficient in both L- and D-LDH.

Publications and Presentations:

1. Picataggio, S.K., M. Zhang, M.A. Franden, J.D. McMillan, and M. Finkelstein. Recombinant Lactobacillus for fermentation of xylose to lactic acid. United States patent 5,798,327, issued on August 25, 1998.

Summary Date: March 2000

Develop Lactobacillus for Ethanol Production

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO 80401-3393, http://www.nrel.gov

Principal Investigator: Min Zhang, 303.384.7753, Min_zhang@nrel.gov

Contract Number: N/A

Contract Period: 10/98–09/99

Contract Funding:

FY 1999: $253,443

Objective: To redirect the carbon flow to ethanol by knocking-out the lactic acid production pathway.

Approach/Background: We have used metabolic engineering to introduce the Z. mobilis pdc and adh genes into Lactobacillus MONT4 to promote ethanol production by Lactobacillus. Although the Zymomonas adh and pdc genes were expressed in L. MONT4, the recombinant strain produced only traces of ethanol and only as much lactic acid as the wildtype strain, suggesting that the carbon flow in these recombinant strains is still directed toward lactic acid. To inactivate the lactate production pathway we identified and cloned L. MONT4 lldh and dldh genes. Subsequently we constructed a D-LDH-deficient L. MONT4 mutant through a two-step homologous recombination process.

Our research effort in this fiscal year is to continue to inactivate the lactic-acid-production pathway in order to redirect the carbon flow to ethanol production. A similar approach is used to construct L-LDH-deficient L. MONT4.

Status/Accomplishments: We had limited success using a similar approach to inactivate LLDH. A LLDH-negative derivative of L. MONT4 was obtained using an autoreplicative, unstable integration vector.

The plasmid containing the Z. mobilis adh and pdc genes was introduced into the newly constructed DLDH-deficient L. MONT4 and LLDH-deficient L. MONT4 strains. However, both recombinant strains produced only traces of ethanol, an amount similar to that of L. MONT4 containing the plasmid. Therefore, we decided to construct DLDH- and LLDH-negative L. MONT4 derivatives to direct carbon flow toward ethanol.

We tried several ways to construct a double mutant deficient in both DLDH and LLDH. However, much of work turned out to be unsuccessful. We have begun new efforts based on new strateties for obtaining a DLDH- and LLDH-negative L. MONT4 derivative for ethanol production.

Publications and Presentations: None

Summary Date: March 2000

Development of Arabinose-Fermenting Saccharomyces

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory, Corn Refiners Association, and National Corn Growers Association

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO 80401-3393, http://www.nrel.gov

Principal Investigator:Min Zhang, 303.384.7753, Min_zhang@nrel.gov

Contract Number: N/A

Contract Period: 10/98–09/99

Contract Funding:

FY 1999: $261,248, CRA/NCGA: $80,000

Objective: To develop an arabinose-fermenting yeast

Approach/Background: Saccharomyces cere-visiae, traditionally used for fermenting glucose-based feedstock in the ethanol industry, is unable to ferment the pentose sugars D-xylose and L-arabinose to ethanol. Much effort has been focused in developing Saccharomyces to ferment xylose to ethanol in the past 10 to 15 years. Expression of the bacterial xylose isomerase gene in Saccharomyces was unsuccessful. However, expression of the xylose reductase and xylitol dehydrogenase genes from Pichia and xylulokinase gene from S. cerevisiae, under strong glycolytic promoters in Saccharomyces, yielded a xylose-fermenting yeast. Because L-arabinose is the other major pentose sugar in corn fiber hydrolysates, we are working to develop arabinose-fermenting yeast under a Cooperative Research and Development Agreement with the Corn Refiners Association (CRA) and the National Corn Growers Association (NCGA). L-arabinose metabolism in bacteria involves three enzymes: L-arabinose isomerase (araA), L-ribulokinase (araB), and L-ribulose-5-p 4-epimerase (araD). S. cerevisiae contains the pathway that uses and ferments the final product, xylulose-5-phosphate, which is generated by the sequential action of these three enzymes. We propose to introduce the bacterial arabinose-assimilating genes into yeast. In addition to introducing the arabinose assimilation pathways, we also propose to select and construct an appropriate strain that is favorable for arabinose utilization.

Status/Accomplishments: First we selected an appropriate S. cerevisiae strain that is favorable for arabinose uptake and then modified the strain to eliminate potential formation of byproducts. The strain is designated as BFY002.

We designed and used primers to isolate E. coli araA, araB, and araD genes by PCR from a Zymomonas arabinose-fermenting plasmid. These genes have been cloned into the yeast expression vector pBFY004, which contains TRP1 selection marker. The yeast strain BFY002 has been separately transformed with each of the expression plasmids, and analysis of the tranformants showed that L-ribulokinase (araB) and L-ribulose-5-P-4-epimerase (araD) were expressed at a high level and L-arabinose isomerase (araA) was expressed at a low level.

In order to introduce all three ara genes into the same cell, we constructed URA3 and HIS3 expression vectors by re-engineering the TRP1 plasmid pBFY004. The TRP1 gene was removed and replaced with a SalI restriction site to generate a plasmid designated as pBFY011. The URA3 gene was isolated as a SalI fragment and cloned into the SalI site to construct the plasmid pBFY012. Similarly, a HIS3 expression vector, pBFY013, was constructed by engineer-ing and cloning the HIS3 gene into pBFY011.

The engineered ara genes are being cloned into each of these expression vectors, and appropriate combinations of the generated constructs will be introduced into the strain BFY002. The transformants will be characterized and assayed for growth and fermentation.

Publications and Presentations: None

Summary Date: March 2000

Biocatalyst Development of Improved Zymomonas

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory, Corn Refiners Association, and National Corn Growers Association

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO 80401-3393, http://www.nrel.gov

Principal Investigator:Min Zhang, 303.384.7753, Min_zhang@nrel.gov

Contract Number: N/A

Contract Period: 10/97–09/98

Contract Funding:

FY 1998: $764,000, CRA/NCGA: $80,000

Objective: To continue to develop gene integration systems and improve stability of the recombinant xylose-fermenting Zymomonas and enhance fermentation performance of xylose-fermenting Zymomonas.

Approach/Background: The ethanologenic bacterium Z. mobilis has been metabolically engineered to ferment xylose and arabinose. Several key enzymes (in the xylose assimilation, arabinose assimilation, and pentose phosphate pathways) were introduced into the microorganism for efficient fermentation of hexose and pentose sugars derived from lignocellulosic feedstocks to ethanol. The recombinant Zymomonas strains were con-structed by cloning the genes necessary for pentose metabolism into a shuttle vector, which was introduced into the microorganism using tetracycline (Tc) as a selection maker. The plasmid-bearing xylose and arabinose-fermenting Zymomonas strain is not stable in the absence of antibiotic (Tc) owing to loss of the plasmid that results in low fermentation yield.

In addition, using antibiotics to select recombinant microorganisms at a commercial scale is not only costly but also imposes environmental safety issues. It is preferred not to use antibiotic resistance genes in genetically engineered microorganisms.

To increase the genetic stability of the recombinant Zymomonas in the absence of antibiotic selection, a more desirable approach is to integrate the gene (or genes) of interest into the host genome. While there are a number of methods established for gene integration in E. coli and other well-known microbes, no method has been described for Z. mobilis. In FY 1997, a gene integration system was developed that used miniTn5 in Z. mobilis .

We planned to use the mini Tn5 to try to integrate the xylose-fermenting genes into the Zymomonas genome while continuing to develop additional gene integration methods. Multiple gene integration systems were expected to ensure that all the genes required for xylose and arabinose fermentation can be inserted into Zymomonas genome.

Status/Accomplishments: In FY 1998, a total of four genes necessary for xylose fermentation into the Z. mobilis genome was inserted in the Z. mobilis genome using the mini Tn5. In addition, another transposon transposition system using mini-transposon Tn10 was developed. Also, targeted gene integration and gene inactivation based upon homologous recombination were demonstrated.

The four genes necessary for xylose fermentation were integrated into the Z. mobilis genome using mini-transposon Tn5 with or without Tc. The genes are coordinately expressed and the fully integrated Z. mobilis strains are capable of growing on and fermenting xylose to ethanol as efficiently as the plasmid-bearing strains. This work fits well with the Cooperative Research and Development Agreement with the Corn Refiners Association/ National Corn Growers Association. We further evaluated the stability of these fully integrated xylose-fermenting Zymomonas strains. One of the strains showed high ethanol yields from mixtures of glucose and xylose (greater than 80%) for 90 generations in nonselective media. Several others are very stable for 260 generations although the ethanol yields from those strains are lower (approximately 60%). These results showed that the integrated xylose-fermenting Zymomonas strains are suitable for fermentations in the absence of the antibiotics.

We have demonstrated gene insertion and inactivation into the Zymomonas genome via a homologous recombination. This system is based on a putative lactate dehydrogenase (LDH) gene of Zymomonas in which a tetracycline (Tc)-resistant gene was inserted. The plasmids containing the LDH::Tc cassette were used to transform Z. mobilis and the resulted Tc resistance transformants were analyzed by Southern hybridization. The results showed that the LDH::Tc cassette was inserted into the LDH region on the Zymomonas genome. This is the first demonstration of gene integration based on homologous recombination in Zymomonas. More important, this targeted integration system inactivated LDH and eliminated lactic acid formation.

In addition to the mini Tn5 transposon and the homologous recombination systems described above, we have demonstrated that a mini Tn10 transposon is functional in Zymomonas. Using mini Tn10 we are able to integrate a Tc-resistant gene, transaldolase, and transketolase genes into the Zymomonas genome. Thus far we have developed several integration systems that allowed us to integrate xylose-fermenting genes into the genome and to inactivate unwanted gene (or genes) to eliminate by-product formation. We will use these tools to devise strategies to integrate the arabinose-fermenting genes into the Zymomonas genome.

To examine the limiting steps in xylose metabolism we established an in vitro system using cell-free extract for studies of sugar conversion in Zymomonas. Using a cell-free extract we had shown that the xylose utilization pathway was more susceptible than the arabinose utilization pathway to the effects of ethanol. We found that The E. coli xylose isomerase (XI) activity was greatly affected by elevated ethanol, low pH and xylitol concentrations. We started to search for a better XI that is more resistant to ethanol, low pH, and xylitol than E. coli XI.

Publications and Presentations:

1. Danielson, N., M. Finkelstein, and M. Zhang. 1998. In vitro pentose conversion using xylose and arabinose-fermenting Zymomonas mobilis cell-free extract. Presented at the Twentieth Symposium on Biotechnology for Fuel and Chemicals, Gatlinburg, TN. May 3–7 [1998].
2. Evans, K., A. Mohagheghi, Y.-C. Chou, M. Zhang, and M. Finkelstein. 1998. Cofermentation of glucose, xylose and arabinose by genetically engineered xylose/arabinose-fermenting strains of Zymomonas mobilis strains. Presented at the Twentieth Symposium on Biotechnology for Fuel and Chemicals, Gatlinburg, TN. May 3–7 [1998].
3. Evans, K., M. Zhang, and M. Finkelstein. 1998. Selection of Robust Zymomonas mobilis Hosts for Construction of Xylose Fermenting Strains. Presented at the Twentieth Symposium on Biotechnology for Fuel and Chemicals, Gatlinburg, TN. May 3–7 [1998].
4. Mohagheghi, A., K. Evans, M. Finkelstein, and M. Zhang. 1998. Cofermentation of glucose, xylose, and arabinose by mixed cultures of two genetically engineered Zymomonas mobilis strains. Applied Biochemistry and Biotechnology, 70–72:285–299.
5. Zhang, M., Y.-C. Chou, X. Lai, S. Milstrey, N. Danielson, A. Mohagheghi, K. Evans, and M. Finkelstein. 1998. Recent advances in metabolic engineering in Zymomonas mobilis for pentose utilization. Presented at the Society for Industrial Microbiology Annual Meeting, Denver, CO. Aug. 9–13 [1998].
6. Pentose Fermentation by Recombinant Zymomonas. U.S. Patent 5,712,133 issued January 27, 1998.
7. Recombinant Zymomonas for Pentose Fermentation. U.S. Patent 5,726,053 issued March 10, 1998.

Summary Date: March 2000

Metabolic Engeering of Yeasts for Commercial Ethanol Production

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory and IOGEN Corporation, Canada

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Department of Bacteriology, University of Wisconsin, Madison, WI 53706

Principal Investigator:Thomas W. Jeffries, 608.231.9453, twjeffri@facstaff.wisc.edu

Contract Number: ZDH-9-29009-01

Contract Period: 09/99–08/00

Contract Funding:

FY 1999: $94,780, Iogen $ 25,000

Objective: To develop more-robust biocatalysts in the near term

Approach/Background: Yeasts have been used commercially for ethanol production for many years. Their relatively large size, thick cell walls, resistance to bacterial and viral infections, and ability to produce large yields of ethanol favor them for industrial application. Yeasts, however, have a major drawback that hinders for commercial ethanol production from renewable lignocellulosic biomass: they do not efficiently ferment the wide variety of sugars found in lignocellulosic hydrolysates.

This project is using tools of contemporary molecular biology and bioprocess engineering to improve the fermentation of lignocellulosic sugars by yeasts to the point that the process can compete economically with ethanol from cornstarch. The project is attempting to improve the fermentation performance of Pichia stipitis and Saccharo-myces cerevisiae. These yeasts are two of the best available for the fermentation of five and six carbon sugars. Because the physiologies of these two yeasts are quite different, the research objectives also differ.

Status/Accomplishments: P. stipitis efficiently ferments xylose to ethanol if it is supplied with low levels of oxygen. This supply is difficult to provide on a large scale, so the research and development effort is focused on altering the oxygen requirements of P. stipitis while increasing its fermentation rate.

The work has incorporated a gene from S. cerevisiae that enables P. stipitis to grow anaerobically. It has also cloned and disrupted the genes for two of the major respiration pathways. Disruption of cytochrome c (Cyc1) greatly reduced cell yield and increased by half the specific fermentation rate. By expressing the Cyc1 gene in an antisense construct under the regulation of an oxygen sensitive promoter, it is possible to shut down respiration and induce fermentation of xylose in much the same manner as S. cerevisiae acts on glucose.

A United States patent has recently been approved for this novel approach to metabolic engineering. S. cerevisiae ferments glucose efficiently but it does not produce the enzymes necessary to metabolize xylose. This research and development effort has introduced several genes for xylose metabolism from P. stipitis into S. cerevisiae. By using promoters to express the genes for xylose reductase (Xyl1) and xylitol dehydrogenase (Xyl2) at different levels, we have shown that xylulose accumulates in S. cerevisiae, indicating that native xylulokinase levels are too low for efficient xylose metabolism.

We have cloned and expressed the P. stipitis gene for xylulokinase (Xyl3) in S. cerevisiae and have demonstrated that the structural and kinetic properties of the P. stipitis enzyme are quite different from those of the enzyme found in S. cerevisiae. The P. stipitis enzyme has a lower affinity for its substrate but a much higher turnover number. Additional work is underway to engineer P. stipitis for higher fermentative and lower respiration activities. We are also introducing some of the novel respiration enzymes from P. stipitis into S. cerevisiae. This work is being coordinated with an industrial collaborator in order to identify the optimal fermentation conditions for recombinant S. cerevisiae and P. stipitis strains.

Publications and Presentations: None

Summary Date: March 2000

Development and Commercialization of New Ethanol-Producing Strains

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Department of Microbiology and Cell Science, University of Florida, Gainesville, FL 32611

Principal Investigator: Lonnie Ingram, 352.392.8176

Contract Number: ZDH-9-29009-02

Contract Period: 07/99–07/00

Contract Funding:

FY 1999: $188,360, BCI $ 50,000

Objective: Develop more robust biocatalysts in the near term.

Approach/Background:Excellent progress has been made during the past 10 years in developing recombinant biocatalysts that can be used to convert lignocellulose into fuel ethanol. However, as the first of these biocatalysts become commercialized, opportunities to improve and simplify the process have been identified that require the near-term development of more-robust biocatalysts. This 2-year DOE–industry co-funded research project addresses this need. Robust Gram-positive target organisms have been identified that can metabolize all of the sugars present as polymers in lignocellulose. These organisms naturally produce a variety of fermentation products that are not useful. This work proposes to develop a portable ethanol production operon (PET) that will efficiently express alcohol dehydrogenase and pyruvate decarboxylase in these Gram-positive organisms. Resulting biocatalysts should have increased hardiness and produce ethanol at high efficiency.

Status/Accomplishments: Progress during the first 6 months of this project confirms the feasibility of the research goals for near-term application. We have established methods for inserting DNA into the target organisms and have demonstrated that the heterologous genes are expressed in the new host. Although alcohol dehydrogenase and pyrvate decarboxylase genes were already available, it is often challenging to express these at the required high levels in Gram- positive bacteria.

To help solve this problem, we isolated and characterized an alcohol dehydrogenase gene from a Gram- positive thermophile and expressed this gene in a Gram-negative host and a Gram-positive host. A portable ethanol production operon has been constructed by fusing a Gram-positive gene (that provides promoter activity and a ribosomal-binding site) with Zymomonas mobilis pdc followed by the Gram positive alcohol dehydrogenase gene. This gene has been functionally inserted into two different Gram-positive hosts. Measurable levels of ethanol were produced. Initial expression levels were not as high as needed and yields from this initial construct were low.

We also cloned and sequenced most of a pyruvate decarboxylase gene from a Gram-positive bacterium and should soon have the completed sequence. This gene may prove far easier to express at high levels in Gram-positive hosts. In addition, we have now cloned a second functional pyruvate decarboxylase gene from a Gram-negative bacterium. Subcloning and sequencing of this gene are in progress. Pyruvate decarboxylase is the key enzyme that diverts pyruvate to ethanol instead of other products, and it also competes with biosynthetic needs for pyruvate. Each of the pyruvate decarboxylase genes is expected to have different biochemical properties. Combinations of these genes may prove optimal for future genetic engineering of both Gram-positive and Gram-negative biocatalysts.

Publications and Presentations: None

Summary Date: March 2000

Development of More Robust Zymomonas mobilis Strains for Ethanol Production from Lignocellulosic Hydrolysates

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Unisearch Ltd., University of New South Wales Sydney, New South Wales 2052, Australia

Principal Investigator: Peter L. Rogers, 61 (2) 313 6710, P.Rogers@unsw.edu.au

Contract Number: ACG-8-18029-01

Contract Period: 903/98–04/99

Contract Funding:

FY 1998: $75,400

FY 1999: $22,500

Objective: To develop more robust Zymomonas strains through classical mutagenesis.

Approach/Background:The aims of the current subcontract are to (1) evaluate existing recombinant strains of Zymomonas mobilis for ethanol production from hexose and pentose sugars; (2) develop mutagenesis, selection, and adaptation protocols suitable for use with recombinant strains, and determine if plasmid-bearing strains represent the most appropriate starting materials for subsequent studies; and (3) isolate and characterize strains with desired phenotypes. These strains are expected to yield a 15%–20% improvement over parental strains.

Status/Accomplishments: For FY 1998, objective (1) has been met with the identification of a recombinant strain of Z. mobilis ZM4 (pZB5) that was more robust than the strain previously under evaluation at NREL, viz., CP4 (pZB5). The former strain demonstrated faster specific uptake rates of glucose and xylose, and a greater tolerance to ethanol than CP4 (pZB5). Under conditions of 65 g/L glucose and 65 g/L xylose, an ethanol concentration of more than 60 g/L was achieved in 48 h with a yield of 90% theoretical (Y_p/s = 0.46 g/g).

The strain ZM4 (pZB5) has been evaluated in both batch and continuous culture for further comparison with CP4 (pZB5).

For FY 1999, objectives (2) and (3) were met by developing and characterizing a mutant of ZM4 (parent strain) that was capable of growth and ethanol production at higher acetate levels than ZM4. Acetate can be an important inhibitory compound in lignocellulosic hydrolysates. The mutant strain, which demonstrated at least 20% improvement over ZM4, has been made available to NREL.

The acetate-resistant mutant has been characterized further using ¹³C and ³¹P nuclear magnetic resonance (NMR) techniques. These NMR techniques have been applied also to the recombinant strain ZM4 (pZB5) to analyse the mechanism of acetate inhibition. Adding acetate resulted in intracellular de-energization and acidification, and these appear to be the major mechanisms by which acetic acid exerts its toxic effects.

Publications and Presentations:

1. Kim, I-S, K.D. Barrow, and P.L. Rogers. (2000) (in press). Nuclear magnetic resonance studies of acetic acid inhibition of recombinant Zymomonas mobilis ZM4(pZB5). Applied Biochemistry and Biotechnology, v. 84–86.
2. Above paper based on presentation at Twenty-first Symposium on Biotechnology for Fuels and Chemicals, Fort Collins, CO. May 2–6, 1999.
3. Joachimsthal, E., Haggett, K.D. and Rogers, P.L. 1999. Evaluation of recombinant strains of Zymomonas mobilis for ethanol production from glucose/xylose media. Applied Biochemistry and Biotechnology, 77–79:147–158.
4. Above paper based on presentation at Twentieth Symposium on Biotechnology for Fuels and Chemicals, Gatlinburg, TN. May 3–7, 1998.

Summary Date: March 2000

Development of Portable Ethanol-Production Operons That Can Be Expressed in Gram-Positive Bacteria

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Department of Microbiology and Cell Science, University of Florida, Gainesville, FL 32611

Principal Investigator: Lonnie Ingram, 352.392.8176

Contract Number: ACG-8-18048-01

Contract Period: 05/97–11/98

Contract Funding:

FY 1998: $77,278

Objective: The overall goal of this research is to develop novel thermophilic Gram positive biocatalysts that produce ethanol as their dominant product by fermentating sugars.

Approach/Background: The specific aims for the work during the first year were to identify sources of alcohol dehydrogenase genes and pyruvate decarboxylase genes and begin the construction of vectors that can be used to test expression. We were exceedingly fortunate during the first year and were able to make considerable progress toward the development of a portable operon for ethanol production in Gram positive bacteria. Two thermophilic genes are needed that encode pyruvate decarboxylase and alcohol dehydrogenase, respectively.

Status/Accomplishments: Candidate genes were identified after surveying the previously described genes and properties of the encoded products. These candidates are the pdc gene encoding pyruvate decarboxylase from Zymomonas mobilis and the adh gene encoding alcohol dehydrogenase from Bacillus stearo-thermphilus. Both enzymes are stable up to 60°C. Clones were obtained from our deposit collection containing the Z. mobilis pdc. A gene library was prepared from a Florida strain of B. stearothermophilus, which was first isolated by Xiaokuang Lai. A gene library of B. stearothermophilus DNA was prepared and the adh gene cloned directly by function using aldehyde indicator plates. This gene was sequenced; the promoter, ribosomal-binding region, and terminators were located; and appropriate fragments subcloned. This gene was functionally expressed at high levels in E. coli and the thermostability of the gene product confirmed. The sequence of the B. stearothermophilus adh gene was compared with that of other genes in the data base. This gene was more than 90% identical to a previously reported gene from another strain of B. stearothermophilus and 60% identical to the Z. adhA gene cloned previously in our lab. This gene appears to be typical of zinc-containing alcohol dehydrogenases similar to those used by yeast to produce ethanol. We have subcloned the B. stearothermophilus adh gene, with native promoter in both directions, into an E. coli/Bacillus shuttle vector (pUSH2) and transformed both plasmids (pLOI1751 and pLOI1752) into B. subtilis YB886.

Expression of adh was measured in YB886 (pLOI1751) and YB886 (pLOI1752) by monitoring the reduction of NAD. Note that in plasmid pLOI1751, adh is oriented with the lac promoter. Expression with this plasmid was higher (0.77 µ/mg cell protein) than with pLOI1752 (0.59 Iµ/mg cell protein). These activities are similar to the alcohol dehydrogenase activities produced by Z. mobilis during fermentation and should be sufficient for ethanol production in Gram positive bacteria. Potentially, the B. stearothermophilus adh gene and promoter can be used to construct an operon for ethanol production by placing the pdc gene downstream. We have estimated the ADH specific activity needed to produce 1 mole of ethanol per 24 h, assuming a catalyst density of 2 g cell protein/L. This specific activity is approximately 0.34 Iµ/mg cell protein, approximately half of our best construct. We would indeed be fortunate to develop a bug which could produce ethanol at this rate. Further work will be needed to develop highly expressed operons and functionally express both genes in the rmophilic Gram-positive organisms.

Publications and Presentations: None

Summary Date: March 2000

Development of Stable Xylose-Fermenting Saccharomyces Yeasts for Conversion of Sugars from Softwood Hydrolysates to Ethanol

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Purdue University, LORRE, 1295 Potter Center West Lafayette, IN 47907-1295

Principal Investigator: Nancy W. Y. Ho, 765.494.7046

Contract Number: ACG-8-18069-01

Contract Period: 07/98–06/99

Contract Funding:

FY 1998: $100,000

Objectives: We have two objectives:

• to develop new, stable, xylose-fermenting Saccharomyces yeasts that will be free of any restrictions for licensing to companies; and
• to use the genetically engineered stable yeasts to convert sugars from softwood hydrolysates to ethanol under conditions specified by NREL

Even before this subcontract, the principal investigator’s group developed one stable, genetically engineered yeast 259A (LNH-ST) and had a second one 424A (LNH-ST) on its way to completion. These new stable yeasts would be free of any restrictions for licensing to companies and groups that wish to use them to convert cellulosic biomass to ethanol. The purpose of development of the 259A (LNH-ST) and 424A (LNH-ST) yeasts was to test the effectiveness of our integration method and to quickly develop one or two stable, genetically engineered yeasts that were free of any restrictions for licensing to companies and groups. However, the parent yeasts used to develop these stable yeasts were not selected for their tolerance to inhibitors present in the hydrolysates. Under this subcontract, we searched for those yeasts that were known to have strong resistance to inhibitors present in cellulosic hydrolysates and were also free from any restrictions for licensing to companies.

Status/Accomplishments: We found a few yeast strains from ATCC and from private sector. However, none of them are much better than 424A (LNH-ST). Thus, the latter stable yeast is our current best genetically engineered yeast with multiple copies of XD-XR-XK genes integrated into the host chromosome for fermenting sugars from the hydrolysates of cellulosic biomass to ethanol.

However, one of the most exciting results of this subcontract is that our engineered Saccharomyces yeasts can repeatedly and efficiently coferment a mixture of glucose and xylose to ethanol for numerous cycles requiring very little nutrients. Moreover, our "stable" glucose-and-xylose cofermenting yeasts can also repeatedly coferment sugars present in hydrolysates of cellulosic biomass, such as glucose, xylose, mannose, and galactose. And we also developed a set of conditions that our engineered yeasts can produce 20%–40% more ethanol than the unengineered parent yeast. The effectiveness of our yeasts in cofermenting sugars present in the hydrolysates coupled with the fact that the same batch of our engineered yeast can repeatedly and efficiently coferment sugars from cellulosic hydrolysates to ethanol may lead to the development of a much more cost-effective process for producing ethanol from cellulosic biomass.

Publications and Presentations: None

Summary Date: March 2000

Develop Efficient Gene Expression Systems in Lactobacillus

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Department of Microbiology, University of Georgia, Athens, GA 30602-2605

Principal Investigator: Elliot Altman and Mark Eiteman, 706.542.2900

Contract Number: XCI-9-29059-01

Contract Period: 07/99–07/00

Contract Funding:

FY 1999: $17,422

Approach/Background: Lactobacillus species are widely used in diverse industrial applications. With the ever-increasing demands for the overproduction of both endogenous and heterologous proteins in Lactobacillus, easy-to-use expression vectors are needed. Such vectors would find use both in academic research and applied industrial applications. The U.S. Department of Energy (DOE), for example, is interested in the potential use of Lactobacillus to produce ethanol. However, its research efforts have been hampered by the lack of expression vectors that are available for use in Lactobacillus. Our laboratory is creating a family of three Lactobacillus expression vectors that can be used by researchers to facilitate the overproduction of proteins in Lactobacillus species.

Status/Accomplishments: All three vectors will contain replicons, which will allow them to be propagated in both Lactobacillus species and Escherichia coli. The ability of a Lactobacillus expression vector to be propagated in E. coli will simplify cloning procedures, because the transformation efficiencies that can be achieved in E. coli are one-hundred-fold to one-thousand-fold greater than the transformation efficiencies attainable in Lactobacillus. Vector 1 will be a high-copy vector with a strong promoter to be used for the constitutive overproduction of proteins in Lactobacillus. Vector 2 will be a high-copy vector with a strong inducible promoter to be used where control of protein production is desired. Vector 3 will be identical to vector 2, but additionally the a-lacZ peptide will be included for researchers who rely on blue-white cloning strategies. Because all three vectors will be designed with an identical multiple cloning site, it will be trivial to shuffle cloned genes among the three vectors.

Publications and Presentations: None

Summary Date: March 2000

Development of Second-Generation Ethanologenic Yeast

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Department of Bacteriology, University of Wisconsin, Madison, WI 53706

Principal Investigator: Thomas W. Jeffries, 608.231.9453, twjeffri@facstaff.wisc.edu

Contract Number: XXL-9-29034-02

Contract Period: 07/99–07/00

Contract Funding:

FY 1999: $137,172

Objective: Our objective is to improve microbial catalysts that effectively convert the sugars produced from lignocellulosic biomass to ethanol.

Approach/Background: The fermentation of lignocellulose would benefit greatly if the process were possible at elevated temperatures and ethanol concentrations. Under these conditions the efficiencies of enzymatic hydrolysis and product recovery increase. High temperatures and ethanol concentrations are, however, toxic to most organisms—including yeasts. Ethanol and heat destabilize the cellular membranes and interfere with nutrient transport and energy generation. The research being conducted under this subcontract will clone genes for thermotolerance, ethanol tolerance, L-arabinose utilization, and cellulase secretion by expressing large blocks of genes from yeast artificial chromosomes (YACs). Screening to identify genes for these traits will be reduced by positive selection systems. Various clones exhibiting improved thermotolerance, ethanol tolerance, L-arabinose utilization, or cellulase secretion will be recombined, and new libraries would be screened for the combined traits. Genes known to promote xylose fermentation will also be incorporated into the construct. Once genes for all the desired traits are assembled in a single YAC, it will be transferred to a preferred industrial yeast that exhibits high fermentative activity.

Status/Accomplishments: In preparation for this work, the principal investigator has completed a major review of the literature related to thermal and ethanol tolerance. Aeration and pH are critical environmental factors that affect the efficiency of ethanol production. All yeast and fungi require oxygen for the metabolism of xylose. This requirement appears to be related to energetics, redox balance, nutrient uptake and the biosynthesis of essential cell constituents such as ergosterol, unsaturated fatty acids and uracil. Elevated temperatures and ethanol concentrations disrupt the integrity of cellular membranes, interfere with transport, and increase the frequency of petite mutants.

Concomitantly, the loss of membrane integrity increases demands on ATP generation, and xylose uptake requires active transport. Critical physiological factors that increase heat and ethanol tolerance include high levels of ergosterol and unsaturated fatty acids in the plasma membrane, efficient plasma membrane H⁺-ATPase activity, the capacity to stabilize cellular proteins with trehalose and heat shock proteins, and biochemical mechanisms that destroy oxidative radicals. Only a few of the genes determining ethanol and thermotolerance have been identified, and very little is known about ethanol and thermotolerance in xylose- fermenting yeasts. In order to determine whether lipid composition can affect ethanol production during xylose fermentation, we began nutritional studies to supplement cultures with ergosterol, inositol, and unsaturated fatty acids. We have also purchased a CHEF gel apparatus and begun to separate and characterize the genome of xylose-fermenting and ethanol-tolerant yeasts.

Publications and Presentations: None

Summary Date: March 2000

Second-Generation Ethanologen Development - University of New South Wales

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Unisearch Ltd ,University of of New South Wales, Sydney New South Wales 2052, Australia

Principal Investigator: Peter L. Rogers, 61 (2) 313 6710, P.Rogers@unsw.edu.au

Contract Number: XXL-9-29034-03

Contract Period: 08/99–08/00

Contract Funding:

FY 1999: $108,675

Objective: The objective of this project is to develop a thermophilic microorganism capable of producing 5% w/v ethanol at temperatures greater than or equal to 50^oC, from mixtures of hexose and pentose sugars (primarily glucose and xylose).

Approach/Background: The approach is as follows:

• to screen for xylolytic ethanologenic thermophiles from various high temperature environments in Australia, New Zealand, and Southeast Asia;
• to carry out enzymatic and kinetic evaluations of these strains;
• to develop genetic manipulation techniques for cloning specific enzymes (e.g., PDC, ADH) into thermophiles to enhance ethanol production;
• to evaluate stability and fermentation characteristics of new recombinant strains for ethanol production from lignocellulosic hydrolysates.

Status/Accomplishments: The project commenced in August 1999 and was in progress for only 2 months in FY 1999. At this stage, preliminary screening of thermophiles has begun, as well as kinetic evaluation of a known thermophile Saccharobacter caldoxylolyticus capable of using glucose and xylose at 60^oC. In future genetic manipulation studies, PDC and ADH from Zymomonas mobilis (both enzymes stable at 50^o–55^oC) will be cloned into this thermophile for subsequent ethanol production.

Publications and Presentations: None

Summary Date: March 2000

Second-Generation Ethanologen Development - University of Florida

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Department of Microbiology & Cell Science, University of Florida, Gainesville, FL 32611

Principal Investigator: Lonnie Ingram, 352.392-8176, lingram@micro.ifas.ufl.edu

Contract Number: XXL-9-29034-01

Contract Period: 07/99–07/00

Contract Funding:

FY 1999: $281,142

Objective: Our objective is to develop improved microbial catalysts that effectively convert the sugars produced from lignocellulosic biomass to ethanol.

Approach/Background: During the past 10 years, excellent progress has been made in the development of at least four recombinant ethanologenic biocatalysts for the bioconversion of hemicellulose and cellulose into fuel ethanol. Although these biocatalysts are effective and efficient, all have limitations that help increase the cost of biofuel production. The development of a new "Second Generation Biocatalyst" has the potential to greatly reduce process costs associated with production of fuel ethanol. On the basis of discussions with industry and academia, new traits that will help lower costs include fermentation at low pH (pH 5.0 or below) and thermal tolerance (50^oC or above)—while the efficient conversion of all sugar constituents into beers containing 5% ethanol or higher is maintained. Additional useful traits are the production of cellulolytic enzymes to reduce the cost of saccharification and resistance to toxins generated during the dilute acid hydrolysis of hemicellulose. Based on these criteria, a three-phase plan was established. In Phase I new microbial platforms, candidate organisms that have the desired hardiness, will be isolated and identified. Ethanol production traits are to be genetically engineered into these organisms during Phase II. Optimization and process development will be conducted as the final step of this contract, Phase III.

Status/Accomplishments: The first half of Phase I has now been completed and is focused on the isolation of organisms from nature using enrichment culture techniques. Thus far, we have set up enrichments using samples from 25 sites including hot springs, acidic soils, desert, and lignocellulosic compost. The first set of screens isolated 51 different types of organisms that grow aerobically at 50^oC at pH 5.0 or below using 5% hemicellulose hydrolysate as a sugar source supplemented with 1 g/L yeast extract and mineral salts.

Of these, the 25 that grew best have been studied in more detail. Two thirds of these isolates readily formed spores in broth or on solid media. Eight of these were also capable of growth under anaerobic conditions, a characteristic deemed essential for efficient ethanol production. Two of these stored poorly in the low-temperature freezer, another essential trait. The remaining six candidates were sent for initial identification by fatty acid analysis. At the ultrastructural level, three have a characteristic Gram positive wall structure. The remaining three have an unusual structure that is either Gram negative or includes an S-layer. Several of these appear more resistant to furfural, an important toxin in hydrolysate, than current biocatalysts.

A second round of enrichments has been completed by selection in 25% hemicellulose hydrolysate (+ nutrients). No organism appeared to grow well in 25% hydrolysate, although 15 different types of organisms were isolated which survived a 3-day exposure. These are being further characterized. Further expansion of sites and enrichment methods are planned. Promising candidates will be tested for ethanol tolerance, use of individual sugars, growth characteristics, fermentation products, and ribosomal RNA similarity.

Publications and Presentations: None

Summary Date: March 2000

Improved Recombinant Zymomonas mobilis for Efficient and Rapid Fermentation of Hexose and Pentose Sugars

Research Funded by: U.S. Department of Energy Office of Fuels Development through the National Renewable Energy Laboratory

Project Manager: Robert Wooley 303.384.6825, Robert_Wooley@nrel.gov

Performing Organization: Unisearch Ltd., University of New South Wales, Sydney New South Wales 2052, Australia

Principal Investigator: Peter L. Rogers 61 (2) 313 6710, P.Rogers@unsw.edu.au

Contract Number: XDH-9-29055-01

Contract Period: 09/99–08/00

Contract Funding:

FY 1999: $109,800

Objectives: Our objective is to develop improved pentose-fermenting Zymomonas mobilis that can efficiently and rapidly convert sugars produced from lignocellulosic biomass to ethanol.

Approach/Background: The aims of the project are as follows:

• to enhance the acid tolerance of Z. mobilis recombinant strains via the use of mutagenesis techniques and continuous adaptation in continuous culture;
• to enhance process robustness of recombinant strains of Z. mobilis for use in high- productivity cell recycle continuous culture using glucose/xylose mixtures and lignocellulosic hydrolysates.

Status/Accomplishments: The project began in mid-September 1999 following signing of the subcontract, and it continues in FY 2000. Preliminary studies have been carried out in the cell recycle system with ZM4 (pZB5) and it is planned to evaluate an integrant strain of Z. mobilis following signing of an MTA with NREL.

Publications and Presentations: None

Summary Date: March 2000