Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jin, Y.-S. Articles by Stephanopoulos, G. Search for Related Content PubMed PubMed Citation Articles by Jin, Y.-S. Articles by Stephanopoulos, G. Agricola Articles by Jin, Y.-S. Articles by Stephanopoulos, G.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, December 2005, p. 8249-8256, Vol. 71, No. 12
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.12.8249-8256.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Improvement of Xylose Uptake and Ethanol Production in Recombinant Saccharomyces cerevisiae through an Inverse Metabolic Engineering Approach

Yong-Su Jin, Hal Alper, Yea-Tyng Yang, and Gregory Stephanopoulos^*

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

Received 1 July 2005/ Accepted 30 August 2005

Top Abstract Introduction Materials and Methods Results Discussion References

 
We used an inverse metabolic engineering approach to identify gene targets for improved xylose assimilation in recombinant Saccharomyces cerevisiae. Specifically, we created a genomic fragment library from Pichia stipitis and introduced it into recombinant S. cerevisiae expressing XYL1 and XYL2. Through serial subculturing enrichment of the transformant library, 16 transformants were identified and confirmed to have a higher growth rate on xylose. Sequencing of the 16 plasmids isolated from these transformants revealed that the majority of the inserts (10 of 16) contained the XYL3 gene, thus confirming the previous finding that XYL3 is the consensus target for increasing xylose assimilation. Following a sequential search for gene targets, we repeated the complementation enrichment process in a XYL1 XYL2 XYL3 background and identified 15 fast-growing transformants, all of which harbored the same plasmid. This plasmid contained an open reading frame (ORF) designated PsTAL1 based on a high level of homology with S. cerevisiae TAL1. To further investigate whether the newly identified PsTAL1 ORF is responsible for the enhanced-growth phenotype, we constructed an expression cassette containing the PsTAL1 ORF under the control of a constitutive promoter and transformed it into an S. cerevisiae recombinant expressing XYL1, XYL2, and XYL3. The resulting recombinant strain exhibited a 100% increase in the growth rate and a 70% increase in ethanol production (0.033 versus 0.019 g ethanol/g cells · h) on xylose compared to the parental strain. Interestingly, overexpression of PsTAL1 did not cause growth inhibition when cells were grown on glucose, unlike overexpression of the ScTAL1 gene. These results suggest that PsTAL1 is a better gene target for engineering of the pentose phosphate pathway in recombinant S. cerevisiae.

Top Abstract Introduction Materials and Methods Results Discussion References

 
An important challenge for metabolic engineering is identification of gene targets that have material importance for improvement of cellular function. Advances in molecular biology and genome sequencing have not been matched by equally effective methods for identifying gene targets, which still remains a daunting task lacking a defined search strategy. This problem is compounded by the lack of comprehensive and accurate metabolic models that capture both reaction kinetics and genetic regulation. An alternative method for introducing cellular perturbations to identify targets, termed inverse metabolic engineering (5), takes advantage of recent advances in high-throughput screening and genome sequencing. In this method a perturbation library is first screened for a desirable phenotype, and the genetic modifications responsible for the cellular response are traced by sequencing. The main objective of the inverse approach is to identify targets which, following modification, elicit a desired phenotype rather than randomly evolve a high-product-titer strain with mutations that are hardly traceable. Moreover, recent advances in genomics technologies allow this process to proceed via high-throughput screening (4, 9) and other, microarray-based methods, such as parallel gene trait mapping (8).

Previously, we used a stoichiometric model in a sequential approach for systematic identification of multigene targets (1, 2). In this paper, we further explore the utility of a sequential search through an inverse approach to identify gene targets that can improve xylose fermentation by recombinant Saccharomyces cerevisiae.

To date, most efforts in the engineering of S. cerevisiae for xylose fermentation have focused on manipulation of the initial xylose metabolic pathway by borrowing genes from Pichia stipitis in order to reconstruct an efficient xylose assimilation pathway in S. cerevisiae (15, 19, 28, 29). Since wild-type S. cerevisiae is able to grow and ferment xylulose as a sole carbon source, the XYL1 and XYL2 genes, coding for xylose reductase and xylitol dehydrogenase, respectively, which convert xylose into xylulose by sequential reactions, were the first genes chosen to be expressed in S. cerevisiae. However, recombinant S. cerevisiae expressing XYL1 and XYL2 suffered from inefficient xylose assimilation, while subsequent xylitol accumulation further limited ethanol production (13). Although increased xylulokinase activity, either due to overexpression of the XKS1 gene (coding for endogenous xylulokinase) or due to modulated expression of XYL3 (coding for heterologous xylulokinase), reduced xylitol accumulation and improved ethanol production (11, 16, 17), the performance of recombinant strains is still inferior to that of native xylose-fermenting yeasts, such as the source of the genes, P. stipitis (12).

In this study, we used inverse metabolic engineering to identify factors that limit xylose fermentation in S. cerevisiae. This was attempted by complementing S. cerevisiae with genomic DNA fragments of P. stipitis, as the latter yeast has pathways that are specific to xylose utilization. Through the use of the genes identified in this report (XYL3 and PsTAL1), we were able to improve xylose uptake and ethanol production in S. cerevisiae without the negative effects observed with the use of other pentose phosphate pathway enzymes (ScTAL1 and ScTKL1), such as severe inhibition of growth on glucose (21, 27). Following identification of PsTAL1 as a key gene target, we also observed that overexpression of this gene resulted in almost negligible inhibition of growth on glucose, unlike overexpression of ScTAL1, suggesting that PsTAL1 is a better target for overexpression than ScTAL1 for engineering S. cerevisiae strains capable of rapid fermentation of glucose-xylose mixtures.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Strains and plasmids.
Microbial strains and plasmids used in this study are listed in Table 1. S. cerevisiae YS1020 (13) and YSX3 (= NRRL Y-30602) (16) were provided by Thomas W. Jeffries of the University of Wisconsin-Madison. Escherichia coli DH5 (F^– recA1 endA1 hsdR17 [r_K^– m_K⁺] supE44 thi-1 gyrA relA1) (Invitrogen, Gaithersburg, MD) was routinely used for gene cloning and manipulation. For library amplification, strain DH10B [F^– mcrA (mrr-hadRMS-mcrBC)80dlacZ lacX74 recA1 endA1 deoR (ara, leu)7697 araD139 galU galK mupG rpsL ^–) (Invitrogen, Gaithersburg, MD) was used.

View this table: [in this window] [in a new window]

TABLE 1. Strains and plasmids used in this study

Media and culture conditions.
Yeast and bacterial strains were stored in 15% glycerol at –70°C. E. coli was grown in Luria-Bertani medium; 50 µg/ml of ampicillin was added to the medium when required. Yeast strains were routinely cultivated at 30°C in YP medium (10 g/liter yeast extract, 20 g/liter Bacto peptone) with 20 g/liter glucose. To select transformants using a TRP1 selectable marker, yeast synthetic complete (YSC) medium was used, which contained 6.7 g/liter yeast nitrogen base plus 20 g/liter glucose, 20 g/liter agar, and CSM-Leu-Trp-Ura (Bio 101, Vista, CA), which supplied appropriate nucleotides and amino acids. For measurement of the growth rate under aerobic conditions, cells were grown in 50 ml of medium in a 250-ml flask. For ethanol fermentation under oxygen-limited conditions, cells were grown in 20 ml of medium in a 50-ml flask.

Construction of P. stipitis genomic library.
P. stipitis genomic DNA was isolated from strain CBS 5774. The genomic DNA was partially digested with restriction enzyme Sau3A, and then DNA fragments in the size range from 2 to 8 kbp were isolated following gel electrophoresis. The isolated DNA fragments were ligated into the BamHI site of the pRS424 plasmid (25). The resulting ligation mixture was transformed into ElectroMax DH10B (Invitrogen, Gaithersburg, MD). Based on colony counting after dilution on an X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside) plate (24), more than 10⁵ colonies were rescued for generation of the library, and 85% of them contained inserts. The quality of the library that was constructed was also checked by performing a PCR experiment. We were able to amplify several known P. stipitis genes (XYL1, XYL2, and XYL3) using the library that was constructed as a template.

Yeast transformation.
A spheroplast transformation kit (Bio 101, Vista, CA) and Frozen-EZ yeast transformation (Zymo Research, Orange, CA) were used for yeast transformations. Transformants were selected on YSC medium containing 20 g/liter glucose. Amino acids and nucleotides were added as necessary.

Serial transfer and screening of fast-growing transformants.
Yeast cells were cultivated at 30°C in 50 ml of medium in a 250-ml Erlenmeyer flask during serial transfer of the culture medium. To enrich fast-growing transformants, cultures of transformants were inoculated into new culture medium using a 1% inoculum. After 10 serial transfers, cells from the final culture were plated onto YSC medium with glucose or xylose as a carbon source. Large colonies on the xylose plate were picked and inoculated into 200 µl of YSC medium with glucose in a 96-well plate for storage at –70°C. For quantitative comparison of growth on xylose, 30-µl portions of cell suspensions were drawn into each syringe of a Hydra machine (Robbins Scientific, Sunnyvale, CA), and 1-µl samples were spotted onto a YSC-xylose agar plate (Omni plate; Nalgene, Rochester, NY). After the plates were incubated for 5 days, images of the plates were captured for a comparison of growth.

Insert identification and sequence analysis.
Plasmids were isolated from selected colonies by using Zymoprep (Zymoresearch, Orange, CA) and subsequently transformed into the E. coli DH5 strain for sequencing. Sequence analysis was performed using the DNA Star software (DNASTAR, Madison, WI). Homologues of transaldolase in yeast were identified using a BLAST search (http://www.ncbi.nlm.nih.gov/BLAST/), and then sequences were aligned using ClustalW (http://www.ebi.ac.uk/clustalw/). A phylogenetic tree was generated from the aligned region with bootstrapping. The quality of sequences from the inserts identified was checked by comparing them with the draft sequence of the P. stipitis genome (T. W. Jeffries, personal communication).

Enzymes, primers, and chemicals.
Restriction enzymes, DNA-modifying enzymes, and other molecular reagents were obtained from New England Biolabs (Beverly, MA), Promega (Madison, WI), Stratagene (La Jolla, CA), and Roche Biochemical (Indianapolis, IN). The reaction conditions were those recommended by the suppliers. All general chemicals were purchased from Sigma (St. Louis, MO). Primers for PCR and sequencing were synthesized by Integrated DNA Technologies (Coralville, IA) and Invitrogen (Carlsbad, CA).

Plasmid construction.
Plasmids used in this study are described in Table 1. For creation of a P. stipitis TAL1 overexpression vector, the PsTAL1 open reading frame was amplified from the isolated pTAL1I plasmid with adjacent restriction enzyme sites (SpeI and XhoI) by PCR using primers HA026 (5'-GGACTAGTATGTCCTCCAACTCCCTTGA-3') and HA027 (CCGCTCGAGTTATTAATAGAAGCGAAATA-3'). The amplified SpeI-XhoI fragment then ligated into the SpeI-XhoI sites of pRS424TEF (22), which contained the TEF promoter and CYC1 terminator. For construction of the ScTAL1 overexpression cassette, the open reading frame of ScTAL1 was amplified using primers JIN156 (5'-GGCACTAGTATGTCTGAACCAGCTCAAAA-3') and JIN157 (5'-GGCCTCGAGTTAAGCGGTAACTTTCTTTT-3') and also cloned into the SpeI-XhoI sites of pRSTEF.

Preparation of crude extract and enzyme assay.
PsTAL1-overexpressing S. cerevisiae was grown to the exponential phase in YSC medium supplemented with appropriate amino acids and nucleotides and 20 g/liter of glucose. Cells were harvested by centrifugation. The pellet was washed and suspended in buffer (100 mM of phosphate buffer, 1 mM EDTA, 5 mM ß-mercaptoethanol; pH 7.0). The suspended cells were mixed with glass beads (Sigma, St. Louis, MO), vortexed at the maximum rate by using 30- to 120-s bursts, and then cooled on ice for a similar period. This procedure was repeated for up to 10 min of vortexing with periodic microscopic examination to determine cell breakage. The crude extract, collected after centrifugation for 10 min at 15,000 x g, was used for the enzyme assay. Transaldolase activity was measured by monitoring NADH oxidation at 340 nm as previously described (6). One unit of activity was defined as the amount of enzyme that converted 1 µmol of fructose-6-phosphate and erythrose-4-phosphate into glyceraldehyde-3-phosphate and sedoheoptulose-7-phosphate per min at 30°C. The protein concentration was determined by the bicinchoninic acid method (Pierce, Rockford, IL).

Analytical methods.
Glucose, xylose, and xylitol concentrations were determined by high-performance liquid chromatography (Hewlett-Packard, Wilmington, DE) with an Aminex HPX-87H column (Bio-Rad, Hercules, CA). The ethanol concentration was determined with an enzymatic assay kit (R-Biopharm, South Marshall, Mich.). Cell growth was monitored by determining the optical density at 600 nm (OD₆₀₀). One OD₆₀₀ unit was equivalent to 0.17 g cells/liter for S. cerevisiae.

Nucleotide sequence accession number.
The sequence of the coding region and the 5' and 3' flanking regions has been deposited in the GenBank database under accession number AY854959.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Enrichment of fast-growing transformants by serial transfer on xylose minimal medium.
We attempted to identify P. stipitis genes that enhance xylose utilization in recombinant S. cereviaiae by constructing a genomic library using P. stipitis genomic DNA fragments ranging in size from 2 to 8 kbp on a multicopy plasmid (pRS424). The resulting library was then transformed into the S. cerevisiae YS1020 recombinant strain containing integrated copies of XYL1 and XYL2 in the chromosome (13). Based on the hypothesis that genes which increase xylose assimilation likewise increase the growth rate of transformants on xylose, we performed serial subculturing to enrich the population. After 10 transfers on xylose minimal medium, cells were plated onto agar plates containing either glucose or xylose to check if the expected enrichment had occurred. As shown in Fig. 1, the colony sizes on the glucose plates were fairly homogeneous. However, the colony sizes on the xylose plates varied greatly. This result suggested that transformants which grew faster on xylose were enriched during the serial transfer.

View larger version (20K): [in this window] [in a new window]

FIG. 1. Distributions of colony sizes for the enriched populations after series of serial transfers on YSC agar plates with glucose and xylose. On a glucose plate (A), the colony sizes of the enriched population (total number of colonies counted, 351) showed a uniform distribution, with a mean area of 4.0 mm2. However, on a xylose plate (B), the colony sizes of the enriched population (total number of colonies counted, 1,191) showed a distribution that can be described as a mixed population with two different means, 1.5 and 4.0 mm2.

Identification of P. stipitis genes enhancing growth of recombinant S. cerevisiae expressing XYL1 and XYL2.
In order to characterize the P. stipitis genes present in the fast-growing transformants on the xylose plates, we picked 204 colonies that were significantly larger (average size, 4.5 mm²) than the background colonies (average size, 1.6 mm²) (Fig. 1). These colonies were assayed on xylose plates to filter out false-positive colonies which might have been included in the previous screening. For this growth assay, transformants were cultured in a 96-well plate and then spotted onto an agar plate containing xylose as the sole carbon source. As shown in Fig. 2, fast-growing cells on xylose were easily identified by the 96-well format spotted growth assay. Replicate growth assays were performed with different amounts of xylose in the plates, and all conditions yielded similar results (Fig. 2). Sixteen of the largest colonies were selected for further characterization, and their plasmids were isolated and sequenced. The insert sizes ranged from 2.0 kbp to 5.1 kbp. A sequence analysis and BLAST search identified four independent inserts (Table 2). Of the 16 plasmids isolated, 10 contained PsXYL3, while 1 plasmid contained an insert that exhibited a high level of homology with the sequence of S. cerevisiae TAL1. These results confirmed previous findings which indicated that an enhanced f D-xyluokinase activity, encoded by either XYL3 or XKS1, increases the efficiency of xylose assimilation in recombinant S. cerevisiae (11, 16, 17). Enhancement of the growth of recombinant S. cerevisiae in xylose media after overexpression of endogenous S. cerevisiae TAL1 has also been reported previously (29). Three additional plasmids contained inserts, but the BLAST search did not identify similarities to any known protein sequences. It is worth noting that the same unknown insert was recovered from two different transformants. While it is conceivable that the unknown inserts may actually code for a regulatory factor, we could not exclude the possibility of adaptation or accumulated chromosomal accumulations which benefited xylose utilization, since we also observed two fast-growing transformants that contained the control plasmids without inserts. To eliminate the possibility of cellular adaptation, we transformed the isolated plasmids into the parental strain again and observed the same increased-growth phenotype (data not shown).

View larger version (17K): [in this window] [in a new window]

FIG. 2. Growth assay for the 204 colonies screened on a YSC agar plate with xylose. Cells grown on glucose in a 96-well plate were spotted onto an agar plate with xylose. Replicate growth assays were performed (plates B and D are replicates of plates A and C, respectively). For each plate the image was captured and visualized in black and white for easy comparison.

View this table: [in this window] [in a new window]

TABLE 2. Isolated DNA fragments that allowed recombinant S. cerevisiae YS1020 to grow faster on xylose

Identification of P. stipitis gene enhancing growth of recombinant S. cerevisiae expressing XYL1, XYL2, and XYL3.
Although we identified four independent inserts, we regarded the XYL3 gene the consensus target in an XYL1 XYL2 overexpression genetic background because more than 60% (10 of 16) of the plasmids isolated contained this gene. We then extended our search for gene targets by conducting a new search in the genetic background, incorporating the modification suggested by the previous consensus target, XYL3. To do this, we used the S. cerevisiae YSX3 strain, which optimally expresses XYL3 in the XYL1 XYL2 overexpression background (16). Following transformation of this strain with the same P. stipitis genomic library used previously, a faster enrichment process was employed, in which the amount of culture medium transferred was gradually reduced from 1% to 0.1% during the serial transfer. After 10 transfers, cells were plated on a minimal medium plate containing xylose as the sole carbon source. Unlike the previous enrichment, the colony sizes were fairly homogeneous. Fifteen colonies were randomly picked, and plasmids were isolated from them to identify the inserts. All of the plasmids isolated from these colonies harbored the same 2.0-kbp fragment containing the PsTAL1 sequence identified in the previous search. The isolated plasmid was designated pTAL1I and used for further characterization.

Sequence analysis of the identified fragment containing PsTAL1.
Since the PsTAL1 gene is approximately 1 kbp long, the entire insert of pTAL1I was fully sequenced to possibly identify unknown open reading frames other than PsTAL1. There was one other meaningful, but truncated, open reading frame, which coded for 193 amino acids, in the insert. NCBI Conserved Domain (20) and BLAST searches (3) with the 193-amino-acid sequence revealed good homology (96 of 199 amino acids; 48% identity) with the sequence of S. cerevisiae riboflavin kinase (Fmn1p). The organization of open reading frames in the 2.0-kbp fragment is shown in Fig. 3A. The PsTAL1 sequence contained the complete open reading frame consisting of 972 nucleotides coding for 323 amino acids. As shown in Fig. 3A, the PsTAL1 sequence identified is likely to be transcribed from a rather short promoter (about 100 bp) which is truncated during library construction. A phylogenetic analysis revealed that the P. stipitis transaldolase protein had diverged significantly from those of Saccharomyces species. Only 71% of the amino acid sequence encoded by the translated PsTAL1 was identical to the homologue (ScTAL1) from S. cerevisiae. As shown in Fig. 3B, the transaldolase from Candida albicans is the most similar transaldolase at the protein sequence level. It is interesting that the amino acid sequences of transaldolases from all of xylose-fermenting yeasts (Pichia, Candida, and Debaryomyces species) are well conserved, although there were a couple of major evolutionary events which resulted in scattering in the hemiascomycete phylogenetic tree based on 25S rRNA gene sequences (7).

View larger version (22K): [in this window] [in a new window]

FIG. 3. Location of PsTAL1 in the isolated plasmid, pTAL1I (A), and phylogenetic tree of transaldoase proteins found in the GenBank database. The homologues of transaldolase were identified by a BLAST search, and the protein sequences found were then aligned by CLUSTALW. The tree was generated from the aligned region with bootstrapping. aa, amino acids; ORF, open reading frame.

Overexpression of PsTAL1 in the engineered S. cerevisiae YSX3 strain is beneficial for xylose assimilation and neutral on growth on glucose, unlike overexpression of ScTAL1.
To investigate the impact of PsTAL1 overexpression, we cloned the PsTAL1 open reading frame from pTAL1I into pRS424TEF, in which PsTAL1 could be expressed under control of a constitutive promoter. The resulting plasmid, pTAL1M, was transformed into the S. cerevisiae YSX3 strain. First, we tested the in vitro enzymatic activities of transaldolase in the YSX3-C and YSX3-TAL1M strains. The YSX3-TAL1M strain exhibited transaldolase activity that was 14 times higher (1.746 versus 0.124 U/mg). Second, we determined the growth rate of the transformant with pTAL1M on xylose to compare it with the growth rate of the transformant with the isolated plasmid, pTAL1I. Table 3 summarizes the specific growth rates of the transformants on glucose and xylose. In glucose medium, there was no significant difference between the specific growth rates of the transformants and the specific growth rate of the parental strain transformed with a control plasmid. However, in xylose medium the specific growth rates of the transformants with pTAL1I and pTAL1M were both doubled compared to that of the parental strain. These results confirmed that overexpression of PsTAL1 was responsible for the rapid-growth phenotype of the S. cerevisiae strain isolated. Overexpression of ScTAL1 in the same strain background with an identical construct except for the open reading frame resulted in a 25% decrease in the growth rate on glucose and only a 43% increase in the growth rate on xylose (Table 3).

View this table: [in this window] [in a new window]

TABLE 3. Comparison of specific growth rates of engineered S. cerevisiae strains on minimal medium with glucose and xylose as sole carbon sources

Overexpression of PsTAL1 dramatically improves xylose uptake, cell growth, and ethanol production in engineered S. cerevisiae.
The profiles of xylose consumption and cell growth for the PsTAL1-overexpressing strain (YSX3-TAL1M) and its parental strain with a control plasmid (YSX3-C) were compared in 50 ml of minimal medium containing either 20 or 40 g/liter of xylose in 250-ml flasks shaken at 200 rpm. Figure 4 shows the profiles for cell mass, xylose consumption, and xylitol production for the YSX3-C and YSX3-TAL1M strains. No ethanol production was observed under these conditions. Apparently, neither strain was able to consume all of the xylose in the experiment with 40 g/liter of xylose. However, with 20 g/liter of xylose, YSX3-TAL1M was able to consume most of the added xylose within 120 h, while YSX3-C was able to consume only one-half that amount in the same time. Table 4 summarizes the xylose consumption and product yields for both strains at 72 h. The YSX3-TAL1 strain consumed more than twice as much xylose as the YSX3-C strain with 20 g/liter and 40 g/liter of xylose (2.6- and 2.8-fold more, respectively). To investigate the effect of PsTAL1 overexpression on ethanol production from xylose, the YSX3-C and YSX3-TAL1M strains were first grown on the minimal medium with 20 g/liter of glucose under aerobic conditions. Cells were harvested and inoculated again into 20 ml of xylose minimal medium with 40 g/liter of xylose in a 50-ml flask for a fermentation study. The initial OD₆₀₀ of these cultures were 12 to 15, and the ethanol concentrations were monitored (Fig. 5). The specific ethanol productivity of the YSX3-TAL1 strain was 70% higher than that of the YSX3-C strain (0.019 versus 0.033 g ethanol/g cell · h).

View larger version (23K): [in this window] [in a new window]

FIG. 4. Xylose uptake and cell growth profile for S. cerevisiae strains YSX3-C (A and C) and YSX3-TAL1M (B and D). Cells were cultured in 50 ml of YSC medium with 20 g/liter of xylose (A and B) and 40 g/liter of xylose (C and D). The data are the averages of two replicate experiments. Symbols: , cell mass; , xylose; , xylitol.

View this table: [in this window] [in a new window]

TABLE 4. Comparison of xylose consumption and product yields (cells, xylitol, and ethanol) under aerobic conditions after 72 ha

View larger version (15K): [in this window] [in a new window]

FIG. 5. Comparison of ethanol production by S. cerevisiae strains YSX3-C () and YSX4-TAL1M () under oxygen-limited conditions. Cells were grown on glucose, harvested, and inoculated into 20 ml of YSC medium with 40 g/liter of xylose. The initial OD600 was about 12 to 15. EtOH, ethanol.

Top Abstract Introduction Materials and Methods Results Discussion References

 
We used multicopy cloning of P. stipitis genomic DNA fragments and enrichment of a desired population to identify gene targets whose overexpression is responsible for improved xylose assimilation. Since the cloned DNA fragments were present in a multicopy plasmid, we expected not only overexpression of target genes in the plasmid but also titration of DNA binding sites, which provided another level of perturbation. Actually, two of the four independent inserts screened from the first enrichment experiment did not contain a meaningful open reading frame (more than 200 amino acids). These results illustrate that multicopy cloning is capable of identifying unknown interactions, as well as straightforward overexpression of a gene.

The two enrichment studies described in this paper showed that an iterative-sequential search is a practical way of finding multiple gene targets. XYL3 was the consensus target in the first screening, although we also identified other unknown inserts and PsTAL1. In the next round of the sequential search in the XYL3 expression background, we found only PsTAL1. The type of enrichment strategy is also important, as it determines the resolution of gene targets obtained from the search. Whereas we were able to find multiple targets after the first screening, in which we used a rather liberal enrichment procedure, only a single target was found after the second screening, in which a much stricter enrichment procedure was used. Thus, one could propose a generous enrichment procedure at the early stage of screening to ensure that more targets are included, followed by more stringent enrichment procedures in subsequent rounds to reduce false positives. In another example, we also concluded that an iterative-sequential search was effective for identification of multiple knockout targets for improvement of lycopene production in E. coli (1).

Another topic of interest is the toxicity of XYL3 overexpression in cells grown on xylose, as observed in previous studies (16, 17, 23). This toxicity was not observed with use of the isolated plasmid containing the XYL3 open reading frame. According to sequence analysis of the insert containing XYL3, about 500 bp upstream of the start codon of the XYL3 open reading frame was included in the plasmid. A previous study reported mild toxicity when a fragment containing the 900-bp upstream region from the start codon was used (16). This suggests that there is a chance that the toxicity problem can be avoided by having a partially active promoter activity caused by the truncated promoter. The selection pressure or screening procedure should eliminate strains having suboptimal expression of the target gene.

Improvement of xylose utilization in engineered S. cerevisiae by overexpression of the ScTAL1 gene was reported previously. However, the results were inconsistent and varied with the background of the strain. Walfridsson et al. (29) reported significant improvement of growth after overexpression of ScTAL1 in the XYL1 XYL2 background. However, Johansson and Hahn-Hägerdal (18) reported that overexpression of ScTAL1 did not influence the xylose fermentation rate or the specific growth rate in the TMB3001 strain, which contains XYL1 and XYL2 and overexpresses endogeneous XKS1. We clearly observed the beneficial effect of PsTAL1 overexpression in engineered S. cerevisiae both in terms of growth and in terms of ethanol production from xylose (Fig. 4 and Table 3), unlike the results of Johansson and Hahn-Hägerdal. There are two differences between our engineered strain and the strain of these workers. First, the parental strains were different, which may have affected the engineered phenotype. Second, we constructed a functional xylose metabolic pathway by using only heterologous genes (XYL1, XYL2, XYL3, and TAL1) originating from P. stipitis. It is conceivable that a series of metabolic enzymes that evolved together could be more effective than a combination of enzymes from different sources. Actually, the protein sequences encoded by XYL3 and TAL1 from P. stipitis and S. cerevisiae are very different (14) (Fig. 3B), suggesting that there are differences in their kinetic and regulatory properties. Therefore, the effects of overexpression of these genes on the phenotype could be different. For instance, the growth inhibition due to overexpression of PsTAL1 on glucose was negligible compared to that due to overexpression of ScTAL1. Growth inhibition on glucose by amplification of pentose phosphate pathway enzyme genes (ScTAL1 and ScTKL1) has been reported previously. Sundstrom et al. (27) observed that growth on glucose is reduced in recombinant S. cerevisiae overexpressing ScTKL1, and Meinander et al. (21) also reported a significant decrease in the growth rate on glucose after overexpression of ScTAL in the XYL1 XYL2 background. They observed an almost 60% decrease in the specific growth rate on glucose after overexpression of ScTAL1 (Table 5). We observed a similar detrimental effect of ScTAL1 overexpression, but the level of the growth inhibition due to ScTAL1 overexpression was lower than the level that Meinander et al. observed (25% versus 60%). This may have been due to the fact that our overexpression construct had a much lower transaldolase specific activity than their construct. An interesting point is that we did not observe such a toxic effect for PsTAL1 overexpression, even though the specific activity of transaldolase due to PsTAL1 overexpression was very high (Table 5).

View this table: [in this window] [in a new window]

TABLE 5. Comparison of specific activities of transaldolase and specific growth rates on glucose for recombinant S. cerevisiae strains on minimal medium with glucose and xylose as sole carbon sources

Overexpression of the P. stipitis TAL1 gene in engineered S. cerevisiae resulted in a 100% increase in the specific growth rate on xylose but only a 70% increase in ethanol production. It is possible that the difference was due to the aerobic conditions employed for enrichment of the fast-growing cells rather than due to screening better-fermenting cells under oxygen-limited conditions. Apparently, cell growth and ethanol production in yeast are conflicting phenotypes. Therefore, another enrichment procedure may be required to identify gene targets that specifically limit ethanol production under oxygen-limited conditions (26).

Through the examples illustrated in this study, we demonstrated the ability of inverse metabolic engineering to identify gene targets responsible for a particular metabolic phenotype, which often depends on changes in multiple genes. Two main targets identified in this study, XYL3 and PsTAL1, confirmed previous findings and suggested a new direction for further engineering of S. cerevisiae for xylose fermentation. Especially, the newly cloned PsTAL1 gene was found to be a better target than ScTAL1 for overexpression, since it had a less inhibitory effect when cells were cultured on glucose. Two obstacles limited our search. First, the maximum number of heterologous gene expression is limited by the number of available auxotrophic markers. Second, the absence of genome sequence information impeded the insert characterization process. Recently developed molecular genetics tools which allow rescue of markers for multiple rounds of genetic modification (10) and the upcoming genome sequence of P. stipitis (Jeffries, personal communication) should increase the feasibility and throughput of engineering of recombinant S. cerevisiae as described in this study.

High-throughput phenotyping and screening increase the possibility of using inverse metabolic engineering for systems other than those which have defined selection pressure (for example, growth advantage). In order to employ such search techniques, experimental tools and search methodologies must be enhanced. The most critical component impeding the process appears to be the development of effective methods for selecting individual cells with desired phenotypes. Advances in this area should further increase the prospect of using inverse metabolic engineering for industrially relevant problems.

 
We acknowledge support for this work provided by the DuPont-MIT Alliance and the Singapore-MIT Alliance.

 
* Corresponding author. Mailing address: Department of Chemical Engineering, Massachusetts Institute of Technology, Room 56-469, Cambridge, MA 02139. Phone: (617) 253-4583. Fax: (617) 253-3122. E-mail: gregstep{at}mit.edu.

Y.-S.J. and H.A. contributed equally to this work.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Alper, H., Y. S. Jin, J. F. Moxley, and G. Stephanopoulos. 2005. Identifying gene targets for the metabolic engineering of lycopene biosynthesis in Escherichia coli. Metab. Eng. 7:155-164.[CrossRef][Medline]
2
2. Alper, H., K. Miyaoku, and G. Stephanopoulos. 2005. Construction of lycopene-overproducing E. coli strains by combining systematic and combinatorial gene knockout targets. Nat. Biotechnol. 23:612-616.[CrossRef][Medline]
3
3. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.[CrossRef][Medline]
4
4. Badarinarayana, V., P. W. Estep, 3rd, J. Shendure, J. Edwards, S. Tavazoie, F. Lam, and G. M. Church. 2001. Selection analyses of insertional mutants using subgenic-resolution arrays. Nat. Biotechnol. 19:1060-1065.[CrossRef][Medline]
5
5. Bailey, J. E., A. Sburlati, V. Hatzimanikatis, K. Lee, W. A. Renner, and P. S. Tsai. 2002. Inverse metabolic engineering: a strategy for directed genetic engineering of useful phenotypes. Biotechnol. Bioeng. 79:568-579.[CrossRef][Medline]
6
6. Bergmeyer, H.-U. (ed.). 1974. Methods of enzymatic analysis. Academic Press, New York, N.Y.
7
7. Dujon, B., D. Sherman, G. Fischer, P. Durrens, S. Casaregola, I. Lafontaine, J. De Montigny, C. Marck, C. Neuveglise, E. Talla, N. Goffard, L. Frangeul, M. Aigle, V. Anthouard, A. Babour, V. Barbe, S. Barnay, S. Blanchin, J. M. Beckerich, E. Beyne, C. Bleykasten, A. Boisrame, J. Boyer, L. Cattolico, F. Confanioleri, A. De Daruvar, L. Despons, E. Fabre, C. Fairhead, H. Ferry-Dumazet, A. Groppi, F. Hantraye, C. Hennequin, N. Jauniaux, P. Joyet, R. Kachouri, A. Kerrest, R. Koszul, M. Lemaire, I. Lesur, L. Ma, H. Muller, J. M. Nicaud, M. Nikolski, S. Oztas, O. Ozier-Kalogeropoulos, S. Pellenz, S. Potier, G. F. Richard, M. L. Straub, A. Suleau, D. Swennen, F. Tekaia, M. Wesolowski-Louvel, E. Westhof, B. Wirth, M. Zeniou-Meyer, I. Zivanovic, M. Bolotin-Fukuhara, A. Thierry, C. Bouchier, B. Caudron, C. Scarpelli, C. Gaillardin, J. Weissenbach, P. Wincker, and J. L. Souciet. 2004. Genome evolution in yeasts. Nature 430:35-44.[CrossRef][Medline]
8
8. Gill, R. T. 2003. Enabling inverse metabolic engineering through genomics. Curr. Opin. Biotechnol. 14:484-490.[CrossRef][Medline]
9
9. Gill, R. T., S. Wildt, Y. T. Yang, S. Ziesman, and G. Stephanopoulos. 2002. Genome-wide screening for trait conferring genes using DNA microarrays. Proc. Natl. Acad. Sci. USA 99:7033-7038.[Abstract/Free Full Text]
10
10. Guldener, U., S. Heck, T. Fielder, J. Beinhauer, and J. H. Hegemann. 1996. A new efficient gene disruption cassette for repeated use in budding yeast. Nucleic Acids Res. 24:2519-2524.[Abstract/Free Full Text]
11
11. Ho, N. W. Y., Z. D. Chen, and A. P. Brainard. 1998. Genetically engineered Sacccharomyces yeast capable of effective cofermentation of glucose and xylose. Appl. Environ. Microbiol. 64:1852-1859.[Abstract/Free Full Text]
12
12. Jeffries, T. W., and Y. S. Jin. 2004. Metabolic engineering for improved fermentation of pentoses by yeasts. Appl. Microbiol. Biotechnol. 63:495-509.[CrossRef][Medline]
13
13. Jin, Y. S., and T. W. Jeffries. 2003. Changing flux of xylose metabolites by altering expression of xylose reductase and xylitol dehydrogenase in recombinant Saccharomyces cerevisiae. Appl. Biochem. Biotechnol. 105-108:277-285.
14
14. Jin, Y. S., S. Jones, N. Q. Shi, and T. W. Jeffries. 2002. Molecular cloning of XYL3 (D-xylulokinase) from Pichia stipitis and characterization of its physiological function. Appl. Environ. Microbiol. 68:1232-1239.[Abstract/Free Full Text]
15
15. Jin, Y. S., T. H. Lee, Y. D. Chio, Y. W. Ryu, and J. H. Seo. 2000. Conversion of xylose to ethanol by recombinant Saccharomyces cerevisiae containing genes coding for xylose reductase and xylitol dehydrogenase from Pichia stiptis. J. Microbiol. Biotechnol. 10:564-567.
16
16. Jin, Y. S., N. Ni, J. M. Laplaza, and T. W. Jeffries. 2003. Optimal growth and ethanol production from xylose by recombinant Saccharomyces cerevisiae require moderate D-xylulokinase activity. Appl. Environ. Microbiol. 69:495-503.[Abstract/Free Full Text]
17
17. Johansson, B., C. Christensson, T. Hobley, and B. Hahn-Hägerdal. 2001. Xylulokinase overexpression in two strains of Saccharomyces cerevisiae also expressing xylose reductase and xylitol dehydrogenase and its effect on fermentation of xylose and lignocellulosic hydrolysate. Appl. Environ. Microbiol. 67:4249-4255.[Abstract/Free Full Text]
18
18. Johansson, B., and B. Hahn-Hägerdal. 2002. The non-oxidative pentose phosphate pathway controls the fermentation rate of xylulose but not of xylose in Saccharomyces cerevisiae TMB3001. FEMS Yeast Res. 2:277-282.[Medline]
19
19. Kötter, P., and M. Ciriacy. 1993. Xylose fermentation by Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol. 38:776-783.[CrossRef]
20
20. Marchler-Bauer, A., J. B. Anderson, C. DeWeese-Scott, N. D. Fedorova, L. Y. Geer, S. He, D. I. Hurwitz, J. D. Jackson, A. R. Jacobs, C. J. Lanczycki, C. A. Liebert, C. Liu, T. Madej, G. H. Marchler, R. Mazumder, A. N. Nikolskaya, A. R. Panchenko, B. S. Rao, B. A. Shoemaker, V. Simonyan, J. S. Song, P. A. Thiessen, S. Vasudevan, Y. Wang, R. A. Yamashita, J. J. Yin, and S. H. Bryant. 2003. CDD: a curated Entrez database of conserved domain alignments. Nucleic Acids Res. 31:383-387.[Abstract/Free Full Text]
21
21. Meinander, N. Q., I. Boels, and B. Hahn-Hägerdal. 1999. Fermentation of xylose/glucose mixtures by metabolically engineered Saccharomyces cerevisiae strains expressing XYL1 and XYL3 from Pichia stipitis with and without overexpression of TAL1. Bioresour. Technol. 68:79-87.[CrossRef]
22
22. Mumberg, D., R. Muller, and M. Funk. 1995. Yeast vectors for the controlled expression of heterologous proteins in different genetic backgrounds. Gene 156:119-122.[CrossRef][Medline]
23
23. Rodriguez-Pena, J. M., V. J. Cid, J. Arroyo, and C. Nombela. 1998. The YGR194c (XKS1) gene encodes the xylulokinase from the budding yeast Saccharomyces cerevisiae. FEMS Microbiol. Lett. 162:155-160.[Medline]
24
24. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
25
25. Sikorski, R. S., and P. Hieter. 1989. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122:19-27.[Abstract/Free Full Text]
26
26. Stephanopoulos, G. 2002. Metabolic engineering by genome shuffling. Nat. Biotechnol. 20:666-668.[CrossRef][Medline]
27
27. Sundstrom, M., Y. Lindqvist, G. Schneider, U. Hellman, and H. Ronne. 1993. Yeast TKL1 gene encodes a transketolase that is required for efficient glycolysis and biosysnthesis of aromatic amino acids. J. Biol. Chem. 268:24346-24352.[Abstract/Free Full Text]
28
28. Tantirungkij, M., T. Izuishi, T. Seki, and T. Yoshida. 1994. Fed-batch fermentation of xylose by a fast-growing mutant of xylose-assimilating recombinant Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol. 41:8-12.
29
29. Walfridsson, M., J. Hallborn, M. Penttila, S. Keranen, and B. Hahn-Hagerdal. 1995. Xylose-metabolizing Saccharomyces cerevisiae strains overexpressing the TKL1 and TAL1 genes encoding the pentose phosphate pathway enzymes transketolase and transaldolase. Appl. Environ. Microbiol. 61:4184-4190.[Abstract]

---

Applied and Environmental Microbiology, December 2005, p. 8249-8256, Vol. 71, No. 12
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.12.8249-8256.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Nevoigt, E. (2008). Progress in Metabolic Engineering of Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 72: 379-412 [Abstract] [Full Text]  
• Guimaraes, P. M. R., Francois, J., Parrou, J. L., Teixeira, J. A., Domingues, L. (2008). Adaptive Evolution of a Lactose-Consuming Saccharomyces cerevisiae Recombinant. Appl. Environ. Microbiol. 74: 1748-1756 [Abstract] [Full Text]  
• Lu, C., Jeffries, T. (2007). Shuffling of Promoters for Multiple Genes To Optimize Xylose Fermentation in an Engineered Saccharomyces cerevisiae Strain. Appl. Environ. Microbiol. 73: 6072-6077 [Abstract] [Full Text]  
• Borden, J. R., Papoutsakis, E. T. (2007). Dynamics of Genomic-Library Enrichment and Identification of Solvent Tolerance Genes for Clostridium acetobutylicum. Appl. Environ. Microbiol. 73: 3061-3068 [Abstract] [Full Text]  
• Tyo, K. E., Zhou, H., Stephanopoulos, G. N. (2006). High-Throughput Screen for Poly-3-Hydroxybutyrate in Escherichia coli and Synechocystis sp. Strain PCC6803.. Appl. Environ. Microbiol. 72: 3412-3417 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jin, Y.-S. Articles by Stephanopoulos, G. Search for Related Content PubMed PubMed Citation Articles by Jin, Y.-S. Articles by Stephanopoulos, G. Agricola Articles by Jin, Y.-S. Articles by Stephanopoulos, G.