High level secretion of cellobiohydrolases by Saccharomyces cerevisiae
- Marja Ilmén1,
- Riaan den Haan2,
- Elena Brevnova3,
- John McBride3,
- Erin Wiswall3,
- Allan Froehlich3,
- Anu Koivula1,
- Sanni P Voutilainen1,
- Matti Siika-aho1,
- Daniël C la Grange2,
- Naomi Thorngren3,
- Simon Ahlgren3,
- Mark Mellon3,
- Kristen Deleault3,
- Vineet Rajgarhia3, 4,
- Willem H van Zyl2 and
- Merja Penttilä1Email author
© Ilmén et al; licensee BioMed Central Ltd. 2011
Received: 4 June 2011
Accepted: 12 September 2011
Published: 12 September 2011
The main technological impediment to widespread utilization of lignocellulose for the production of fuels and chemicals is the lack of low-cost technologies to overcome its recalcitrance. Organisms that hydrolyze lignocellulose and produce a valuable product such as ethanol at a high rate and titer could significantly reduce the costs of biomass conversion technologies, and will allow separate conversion steps to be combined in a consolidated bioprocess (CBP). Development of Saccharomyces cerevisiae for CBP requires the high level secretion of cellulases, particularly cellobiohydrolases.
We expressed various cellobiohydrolases to identify enzymes that were efficiently secreted by S. cerevisiae. For enhanced cellulose hydrolysis, we engineered bimodular derivatives of a well secreted enzyme that naturally lacks the carbohydrate-binding module, and constructed strains expressing combinations of cbh1 and cbh2 genes. Though there was significant variability in the enzyme levels produced, up to approximately 0.3 g/L CBH1 and approximately 1 g/L CBH2 could be produced in high cell density fermentations. Furthermore, we could show activation of the unfolded protein response as a result of cellobiohydrolase production. Finally, we report fermentation of microcrystalline cellulose (Avicel™) to ethanol by CBH-producing S. cerevisiae strains with the addition of beta-glucosidase.
Gene or protein specific features and compatibility with the host are important for efficient cellobiohydrolase secretion in yeast. The present work demonstrated that production of both CBH1 and CBH2 could be improved to levels where the barrier to CBH sufficiency in the hydrolysis of cellulose was overcome.
Keywordsbiofuels cellulolytic yeast UPR
The baker's yeast Saccharomyces cerevisiae has been extensively studied as a production host for heterologous proteins and other valuable compounds [1–3]. Due to its long use and beneficial properties as a robust production host in large scale, especially commercial ethanol production, interest in metabolic engineering and utilization of the engineered S. cerevisiae in the production of fuels and other bulk chemicals from renewable resources keeps increasing. S. cerevisiae is expected to continue as a prominent host in future biorefineries that aim to effectively convert currently unutilized plant materials to useful products.
A low-cost bioprocess to produce bulk fuels and chemicals requires several changes to be made in the metabolism of S. cerevisiae. One of these is the utilization and fermentation of all biomass derived sugars. Consequently, engineering S. cerevisiae for pentose sugar fermentation, particularly D-xylose and L-arabinose derived from lignocellulosic raw materials, has been one of the successfully met challenges in the development of second generation bioethanol production (reviewed in ). Lignocellulose hydrolysis to fermentable sugars is currently achieved by biomass pretreatment and the addition of separately produced enzyme preparations into the process. The enzymes are often also allowed to act during the actual fermentation leading to simultaneous saccharification and fermentation (SSF) [5, 6]. While SSF has benefits such as uptake of the released glucose by the fermenting organism, which counteracts glucose inhibition of cellulases, the high cost of added enzymes is still a major factor in the process economics. Since the conversion of lignocellulosic raw material into monomer sugars is limited by the rate and extent of conversion of the plant polysaccharides by enzymes, engineering of yeast to secrete rate limiting enzymes would complement extensive efforts undertaken to engineer existing enzymes, and further streamline the process towards a consolidated bioprocessing (CBP) and lower production costs [6, 7].
Hydrolysis of cellulose, in particular the complete hydrolysis of the more crystalline parts, is considered a key challenge in biomass hydrolysis [8, 9]. Cellulose is hydrolyzed with mixtures of three different types of enzymes that hydrolyze the β-1,4-glycosidic bonds, cellobiohydrolases (CBHs, EC 220.127.116.11), endo-β-1,4-glucanases (EG, EC 18.104.22.168) and β-glucosidase (EC 22.214.171.124) (for review see ). CBHs are instrumental in the hydrolysis of natural cellulose that contains highly ordered crystalline regions. These enzymes act from cellulose chain ends, releasing mainly cellobiose. The endoglucanases attack more amorphous parts and hydrolyze cellulose from the middle of the chains, acting in synergy with CBHs to hydrolyze the substrate to small oligosaccharides. Finally, β-glucosidase hydrolyses cello-oligosaccharides to glucose.
The most studied cellulolytic fungus, Trichoderma reesei, produces up to about 80% of the total secreted protein as CBH, and the best production strains can secrete tens of grams per liter of these enzymes . There are two major fungal CBH classes, separated into the glycosyl hydrolase families GH-7 (also called CBH1) and GH6 (CBH2) based on their sequence similarity and predicted structural and functional relationships (http://www.cazy.org/; ). The catalytic domains of these two enzyme classes are structurally different but both share a tunnel-like active site. Many fungal CBHs have a separate, small cellulose-binding module (CBM) belonging to the CBM-1 family http://www.cazy.org/. In GH-7 CBHs, the CBM is attached to the C-terminus via a flexible linker, and in GH-6 enzymes to the N-terminus. The CBM is considered to be essential for hydrolysis of crystalline cellulose . Several studies indicate that CBH1 and CBH2 types of enzymes also act in synergy in cellulose hydrolysis [8, 12, 13].
Cellulases were among the first heterologous proteins expressed in yeast [14–17] and since then several reports have shown that S. cerevisiae can secrete fungal hydrolytic enzymes, including CBHs. The naturally N- and O-glycosylated CBH enzymes are typically hyperglycosylated with high mannose glycans in S. cerevisiae [15, 18–20]. The examples of CBH expressed in yeast include CBH1 (Cel7A) and CBH2 (Cel6A) of T. reesei [15, 19, 21, 22], and CBHs of other fungi [18–20, 23, 24]. Although the enzymes retain activity when expressed in S. cerevisiae, there are results indicating that the activity of the yeast-produced enzymes is impaired in comparison to the native proteins, which in some cases could be due to overglycosylation [15, 23, 25] or misfolding .
Relatively high protein production levels of 1-10% of cellular protein have been reported in S. cerevisiae . However, there have also been reports of poor levels of protein secretion, especially when expressing cellulase-encoding genes . In order to maximize production of a heterologous protein, the gene copy number, codon usage and the choice of promoters are important for obtaining sufficient transcript levels in the host. Subsequently, stable transcripts are required to maintain high levels of translation of the heterologous gene. Production of secreted proteins also requires that they are able to enter the secretory pathway, are correctly folded and processed there, and finally are secreted in an active form into the extracellular medium. Heterologous protein production is known to be limited by cellular stress reactions that can largely influence productivity . For example, the accumulation of unfolded proteins in the endoplasmic reticulum (ER) causes stress and induces the unfolded protein response (UPR) that coordinates the physiological responses to ER stress . It is well established that Hac1p mediates the UPR in S. cerevisiae [30, 31]. The constitutively synthesized HAC1 mRNA is spliced in response to ER stress, resulting in the synthesis of the active DNA-binding transcription factor Hac1p. This activates the expression of genes coding for chaperones, foldases and components of the ER associated degradation system in order to adapt to the situation by increasing the protein folding capacity of the ER and by clearing misfolded proteins from the ER . Moreover, autoregulation of the HAC1 gene is required for sustained activation of the UPR and resistance to ER stress .
In this work we carried out a comprehensive study of expression of various cbh genes to identify enzymes that can be efficiently secreted by S. cerevisiae in an active form into extracellular medium, which is fundamental for a successful CBP or SSF process. With the aim of enhancing cellulose hydrolysis further from the levels obtained during screening, we engineered bimodular derivatives of a well secreted CBM-less enzyme and constructed strains expressing various combinations of the cbh1 and cbh2 genes. We examined the physiological impact of CBH production in S. cerevisiae expressing the different cbh genes and discovered correlations with the UPR. Finally, we report fermentation of microcrystalline Avicel™ cellulose to ethanol by selected CBH-expressing S. cerevisiae strains with the aid of externally added β-glucosidase.
Expression and secretion of CBH1 and CBH2
Cellobiohydrolases expressed in S. cerevisiae in this study
Species and gene name
GenBank accession number
Recombinant yeast strain abbreviation
cbh1 encoding genes: expressed under transcriptional control of S. cerevisiae ENO1 promoter
Humicola grisea cbh1 a, b
Thermoascus aurantiacus cbh1 a, b
Talaromyces emersonii cbh1 a, b
Neosartorya fischeri cbh1 c, d
Penicillium janthinellum cbh1 c, d
Gibberella zeae cbh1 c, d
Nectria haematococca cbh1 c, d
Fusarium poae cbh1 c, d
Aspergillus terreus cbh1 c, d
Penicillium chrysogenum cbh1 c, d
Neurospora crassa cbh1 c, d
Chaetomium thermophilum cbh1 b, e
Acremonium thermophilum cbh1 b, e
Trichoderma reesei cbh1 f
cbh2 encoding genes: expressed under transcriptional control of S. cerevisiae PGK1 promoter
Cochliobolus heterostrophus C4 cel7 b, c
Gibberella zeae K59 cel6 b, c
Irpex lacteus MC-2 cex b, c
Volvariella volvacea cbhII-I b, c
Piromyces sp. E2 cel6A b, c
Talaromyces emersonii cbh2 a, b
Trichoderma reesei cbh2 a, f
Chrysosporium lucknowense cbh2b b, d
Acremonium cellulolyticus cbh2 b, e
Chaetomium thermophilum cbh2 b, e
The first screening for CBH production in YPD cultivations was carried out using two enzymatic methods. To assess enzymatic hydrolysis of polymeric insoluble cellulose, secreted CBH1 and CBH2 activity was measured by incubating the cell-free yeast culture supernatants with Avicel PH105 cellulose in the presence of β-glucosidase (Novozyme 188) to hydrolyze the cellobiose released from cellulose to glucose, followed by determination of the reducing sugars formed. CBH1 production was also measured by activity on the soluble fluorescent substrate 4-methylumbelliferyl β-D-lactoside (MULac); an analogous substrate is not available for CBH2. The activity of the highly expressed T.e.CBH1 reached its maximum after three days growth on YPD medium and therefore day three samples were analyzed for all strains.
Results of CBH1 production by a subset of the best strains and reference strains grown in SCD-URA medium were consistent with the results in YPD medium with regard to activity-based ranking of the best strains. The culture supernatant of Sc[T.e.cbh1] had by far the highest activity on the soluble substrate MULac in both media (Figure 1A and Additional file 1), while Sc[H.g.cbh1] had relatively low activity on MULac, even though it had the highest activity on crystalline cellulose (Figure 1B and Additional file 1). Measurement of protein concentration in the cell-free SCD-URA culture supernatants and SDS-PAGE analyses confirmed that T.e.CBH1 enzyme was abundantly produced relative to the other CBH1 enzymes (Figure 1C). All CBH1 proteins contained N-glycans, as their mobility in the gel was altered following enzymatic N-glycan removal by endoH treatment (Figure 1C) which shows that hyperglycosylation of the CBH1 enzymes occurs in yeast. Avicel conversion by Ac.t.CBH1 and T.r.CBH1 was the least efficient and the enzymes could be visualized as distinct bands only after the removal of N-linked glycans followed by SDS-PAGE (Figure 1C). Even after the removal of N-glycans all the protein bands had a larger molecular weight than predicted based on the amino acid sequence, as could be expected based on several earlier studies. For example, it is likely that O-glycosylation of the CBH1 enzymes takes place in S. cerevisiae as it does in the native organisms.
Since the CBH1 enzyme appeared in most strains as the major band in the SCD-URA culture supernatant in SDS-PAGE, and the level of host proteins secreted in these conditions was not noticeably changed as a result of cbh1 expression, it seemed reasonable to estimate the amount of secreted CBH1 by measuring the protein concentrations with the BioRad protein assay in the cell-free culture supernatants and subtracting the values for the empty vector control from those of the CBH-producing strains. Estimations of T.e.CBH1 protein concentration based on total protein and the concentration of active T.e.CBH1 based on specific activity on MULac (Figure 1C) were fairly consistent, as they were for the Ac.t.CBH1, which was produced at a low level but was evidently capable of cellulose conversion. In comparison, the enzymatic activity of the T. reesei CBH1 was not proportional to the amount of protein measurement (Figure 1C), suggesting that only a small fraction of the secreted enzyme pool was enzymatically active, similarly to the T.r.CBH1 expressed in Pichia pastoris . The concentration of active C.t.CBH1 also was lower than the concentration estimated from the secreted protein measurement (Figure 1C).
Improvement of cellulose hydrolysis by CBM attachment to CBH1
Fusion genes created with T. emersonii cbh1 for expression in S. cerevisiae
Origin of CBM
Recombinant yeast strain abbreviation
Primers used for construction (5'-3')
T. reesei cbh1 a
H. grisea cbh1 a
C. thermophilum cbh1 a
T. reesei cbh2 a
406TEM CBH1 NCBM-L CCTCCGAATTCATGCTAAGAAGAGCTTTACTATTGA-GCTCTTCTGCTATCTTGGCCGTTAAGGCTCAAGCCTGCTCCTCTGTTTGG
407TEM CBH1 NCBM-R AAACTTCAAGTCACGTGGACATTGAGAGTCACAG
T. reesei cbh2 b
T. reesei cbh1 b
Co-secretion of CBH1 and CBH2
Combinations of cellobiohydrolases expressed in S. cerevisiae in this study
Recombinant yeast strain abbreviation
T. emersonii cbh1 a
T. reesei cbh2 b
Sc[Tecbh1 & Trcbh2]
T. emersonii cbh1-T.r.CBM-C a
T. reesei cbh2 b
Sc[Tecbh1-TrCBM-C & Trcbh2]
C. thermophilum cbh1 a
T. reesei cbh2 b
Sc[Ctcbh1 & Trcbh Clcbh2b]
H. grisea cbh1 a
T. reesei cbh2 b
Sc[Hgcbh1 & Trcbh2]
T. aurantiacus cbh1 a
T. reesei cbh2 b
Sc[Tacbh1 & Trcbh2]
T.r.CBM-N-T. emersonii cbh1 b
T. reesei cbh2 b
Sc[Tecbh1-TrCBM-N & Trcbh2]
T. emersonii cbh1 a
C. lucknowense cbh2b a
Sc[Tecbh1 & Clcbh2b]
C. thermophilum cbh1 a
C. lucknowense cbh2b a
Sc[Ctcbh1 & Clcbh2b]
T. emersonii cbh1-T.r.CBM a
C. lucknowense cbh2b a
H. grisea cbh1 a
C. lucknowense cbh2b a
Sc[Hgcbh1 & Clcbh2b]
Two of the best performing combinations were studied in more detail by comparing Avicel hydrolysis for several dilutions of the cell-free yeast culture supernatants. The dilutions displaying the most similar cellulose conversion rates are plotted in Additional file 3C to enable comparison between the samples. These data show that culture supernatant of Sc[Tecbh1-TrCBM-C] was two and a half times, Sc[Tecbh1-TrCBM-C & C.l.cbh2b] was over six times, and Sc[H.g.cbh1 & C.l.cbh2b] was four and a half times more efficient in cellulose conversion than the T.e.CBH1.
Consequences of CBM fusion and co-expression two cbhgenes on CBH production
In an attempt to explain why some CBHs are secreted at high concentration while others are not, and why co-expression of two enzymes alters CBH production levels relative to single enzyme production levels, we investigated relative differences in cbh mRNA levels, in copy number of the expression vector, and in secretion stress-induced responses in a set of eight strains including both high and low cellulase producers. S. cerevisiae expressing T.r.cbh1, T.e.cbh1, Tecbh1-TrCBM-C, Tecbh1-TrCBM-C & T.r.cbh2, Tecbh1-TrCBM & C.l.cbh2b, T.r.cbh2, C.l.cbh2b and the empty vector control strain were grown in YPD medium for three days and sampled daily for RNA isolation and enzyme activity measurements.
Among the strains co-expressing cbh1 and cbh2 the copy numbers were relatively low, but interestingly there was a notable exception. The strain expressing T.e.cbh1 & C.l.cbh2b, the combination of the most highly expressed cbh1 and cbh2 genes, had a copy number intermediate to strains Sc[T.e.cbh1] and Sc[C.l.cbh2b] (Additional file 4A). Thus, irrespective of its larger size, the plasmid with two cbh expression cassettes, each about 3 kb, existed in more copies than a plasmid with only one expression cassette.
The expression of the unfolded protein response regulator HAC1 was studied as it is a sensitive indicator of UPR induction. The HAC1 u transcript that does not code for a functional protein was detected in all cells (Additional file 4B). The spliced HAC1 i mRNA coding for the UPR-inducible transcription factor was not detected in the strain containing the empty vector, while it appeared in each of the CBH strains, indicating that UPR was induced (Figure 5C and Additional file 4B). Expression of two other genes, KAR2(Bip) and PDI1, that are known to be induced by UPR, were also analyzed and their transcript levels were elevated relative to the empty vector control providing additional evidence for UPR in the cells (Additional file 4B).
The level of the HAC1 i mRNA varied between the strains so that the Sc[C.l.CBH2b] strain had the lowest HAC1 i levels at each time point throughout the cultivation suggesting that expression of this protein was the least stressful for the cell's secretion machinery. Similarly, the strain Sc[T.e.cbh1] producing an efficiently secreted enzyme also had relatively low levels of HAC1 i . Comparison between the Sc[T.e.cbh1] and Sc[Tecbh1-TrCBM-C] strains showed that the strain expressing the bi-modular enzyme had 2-3 fold higher HAC1 i mRNA levels suggesting that production of the fusion protein caused a higher ER stress. Furthermore, the production of T.r.CBH1 protein caused a relatively strong UPR induction as judged by the HAC1 i mRNA level which suggests that the post-translational processing in the secretory pathway was impaired resulting in secretion of less than 1 mg/L of active T.r.CBH1 protein.
CBH1 and CBH2 production in bioreactor
Strains M0759 expressing Tecbh1-TrCBM-C and M0969 expressing C.l.cbh2b (Tables 1 and 2), derived from the industrial background strain M0749 and disrupted in both copies of the FUR1 gene, were evaluated for their ability to accumulate CBH1 and CBH2 proteins during aerobic glucose fed-batch high cell density cultivation.
The experiment was also performed for Y294 derived laboratory strains expressing Tecbh1-TrCBM-C or C.l.cbh2b. The laboratory strains reached three-fold less DCW at the end of glucose feed. Proportionally, the strains yielded three- to four-fold less protein per volume. Therefore, even though industrial strains were capable of reaching significantly higher biomass in aerobic bioreactor propagation conditions, DCW normalized protein production was similar for industrial and laboratory strains.
Anaerobic Avicel fermentation with the aid of externally added β-glucosidase
As shown above, glucose accumulated during the incubation of cell-free culture supernatants of Sc[cbh] strains with Avicel cellulose and Novozyme 188 β-glucosidase in vitro (Figures 1, 2, 3, 4). In order to demonstrate the ability of the cellulolytic yeast to convert crystalline cellulose to soluble sugars, and further to ethanol, in vivo under typical yeast cultivation conditions, the following experiment was carried out. The strains expressing Tecbh1-TrCBM-C and C.l.cbh2b separately or in combination and the empty vector control were grown aerobically on YP-2% glucose medium for three days, and then Avicel cellulose was added to 20 g/L into the cultivation, and the incubation was continued under anaerobic conditions to prevent consumption of the ethanol that would be produced. Since the recombinant S. cerevisiae does not metabolize cello-oligosaccharides, Novozyme 188 β-glucosidase was added into the cultivations to enable conversion of cellobiose to glucose and subsequent fermentation, while parallel control flasks were not supplemented with Novozyme 188.
High level functional expression and secretion of exoglucanases are requirements for enabling non-cellulolytic organisms such as S. cerevisiae to utilize crystalline cellulose substrates . The difficulty of producing CBHs in sufficient quantities is considered as a major hurdle in the development of yeast as a CBP organism [6, 19, 36]. In this study we have attempted to alleviate this problem through identifying gene candidates that are compatible with expression in yeast. To this end we screened 14 cbh1 (Cel7A) and 10 cbh2 (Cel6A) encoding genes from ascomycetous origin by functional expression in S. cerevisiae. Somewhat surprisingly, despite sequence conservation and similar folding within CBH1s and also within CBH2s from different species, a wide range of enzyme production levels was observed even though the same regulatory sequences for all the cbh1 genes (ENO1 promoter and terminator) and cbh2 genes (PGK1 promoter and terminator), the same vector backbone and the same host cells were used. The T.r.CBH1 included in our study for comparison was produced at levels comparable to earlier reports of 0.2-5.0 mg/L of T.r.CBH1 [15, 19, 25], confirming the disappointingly poor production. However, we identified several other cbh1 genes that were expressed more efficiently: the activity of T.e.CBH1 and its derivative with the T.r.CBM attached to its C-terminus exceeded that of T.r.CBH1 by at least two orders of magnitude, yielding 100-200 mg/L in shake flasks and 300 mg/L in high cell density conditions. This shows a large improvement compared with a recent report of 5-10 mg/L T.e.CBH1 . Heinzelman et al.  expressed cbh1 genes of T. reesei, T. emersonii, A. thermophilum, C. thermophilum and T. aurantiacus. The activity based ranking is similar to our results, with the exception of T. aurantiacus, however the enzyme secretion levels were higher in our work, which may result from differences in codon optimization, or strain and construct specific effects or, in the case of T. aurantiacus, a difference in the amino acid sequences chosen initially. The highest CBH level secreted, 1000 mg/L C.l.CBH2b, corresponding to 4% of the total cellular protein, was produced in high cell density conditions, exceeded any previous reports on CBH production in S. cerevisiae. In the shake flask cultivations on defined medium the difference between secreted C.l.CBH2b and T.r.CBH2 was two- to three-fold, the T.r.CBH2 level being comparable with earlier reported levels of 10-100 mg/L [15, 16]. In contrast to earlier conclusions, the present work proves that S. cerevisiae is capable of secreting CBHs at high levels that compare well with the highest heterologous protein production levels described for S. cerevisiae [27, 37, 38].
The differences in secreted enzyme levels can be explained to a large extent by the differences in plasmid copy number, which were highest for the strains secreting the highest enzyme levels, Sc[T.e.cbh1], Sc[C.l.cbh2b], and Sc[Tecbh1-TrCBM-C & C.l.cbh2b]. While the copy number and segregation of the endogenous 2-micron circle is under strict control , little is known about copy number control of artificial 2-micron plasmids [40, 41] even though they are widely used essential molecular biology tools, and the best option to ensure high expression level, which is necessary for addressing secretability of the proteins of interest. The suggestion that the plasmid size may affect its copy number and stability  seems unlikely in our case since the sizes differed by less than 1% within the cbh1 expressing plasmids and within the cbh1 and cbh2 co-expressing plasmids. It appears that the cbh gene inserts influence the plasmid copy number significantly, but the mechanism by which this occurs remains unknown. Possibly the cbh inserts affect plasmid replication or transcription, or indirect cellular effects caused by the cbh translation products may be involved.
Among the strains expressing T.e.cbh1 or its derivatives, the plasmid copy number, CBH1 enzyme activity and cbh1 mRNA levels were consistent, all of which were the highest for the strain Sc[T.e.cbh1] followed by strains expressing Sc[Tecbh1-TrCBM-C], Sc[Tecbh1-TrCBM-C & C.l.cbh2b] and Sc[Tecbh1-TrCBM-C & T.r.cbh2] in that order. With regard to the last two strains, it appeared that the presence of C.l.cbh2b allowed higher plasmid, mRNA and activity levels than T.r.cbh2 did, similarly to strains Sc[C.l.cbh2b] and Sc[T.r.cbh2]. Moreover, it appears that the attachment of the linker and CBM moieties to the T.e.CBH1 catalytic domain decreased the plasmid copy number, transcript and enzyme levels. The results indicate that individual gene and/or protein specific features and compatibility with the host are important, not only for efficient production of the individual protein but also for efficient production of the accompanying protein, when two or more genes are simultaneously expressed from one plasmid. From this point of view, expression of integrated gene copies could be a useful strategy; however, high level expression is likely to require integration of each gene in multiple copies. Multicopy integration has also been applied in the construction of yeast strains expressing cellulases [42, 43].
The T.r.cbh1 mRNA level and plasmid copy number were comparable to those of strain Sc[Tecbh1-TrCBM-C & T.r.cbh2] and therefore it could be expected that a reasonable amount of T.r.CBH1 activity would have been detected. Because activity was barely detectable, it implies that post-transcriptional events have a major impact on the production and secretion of this enzyme from S. cerevisiae. It is possible that the activity of TrCBH1 was impaired due to hyperglycosylation or incorrect folding to some extent, as has been suggested previously [15, 25, 26].
The secretion of heterologous proteins is believed to be limited by processes occurring in the ER [32, 44]. The machinery required for proper protein folding may become saturated when heterologous proteins are over expressed, causing accumulation of misfolded or aggregated proteins in the ER. The UPR regulates gene expression in response to ER stress, resulting in selective induction of genes that are essential under folding stress and in specific remodeling of the secretory pathway to improve the protein folding capacity . The transcription factor Hac1p is the central regulator mediating the UPR [30, 31]. It has been suggested before that some components of the UPR pathway are involved in the secretion of T. reesei endoglucanase, which was based on a comparison between two S. cerevisiae strains, one with an intact HAC1 gene and the other with a disrupted hac1 gene preventing the UPR . Since the effects of cellulase production on the secretory machinery have not been studied before in S. cerevisiae, we undertook northern analysis of the UPR-related genes to provide insight into the physiological responses associated with CBH production. Especially HAC1 i , but also PDI1, and KAR2 mRNA levels were elevated in the strains expressing certain cbh1 and/or cbh2 genes relative to the empty vector control, which showed that the expression of the CBHs studied caused ER stress and activated the UPR in the cells in order to adapt to the prevailing conditions. This is to our knowledge the first demonstration of UPR activation in S. cerevisiae in response to cellulase expression.
The strength of UPR activation varied depending on the gene as judged from the expression levels of the HAC1 i mRNA. Interestingly, a negative correlation between the HAC1 i mRNA levels and the amount of secreted active enzyme was found. For example, the Sc[C.l.cbh2b] strain which secreted the highest amount of CBH had the lowest HAC1 i mRNA level of all the cbh-expressing strains studied. This suggests that the protein folding capacity was greater in strain Sc[C.l.cbh2b] than for the other strains, which enabled high level secretion of C.l.CBH2b. The Sc[Tecbh1-TrCBM-C] also had a relatively high HAC1 i mRNA level and low enzyme level compared with the strain Sc[T.e.cbh1], which suggests that additional ER protein folding capacity was required to produce the bimodular enzyme, in which the formation of two additional disulfide bridges is necessary for correct folding of the CBM. Furthermore, it appears that the T.r.CBH1 and T.r.CBH2 proteins were more potent than, for example, T.e.CBH1 or C.l.CBH2b in causing secretion stress and inducing the UPR, even though the gene copy number and resultant expression level of T.r.cbh1 and T.r.cbh2 were lower than those of the abundantly expressed T.e.cbh1 or C.l.cbh2b. It may be possible that, despite the strong UPR, the ER protein folding capacity was not sufficient to enable efficient secretion of, for example, T.r.CBH1.
As concluded above, the plasmid copy number could explain secreted CBH levels to a large extent. At the same time, the expression of certain genes induced a stress response in the ER and upregulation of the UPR correlated with low plasmid copy number. It would appear that some CBHs are thus more compatible with high-level expression and production in S. cerevisiae than others, although which features lead to incompatibility, marked by low levels of plasmid, mRNA and secreted protein and strong induction of UPR, are difficult to define. The stress response indicated that CBH production was a burden to the cells. One way to relieve the stress could be downregulation of CBH production either through UPR or through decreasing the plasmid copy number. Whether there is a link between ER stress and plasmid copy number control or whether they occur independently of each other cannot be concluded from our data, but would require a separate study.
S. cerevisiae, the most efficient ethanol producer on a large industrial scale, was shown to be capable of high level CBH expression. This indicates that it is a promising organism for conversion of cellulosic biomass to ethanol. The main obstacle in the way of applying CBP with S. cerevisiae is considered to be the sufficiency of CBH production, estimated to require approximately 20- to 120-fold improvement . The progress made in the present work demonstrated that production of both CBH1 and CBH2 could be improved to that level and that the barrier of CBH sufficiency was overcome. Data demonstrating that yeast co-expressing CBH1 and CBH2 could ferment Avicel cellulose to ethanol with the aid of externally added β-glucosidase supports this conclusion. Simultaneous expression of CBHs with endoglucanases and β-glucosidase is the next step to enable S. cerevisiae to directly convert cellulose to ethanol and to grow on cellulose under CBP conditions. It can be envisioned that cellulolytic S. cerevisiae will also be suitable for other biorefinery process concepts, exploiting the capability of yeast to convert cellulose to useful products other than ethanol.
Strains, media and culture conditions
Escherichia coli strains XL1 Blue MRF' (Stratagene, La Jolla, CA, USA) and DH5α were used for cloning. S. cerevisiae Y294 (αleu2-3,112 ura3-52 his3 trp1-289) [ATCC 201160] was used as the host for CBH expression. S. cerevisiae M0749 (Mascoma proprietary industrial strain)  was used as the host for larger scale production for CBH1 and CBH2 enzyme purification. Yeast were grown at 30°C with shaking in YPD and SCD media supplemented with the necessary amino acids as required (Additional file 5).
Plasmid and strain construction
Standard DNA techniques  were used in the study. Details about plasmids and recombinant strains used and constructed are summarized in Tables 1, 2 and 3. The nucleotide sequences of the cbh genes were codon-optimized for expression in S. cerevisiae and synthesized by de novo gene synthesis providers (Table 1). Synthetic genes were subsequently cloned onto yeast expression vectors containing the URA3 selection marker and the 2-micron sequence for autonomous replication. The cbh1 genes were expressed under transcriptional control of the S. cerevisiae enolase gene (ENO1) promoter, and the cbh2 genes under the S. cerevisiae 3-phosphoglycerate kinase gene (PGK1) promoter and terminator. To attach carbohydrate-binding modules (CBMs) to the CBM-less T.e.CBH1, PCR was used (Table 2). For simultaneous expression of cbh1 and cbh2 genes the ENO1p-cbh1-ENO1t and the PGK1p-cbh2-PGK1t expression cassettes were cloned into the same vector in different combinations (Table 3). S. cerevisiae was transformed with the lithium acetate dimethylsulfoxide method  and selected for uracil prototrophy on SCD-URA. Autoselective strains were constructed to ensure maintenance of the URA3-bearing expression vectors in complex medium (Additional file 5).
To determine exoglucanase activity on a polymeric insoluble substrate, 300 μL of the yeast culture supernatant was added to deep-well microtiter plates with each well containing 300 μL of 2% Avicel PH-105 cellulose (FMC Biopolymer, Mechanicsburg, PA, USA), 0.05 M acetate buffer pH 5.0, 0.04% sodium azide and 0.3 μL Novozyme-188 (Sigma-Aldrich, St Louis, MO, USA) at approximately 1000 rpm at 35°C. The amount of sugars released during 24 h and 28 h incubations was determined using a modified 3,5-dinitrosalicylic acid (DNS) method (Additional file 5). Glucose was used to set a standard curve in the range of 0.125 to 4 g/L from which the amount of glucose released during the assay was determined. The amount of activity was expressed as the percentage of Avicel hydrolyzed.
Secreted activities for the strains producing CBH1 (GH7) enzymes were determined using soluble 4-methylumbelliferyl-β-D-lactoside (MULac, Sigma) (Additional file 5).
TeCBH1-TrCBM-C and C.l.CBH2b were purified using chromatography methods for use as protein standards in the HPLC assay (Additional file 5).
Other protein analysis methods
Protein concentrations in shake flask cultivations were measured with BioRad protein reagent (Additional file 5). Endoglycosidase H (Roche, Mannheim, Germany) was used to remove N-linked glycans (Additional file 5). Protein samples were separated on SDS-PAGE gels (BioRad) and visualized with silver staining.
For determination of the concentration of CBHs produced in bioreactor cultivations, a phenyl reversed phase method was developed on HPLC, and the purified TeCBH1-TrCBM-C and C.l.CBH2b were used for generating a standard curve (Additional file 5).
Determination of plasmid copy number
Yeast DNA was isolated by phenol extraction from cells grown overnight in YPD (Additional file 5). Radioactive URA3 hybridization signals on Southern blots were quantified as described in Additional file 5 and the plasmid copy number was determined as the ratio between the plasmid-borne and the genomic copy of URA3. The copy number is expressed in relative units.
Gene expression studies
Yeast were grown in 50 mL YPD medium in 250 mL Erlenmeyer flasks at 30°C at 250 rpm and 2 mL samples were removed periodically. Cells were harvested by centrifugation, frozen in dry ice and stored at -70°C. RNA was isolated using the Trizol reagent (Invitrogen cat. no. 15596-018). Northern blots were prepared and hybridized with T.e.cbh1, T.r.cbh2, C.l.cbh2b, T.r.cbh1 CBM, HAC1, ACT1, KAR2 and PDI1 probes (Additional file 6) using conventional techniques . Radioactive hybridization signals were detected and quantified as above.
Bioreactor propagation of CBH-producing yeast strains
Aerobic glucose fed-batch high cell density cultivation was performed in 2 liters working volume bioreactors with strains M0759 and M0969 (Additional file 5).
Avicel fermentation to ethanol
The yeast strains were grown in YPD medium for four days (Additional file 5). Subsequently, 25 mL of each culture was added to McCartney bottles containing 0.5 g of Avicel PH-105 to attain a concentration of 20 g/L. In addition, 100 μL of the β-glucosidase preparation Novozyme 188 (Sigma) was added so that for each strain there were triplicate bottles with and without added enzyme. The bottles were sealed with rubber lined caps to maintain the cultures anaerobically and stirred on magnetic stirrers for seven days. Samples were taken on days 0, 2, 4 and 7 and cellobiose, glucose and ethanol content was determined with HPLC (Additional file 5).
dry cell weight
high performance liquid chromatography
polymerase chain reaction
synthetic complete dextrose medium
simultaneous saccharification and fermentation
unfolded protein response
yeast extract peptone dextrose.
The authors thank Merja Helanterä, Chhayal Gandhi, Arja Kiema, and Ulla Vornamo for skilled technical assistance, Dominik Mojzita for constructing pTeHg, Brian Firliet for protein concentration measurements by HPLC, and Annatjie Hugo help in sugar concentration measurements by HPLC.
- Romanos MA, Scorer CA, Clare JJ: Foreign gene expression in yeast: a review. Yeast. 1992, 8: 423-488. 10.1002/yea.320080602.View ArticleGoogle Scholar
- Vasavada A: Improving productivity of heterologous proteins in recombinant Saccharomyces cerevisiae fermentations. Adv Appl Microbiol. 1995, 41: 25-54.View ArticleGoogle Scholar
- Idiris A, Tohda H, Kumagai H, Takegawa K: Engineering of protein secretion in yeast: strategies and impact on protein production. Appl Microbiol Biotechnol. 2010, 86: 403-417. 10.1007/s00253-010-2447-0.View ArticleGoogle Scholar
- Hahn-Hägerdal B, Karhumaa K, Fonseca C, Spencer-Martins I, Gorwa-Grauslund MF: Towards industrial pentose-fermenting yeast strains. Appl Microbiol Biotechnol. 2007, 74: 937-953. 10.1007/s00253-006-0827-2.View ArticleGoogle Scholar
- Lynd LR, Weimer PJ, van Zyl WH, Pretorius IS: Microbial cellulose utilization: fundamentals and biotechnology. Microbiol Mol Biol Rev. 2002, 66: 506-577. 10.1128/MMBR.66.3.506-577.2002.View ArticleGoogle Scholar
- Lynd LR, van Zyl WH, McBride JE, Laser M: Consolidated bioprocessing of cellulosic biomass: an update. Curr Opin Biotechnol. 2005, 16: 577-583. 10.1016/j.copbio.2005.08.009.View ArticleGoogle Scholar
- van Zyl WH, Lynd LR, den Haan R, McBride JE: Consolidated bioprocessing for bioethanol production using Saccharomyces cerevisiae. Biofuels. 2007, 108: 205-235. 10.1007/10_2007_061.View ArticleGoogle Scholar
- Zhang YH, Lynd LR: Toward an aggregated understanding of enzymatic hydrolysis of cellulose: noncomplexed cellulase systems. Biotechnol Bioeng. 2004, 88: 797-824. 10.1002/bit.20282.View ArticleGoogle Scholar
- Zhou W, Schuttler HB, Hao Z, Xu Y: Cellulose hydrolysis in evolving substrate morphologies I: a general modeling formalism. Biotechnol Bioeng. 2009, 104: 261-274. 10.1002/bit.22389.View ArticleGoogle Scholar
- Tangnu SK, Blanch HW, Wilke CR: Enhanced production of cellulase, hemicellulase, and β-glucosidase by Trichoderma reesei (Rut C-30). Biotechnol Bioeng. 1981, 23: 1837-1849. 10.1002/bit.260230811.View ArticleGoogle Scholar
- Henrissat B, Bairoch A: Updating the sequence-based classification of glycosyl hydrolases. Biochem J. 1996, 316: 695-696.View ArticleGoogle Scholar
- Zhang YH, Lynd LR: A functionally based model for hydrolysis of cellulose by fungal cellulase. Biotechnol Bioeng. 2006, 94: 888-89. 10.1002/bit.20906.View ArticleGoogle Scholar
- Nidetzky B, Steiner W, Hayn M, Claeyssens M: Cellulose hydrolysis by the cellulases from Trichoderma reesei: a new model for synergistic interaction. Biochem J. 1994, 298: 705-710.View ArticleGoogle Scholar
- Penttilä ME, André L, Saloheimo M, Lehtovaara P, Knowles JKC: Expression of two Trichoderma reesei endoglucanases in the yeast Saccharomyces cerevisiae. Yeast. 1987, 3: 175-185. 10.1002/yea.320030305.View ArticleGoogle Scholar
- Penttilä ME, André L, Lehtovaara P, Bailey M, Teeri TT, Knowles JKC: Efficient secretion of two fungal cellobiohydrolases by Saccharomyces cerevisiae. Gene. 1988, 63: 103-112. 10.1016/0378-1119(88)90549-5.View ArticleGoogle Scholar
- Zurbriggen B, Bailey MJ, Penttilä ME, Poutanen K, Linko M: Pilot scale production of a heterologous Trichoderma reesei cellulase by Saccharomyces cerevisiae. J Biotechnol. 1990, 13: 267-278. 10.1016/0168-1656(90)90075-M.View ArticleGoogle Scholar
- Bailey MJ, Siika-aho M, Valkeajärvi A, Penttilä M: Hydrolytic properties of two cellulases of Trichoderma reesei expressed in yeast. Biotechnol Appl Biochem. 1993, 17: 65-76.Google Scholar
- Hong J, Tamaki H, Yamamoto K, Kumagai H: Cloning of a gene encoding thermostable cellobiohydrolase from Thermoascus aurantiacus and its expression in yeast. Appl Microbiol Biotechnol. 2003, 63: 42-50. 10.1007/s00253-003-1379-3.View ArticleGoogle Scholar
- den Haan R, McBride JE, la Grange DC, Lynd LR, van Zyl WH: Functional expression of cellobiohydrolases in Saccharomyces cerevisiae towards one-step conversion of cellulose to ethanol. Enz Microb Technol. 2007, 40: 1291-1299. 10.1016/j.enzmictec.2006.09.022.View ArticleGoogle Scholar
- Voutilainen SP, Murray PG, Tuohy MG, Koivula A: Expression of Talaromyces emersonii cellobiohydrolase Cel7A in Saccharomyces cerevisiae and rational mutagenesis to improve its thermostability and activity. Prot Eng Des Sel. 2010, 23: 69-79. 10.1093/protein/gzp072.View ArticleGoogle Scholar
- Heinzelman P, Snow CD, Wua I, Nguyen C, Villalobos A, Govindarajan S, Minshull J, Arnold FH: A family of thermostable fungal cellulases created by structure-guided recombination. Proc Nat Acad Sci USA. 2009, 106: 5611-5615.View ArticleGoogle Scholar
- Du Plessis L, Rose SH, van Zyl WH: Exploring improved endoglucanase expression in Saccharomyces cerevisiae strains. Appl Microbiol Biotechnol. 2010, 86: 1503-1511. 10.1007/s00253-009-2403-z.View ArticleGoogle Scholar
- Takada G, Kawaguchi T, Sumitani J, Arai M: Expression of Aspergillus aculeatus no. F-50 cellobiohydrolase I (CBH1) and beta-glucosidase 1 (bgl1) genes by Saccharomyces cerevisiae. Biosci Biotehnol Biochem. 1998, 62: 1615-1618. 10.1271/bbb.62.1615.View ArticleGoogle Scholar
- Heinzelman P, Komor R, Kanaan A, Romero P, Yu X, Mohler S, Snow C, Arnold F: Efficient screening of fungal cellobiohydrolase class I enzymes for thermostabilizing sequence blocks by SCHEMA structure-guided recombination. Prot Eng Des Sel. 2010, 23: 871-880. 10.1093/protein/gzq063.View ArticleGoogle Scholar
- Reinikainen T, Ruohonen L, Nevanen T, Laaksonen L, Kraulis P, Jones TA, Knowles JKC, Teeri TT: Investigation of the function of mutated cellulose-binding domains of Trichoderma reesei cellobiohydrolase I. Prot Struct Func Gen. 1992, 14: 475-482. 10.1002/prot.340140408.View ArticleGoogle Scholar
- Boer H, Teeri T, Koivula A: Characterization of Trichoderma reesei cellobiohydrolase Cel7A secreted from Pichia pastoris using two different promoters. Biotechnol Bioeng. 2000, 69: 486-494. 10.1002/1097-0290(20000905)69:5<486::AID-BIT3>3.0.CO;2-N.View ArticleGoogle Scholar
- Schmidt FR: Recombinant expression systems in the pharmaceutical industry. Appl Microbiol Biotechnol. 2004, 65: 363-372. 10.1007/s00253-004-1656-9.View ArticleGoogle Scholar
- Mattanovich D, Gasser B, Hohenblum H, Sauer M: Stress in recombinant protein producing yeasts. J Biotechnol. 2004, 113: 121-135. 10.1016/j.jbiotec.2004.04.035.View ArticleGoogle Scholar
- Schröder M: The unfolded protein response. Mol Biotechnol. 2006, 34: 279-290. 10.1385/MB:34:2:279.View ArticleGoogle Scholar
- Cox JS, Walter P: A novel mechanism for regulating activity of a transcription factor that controls the unfolded protein response. Cell. 1996, 87: 391-404. 10.1016/S0092-8674(00)81360-4.View ArticleGoogle Scholar
- Kaufman RJ: Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene transcriptional and translational controls. Genes Dev. 1999, 13: 1211-33. 10.1101/gad.13.10.1211.View ArticleGoogle Scholar
- Travers KJ, Patil CK, Wodicka L, Lockhart DJ, Weissman JS, Walter P: Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation. Cell. 2000, 101: 249-258. 10.1016/S0092-8674(00)80835-1.View ArticleGoogle Scholar
- Ogawa N, Mori K: Autoregulation of the HAC1 gene is required for sustained activation of the yeast unfolded protein response. Genes Cells. 2004, 9: 95-104. 10.1111/j.1365-2443.2004.00704.x.View ArticleGoogle Scholar
- Calleja GB, Yaguchi M, Levy-Rick S, Seguin JRH, Roy C, Lusena CV: Single-cell protein production from potato starch by the yeast Schwanniomyces alluvius. J Ferm Technol. 1986, 64: 171-175.View ArticleGoogle Scholar
- La Grange DC, Den Haan R, Van Zyl WH: Engineering cellulolytic ability into bioprocessing organisms. Appl Microbiol Biotechnol. 2010, 4: 1195-1208.View ArticleGoogle Scholar
- Xu Q, Singh A, Himmel ME: Perspectives and new directions for the production of bioethanol using consolidated bioprocessing of lignocellulose. Curr Opin Biotechnol. 2009, 20: 364-371. 10.1016/j.copbio.2009.05.006.View ArticleGoogle Scholar
- Park E-H, Shin Y-M, Lim Y-Y, Kwon T-H, Kim D-H, Yang M-S: Expression of glucose oxidase by using recombinant yeast. J Biotechnol. 2000, 81: 35-44. 10.1016/S0168-1656(00)00266-2.View ArticleGoogle Scholar
- Chung BH, Seo DJ, Nam SW: High-level secretory production of recombinant human lipocortin-I by Saccharomyces cerevisiae. Process Biochem. 1999, 35: 97-101. 10.1016/S0032-9592(99)00038-2.View ArticleGoogle Scholar
- Yeh E, Bloom K: Hitching a ride. EMBO reports. 2006, 7: 985-987. 10.1038/sj.embor.7400793.View ArticleGoogle Scholar
- Futcher AB, Cox BS: Copy number and the stability of 2-micron circle-based artificial plasmids of Saccharomyces cerevisiae. J Bacteriol. 1984, 157: 283-290.Google Scholar
- Zhang Z, Moo-Young M, Chisti Y: Plasmid stability in recombinant Saccharomyces cerevisiae. Biotech Adv. 1996, 14: 401-435. 10.1016/S0734-9750(96)00033-X.View ArticleGoogle Scholar
- Cho KM, Yoo YJ, Kang HS: δ-Integration of endo/exo-glucanase and β-glucosidase genes into the yeast chromosomes for direct conversion of cellulose to ethanol. Enz Microb Technol. 1999, 25: 23-30. 10.1016/S0141-0229(99)00011-3.View ArticleGoogle Scholar
- Yamada R, Taniguchi N, Tanaka T, Ogino C, Fukuda H, Kondo A: Cocktail δ-integration: a novel method to construct cellulolytic enzyme expression ratio-optimized yeast strains. Microb Cell Fact. 2010, 9: 32-10.1186/1475-2859-9-32.View ArticleGoogle Scholar
- Arvas M, Pakula T, Lanthaler K, Saloheimo M, Valkonen M, Suortti T, Robson G, Penttilä M: Common features and interesting differences in transcriptional responses to secretion stress in the fungi Trichoderma reesei and Saccharomyces cerevisiae. BMC Genomics. 2006, 7: 32-49. 10.1186/1471-2164-7-32.View ArticleGoogle Scholar
- Valkonen M, Penttilä M, Saloheimo M: Effects of inactivation and constitutive expression of the unfolded-protein response pathway on protein production in the yeast Saccharomyces cerevisiae. Appl Environ Microbiol. 2002, 69: 2065-2072.View ArticleGoogle Scholar
- McBride J, Brevnova E, Ghandi C, Mellon M, Froehlich A, Deleault K, Rajgharia V, Flatt J, van Zyl E, den Haan R, la Grange D, Rose S, Penttila M, Ilmen M, Siika-aho M, Uusitalo J, Hau HH, Rice C, Villari J, Stonehouse E, Gilbert A, Keating J, Xu H: Yeast expressing cellulases for simultaneous saccharification and fermentation using cellulose. 2010, International patent application Pub. No. WO/2010/060056Google Scholar
- Sambrook J, Russell DW: Molecular cloning: a laboratory manual. 2001, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory PressGoogle Scholar
- Hill J, Donald KA, Griffiths DE: DMSO-enhanced whole cell yeast transformation. Nucleic Acids Res. 1991, 19: 5791-10.1093/nar/19.20.5791.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.