Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Transcriptomes of a xylose-utilizing industrial flocculating Saccharomyces cerevisiae strain cultured in media containing different sugar sources

Transcriptomes of a xylose-utilizing industrial flocculating Saccharomyces cerevisiae strain... Lignocellulosic hydrolysates used for bioethanol production contain a mixture of sugars, with xylose being the second most abundant after glucose. Since xylose is not a natural substrate for Saccharomyces cerevisiae, recombi‑ nant S. cerevisiae strongly prefers glucose over xylose, and the fermentation rate and ethanol yield with xylose are both lower than those with glucose. To determine the molecular basis for glucose and xylose fermentation, we used microarrays to investigate the transcriptional difference of a xylose ‑ utilizing industrial strain cultured in both single sugar media and a mixed sugar medium of glucose and xylose. The transcriptomes were nearly identical between glucose metabolizing cells in the glucose alone medium and those in the glucose fermentation phase in the mixed‑ sugar medium. Whereas the transcriptomes highly differed between the xylose metabolizing cells in the xylose alone medium and those in the xylose fermentation phase in the mixed sugar medium, and the differences mainly involved sulfur metabolism. When the transcriptional profiles were compared between glucose fermentation state and xylose fermentation state, we found the expression patterns of hexose transporters and glucose signaling pathway dif‑ fered in response to different sugar sources, and the expression levels of the genes involved in gluconeogenesis, the glyoxylate and tricarboxylic acid cycles and respiration increased with xylose, indicating that the xylose‑ metabolizing cells had high requirements for maintenance energy and lacked the carbon catabolite repression capability. The effect of carbon catabolite repression by glucose lasted after glucose depletion for specific genes to different extents. Keywords: Transcriptome, Saccharomyces cerevisiae, Xylose fermentation, Bioethanol, Glucose and xylose cofermentation In the past two decades, fermentation of xylose to eth- Introduction anol has been achieved in S. cerevisiae by genetic engi- Lignocellulosic biomass has been recognized as a sustain- neering. Through expression of the heterogeneous xylose able source for fuel ethanol production without affecting metabolic pathway—either xylose reductase-xylitol dehy- the food and feed markets. It is of economic interest to drogenase (XR-XDH) or xylose isomerase (XI)—S. cer- convert all lignocellulosic sugar fractions, predominantly evisiae can convert xylose to xylulose, which can then glucose and xylose, into ethanol at sufficiently high rates be natively catabolized (Matsushika et  al. 2009a). The and yields. However, Saccharomyces cerevisiae, which is xylose-utilizing capacity of the recombinant strains can widely used in bioethanol plants due to its high fermen- be further optimized by enhancing the downstream met- tation efficiency and process robustness, cannot ferment abolic pathway rationally or through evolutionary engi- xylose (Batt et al. 1986). neering (Peng et al. 2012). However, recombinant strains strongly prefer glucose over xylose, and therefore the *Correspondence: tangyq@scu.edu.cn co-consumption remains a challenge. What’s more, the College of Architecture and Environment, Sichuan University, No. 24, specific ethanol productivity from xylose was an order of South Section 1, First Ring Road, Chengdu 610065, Sichuan, China © 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Zeng et al. AMB Expr (2016) 6:51 Page 2 of 13 magnitude lower than that from glucose, despite tremen- attenuate the possible impact of different growth rates on dous efforts, and the ethanol yield from xylose was lower transcriptome analysis. This is the first study investigat - than that from glucose as well (Matsushika et al. 2009b). ing transcriptional difference of an industrial S. cerevi - To reveal the major reasons for the suboptimal fermen- siae strain in both single sugar media and mixed sugar tation of xylose by recombinant S. cerevisiae strains, the medium. difference in transcriptional response between xylose and glucose fermentation has been examined in the past Materials and methods decade. It has been recognized that S. cerevisiae does not Strains and batch fermentation conditions sense xylose as a fermentable carbon source (Jin et  al. The recombinant xylose-utilizing S. cerevisiae used in this 2004). Early transcriptional analysis on xylose was con- study was KF7M-16 derived from KF-7 (Kida et al. 1992). ducted during aerobic growth, and revealed that xylose Two integrative plasmids were transformed into KF-7: was neither recognized as a fermentable carbon source pIUX1X2XK (contains XYL1, XYL2 from Scheffersomyces nor as a respirative carbon source (Salusjärvi et al. 2006). stipitis and XKS1 from S. cerevisiae) and pIWBGL1 (con- Using transcriptome and proteome, the difference in car - tains BGL1 from Aspergillus aculeatus) (Li et al. 2015). bon source signaling and catabolite repression was stud- The single and mixed sugar media used in the batch ied in the aerobic batch fermentation of either glucose or fermentation were as follows: 6  % YPD (20  g/L peptone, xylose, and it has been suggested that cells metabolizing 10 g/L yeast extract, and 60 g/L glucose), 4 % YPX (20 g/L xylose were neither in a completely repressed nor in a peptone, 10  g/L yeast extract, and 40  g/L xylose), and derepressed state (Salusjärvi et  al. 2008). The transcrip - 10 % YPDX (20 g/L peptone, 10 g/L yeast extract, 60 g/L tional difference in S. cerevisiae growing anaerobically glucose, and 40 g/L xylose). in either glucose or xylose was subsequently analyzed Batch fermentations were conducted in 300-ml cotton- (Matsushika et  al. 2014; Runquist et  al. 2009), indicat- plugged shake-flasks containing 100 ml medium, at a stir - ing xylose was recognized as a non-fermentable carbon ring speed of 200 rpm controlled by a HS-6DN magnetic source and induced the expression of stress-responsive stirrer (As One, Japan). The temperature was maintained genes. More recently, the specific regulatory response of at 35 °C in a thermostatic water bath. S. cerevisiae to xylose was quantified at a range of culti - Working cultures of yeasts were obtained following an vation times in anaerobic glucose-xylose mixed medium activation period at 30  °C for 24  h on a 2  % YPD plate (Alff-Tuomala et  al. 2016), and xylose was observed to (20 g/L peptone, 10 g/L yeast extract, 20 g/L glucose with delay the glucose-dependent repression of specific genes 2 % agar) and then pre-cultivating the yeast cells at 30 °C in the mixed culture. for 16 h in 5 % YPD (20 g/L peptone, 10 g/L yeast extract, To the best of our knowledge, the transcription and and 50  g/L glucose). The harvested cells were inocu - regulatory responses induced by xylose have been ana- lated into batch fermentation at 5  g-dry cell weight/L lyzed either in single sugar or mixed sugar cultures. As (g-DCW/L). the transcriptional profile is dependent on different host For the pre-experiment, the culture was sampled every strain background (Feng and Zhao 2013a), a systematic 2  h in the fast fermenting stage to monitor cell growth transcriptional analysis of a particular S. cerevisiae strain and metabolites concentration. The experiments for under different fermentation conditions, including in microarray analysis were conducted in biological dupli- media with a single sugar and mixed sugars, would be cate, and samples were collected at the time points indi- useful to identify the genetic factors responsible for the cated in the results section. discrepancy in sugar source utilization efficiency. Indus - trial S. cerevisiae strains generally have a superior ethanol RNA extraction production efficiency and inhibitor tolerance compared After cell collection, the total RNA was extracted using to laboratory strains. In this study, we therefore carried the Takara Yeast RNAiso Kit according to the manufac- out batch fermentations of KF7M-16, an XR- and XDH- turer’s protocol. RNA quality and concentration were expressing industrial flocculating strain, in both sin - measured by agarose gel electrophoresis and NanoDrop gle sugar medium (either glucose or xylose) and mixed 2000/2000C (Thermo Scientific, USA). sugar medium (including the glucose fermentation phase and the xylose fermentation phase), and analyzed the Microarray analysis global transcriptomes based on microarrays to identify Microarray analysis was performed using the 7G Affy - the molecular response to different fermentation states. metrix GeneChip Yeast Genome 2.0 Array (CapitalBio It has been reported that cellular growth rate has great Tec., Beijing). The isolated total RNA was cleaned up influence on transcriptional regulation (Regenberg et  al. with RNeasy Kit (Qiagen, Germany). 100 ng of total RNA 2006), we therefore adopt high-density inoculations to was used for cDNA synthesis and produce biotin-tagged Zeng et al. AMB Expr (2016) 6:51 Page 3 of 13 cRNA with GeneChip IVT Labeling kit (Affymetrix). 15  µg fragmented cRNA, together with contol oligo B2 and eukaryotic hybridization controls, was hybrid- ized to each GeneChip array at 45  °C for 16  h (Affym - etrix GeneChipHybridization Oven 640) according to manufacturer’s instructions. After hybridization, the GeneChip arrays were washed, stained with streptavidin phycoerythrinonan (SAPE) with Affymetrix Fluidics Sta - tion 450, and then scanned with the Affymetrix Gene - Chip Scanner 3000 7G. The data extraction and analysis were carried out using Affymetrix GeneChip Command Console Software. The microarray data can be accessed through GEO accession through GSE80748. To identify altered gene expression, the averages of biological dupli- cates were compared, and fold changes in gene expres- sion ≥2 were considered significant. The gene annotation information was based on the Saccharomyces genome database. The pathway terms were enriched using the KEGG orthology based annotation system (KOBAS). Quantitative real‑time PCR The cDNA was reverse-transcribed from total RNA using the Takara PrimeScript RT reagent Kit with gDNA Eraser (Perfect Real Time). qPCR analysis was performed ® ™ using the Takara SYBR Premix Ex Taq II (Tli RNaseH Plus) following manufacture’s manual. ACT1 was used as the normalization standard. The primers used are listed in Additional file 1: Table S1. Fig. 1 Time‑ dependent fermentation profile of KF7M ‑16 in a YPD medium containing 60 g/L glucose (6 % YPD), b YPX medium Analytical methods for substrate and products containing 40 g/L xylose (4 % YPX), and c YPDX medium containing Metabolite analysis was conducted as previously 60 g/L glucose and 40 g/L xylose (10 %YPDX). Black diamond OD ; described (Tang et  al. 2006). Glucose and xylose in black square glucose; empty square xylose; empty up‑pointing triangle ethanol; and error marks xylitol. The arrows (denoted by G, X, GX1, and the fermentation medium was determined by a HPLC GX2) indicate the times at which samples were taken for transcription equipped with a fluorescence detector (RF-10A ). Etha- XL analysis nol was measured by GC with a FID detector and 2-pro- panol was used as the internal standard. Xylitol was assayed by HPLC equipped with an Aminex HPX-87H column (300 × 7.8 mm) (Bio-Rad, USA) and a refractive YPDX. The utilization rate of xylose was approximately a index detector. magnitude slower than that of glucose (2.88 g/L/h in 4 % YPX in the first 8  h, 17.27  g/L/h in 6  % YPD in the first Results 2  h and 16.80  g/L/h in 10  % YPDX in the first 2  h), and Experimental design and batch fermentation after 24  h of metabolism, KF7  M-16 almost consumed Batch fermentations were carried out in YP medium con- all of the xylose in both the 4  % YPX and 10  % YPDX. taining 60  g/L glucose (6  % YPD) or 40  g/L xylose (4  % It is worth noting that the presence of glucose repressed YPX) as single sugar medium, as well as in YP medium the utilization of xylose, as in 10  % YPDX the xylose containing 60  g/L glucose and 40  g/L xylose as mixed consumption rate was 1.38  g/L/h in the first 4  h and sugar medium (10  % YPDX). To eliminate the effect of 3.03 g/L/h in 4–8 h. Therefore the fermentation profile in different growth rates between glucose and xylose uti - 10 % YPDX can be divided into the glucose fermentation lization on the transcriptional profile, a relatively high phase and the xylose fermentation phase. inoculum density of 5  g-DCW/L was chosen. A pre- For transcription analysis, experiments were con- experiment was conducted to grasp the fermentation ducted in duplicate. Samples of glucose-grown cells were profile of KF7M-16 (Fig.  1). Glucose was depleted in harvested at 2  h while samples of xylose-grown cells the first 4  h of fermentation, both in 6  % YPD and 10  % were harvested at 7  h. The samples taken at 2  h in 6  % Zeng et al. AMB Expr (2016) 6:51 Page 4 of 13 YPD were denoted as G, and the residual glucose was 16.4  ±  1.2  g/L, with 16.4  ±  0.3  g/L ethanol produced at a yield of 0.46  ±  0.01  g/g at that time point. The sam - ples taken at 7  h in 4  % YPX were denoted as X, and at that time point the residual xylose was 20.4  ±  0.1  g/L, with 6.2  g/L ethanol and 1.7  ±  0.1  g/L xylitol produced at yields of 0.35  ±  0.1 and 0.10  ±  0.003  g/g respec- tively. The samples taken at 2  h in 10  % YPDX (during glucose fermentation stage) were denoted as GX1, and during this first 2  h of fermentation, 37.1  ±  5.1  g/L glu- Fig. 2 Overview of differently expressed gene numbers between cose and 2.3  ±  1.3  g/L xylose were consumed, produc- different fermentation conditions. G glucose fermentation state in ing 17.2  ±  1.8  g/L ethanol at a yield of 0.44  ±  0.02  g/g. the glucose alone medium; X xylose fermentation state in the xylose The samples taken at 7  h in 10  % YPDX (during xylose alone medium; GX1 glucose fermentation phase in the mixed sugar; fermentation stage) were denoted as GX2, and the glu- GX2 xylose fermentation phase in the mixed sugar. C1–C4 indicate the specific pairwise comparisons. Averages of biological duplicates cose had been depleted for at least 3  h by then, and the were compared. The numbers indicate differently expressed genes in residual xylose was 19.5  ±  1.0  g/L, with 27.8  ±  1.8  g/L each comparison ethanol as well as 4.6 ± 0.3 g/L xylitol produced at yields of 0.38 and 0.26 ± 0.04 g/g respectively. It’s notable that the ethanol yield from xylose was 24  % lower than that from glucose, and that the xylitol yield from xylose was X vs. G 160 % higher in the mixed sugar medium than in the sin- To reveal the molecular basis for the fermentation abil- gle sugar medium. ity discrepancy between glucose and xylose, the tran- scriptional difference in S. cerevisiae growing in either Overview of transcriptional differences across four the glucose alone or the xylose alone medium has long conditions been analyzed, both aerobically and anaerobically. In To obtain the general information of transcriptional our study with the low aeration condition, a total of 487 responses to the different sugar sources, the microarray genes showed significantly different expression levels data was organized into four relevant pairwise compar- (fold change ≥2) in C2, with 281 upregulated and 206 isons: GX1 vs. G (comparison 1  =  C1); X vs. G (com- downregulated genes (Additional file  1: Table S3). The parison 2  =  C2); GX2 vs. GX1 (comparison 3  =  C3); significantly enriched pathway terms (p value  <  0.1) are and GX2 vs. X (comparison 4  =  C4). The averages of listed in Table 1. biological duplicates were compared. The numbers of Secondary metabolites are organic compounds that are genes with significantly changed expression levels (fold not directly involved in the normal growth or reproduc- change ≥2) between the corresponding conditions are tion of the cells, and often have a role in stress defense. shown in Fig.  2. Differentially expressed genes can be A secondary metabolite forms frequently near or in the identified and enriched for pathway terms using the stationary phase of growth. The KEGG term of “Biosyn - KEGG Orthology Based Annotation System (KOBAS). thesis of secondary metabolites” (ID: sce01110) involves a number of pathways, including carbon metabolism and GX1 vs. G biosynthesis of amino acids. Only 36 genes stood out as being differently expressed in The expressional changes in carbon metabolism C1 (Additional file  1: Table S2), with 16 upregulated and directly reflect the different metabolic states in differ - 20 downregulated genes. This suggested that the transcrip - ent sugar sources. Genes involved in the citrate and gly- tional profile of cells during glucose fermentation phase oxylate cycles, which are repressible by glucose (Gancedo in the mixed sugar medium was nearly identical to those 1998), were upregulated in the xylose alone medium, growing in glucose alone medium. Since xylose is not a indicating that xylose lacked the carbon catabolite natural carbon source for S. cerevisiae, recombinant strains repression capability. Genes involved in the central car- have a stronger preference for glucose rather than xylose. bon metabolism across the four conditions will be further KF7M-16 utilized glucose first in the mixed sugar medium. discussed later. Although a meager amount of xylose was consumed in the The expression levels of the genes involved in sulfur first 2 h of fermentation, the added xylose seemed power - metabolism were downregulated in C2. The detailed less to affect the glucose dominated transcriptional profile. transcriptional changes will be further discussed in C4. Zeng et al. AMB Expr (2016) 6:51 Page 5 of 13 Table 1 Enriched pathways of the differently expressed genes Comparison Term Database ID p value C2 Biosynthesis of secondary metabolites KEGG PATHWAY sce01110 2.21E−05 Carbon metabolism KEGG PATHWAY sce01200 2.46E−04 Histidine biosynthesis PANTHER P02747 5.18E−04 Glyoxylate and dicarboxylate metabolism KEGG PATHWAY sce00630 6.25E−04 Biosynthesis of amino acids KEGG PATHWAY sce01230 1.09E−03 Metabolic pathways KEGG PATHWAY sce01100 2.96E−03 Sulfur metabolism KEGG PATHWAY sce00920 3.70E−03 Citrate cycle ( TCA cycle) KEGG PATHWAY sce00020 5.86E−03 C3 Biosynthesis of secondary metabolites KEGG PATHWAY sce01110 4.19E−03 Heme biosynthesis PANTHER P02746 9.02E−03 Histidine metabolism KEGG PATHWAY sce00340 1.20E−02 Tryptophan metabolism KEGG PATHWAY sce00380 1.42E−02 Porphyrin and chlorophyll metabolism KEGG PATHWAY sce00860 1.42E−02 C4 Sulfur metabolism KEGG PATHWAY sce00920 7.11E−04 superpathway of sulfur amino acid biosynthesis BioCyc PWY‑821 1.44E−03 Arginine biosynthesis PANTHER P02728 1.58E−03 Sulfate reduction I (assimilatory) BioCyc SO4ASSIM‑PWY 9.14E−03 http://bioinfo.capitalbio.com/mas3/ GX2 vs. GX1 uroporphyrinogen (MET1, MET8) showed an increased The transcription profiles in the mixed sugar medium expression level. were compared between the xylose fermentation stage It has been previously suggested that the increased and the glucose fermentation stage. A total of 499 and expression level of the genes related to tryptophan bio- 377 genes were upregulated and downregulated respec- synthesis might confer ethanol stress tolerance to yeast tively in C3 (Additional file  1: Table S4). The number of cells (Hirasawa et  al. 2007). In our study, the expression differently expressed genes in C3 was 80 % more than in level of the genes related to tryptophan degradation C2, although both comparisons were carried out between (BNA1, BNA4, BNA7, ARO8) decreased when the tran- glucose fermentation state and xylose fermentation scriptional profile of xylose fermentation stage in the state. The transcriptional shift induced by different sugar mixed sugar medium was compared to that of the glu- source utilization was more complex in the mixed sugar cose fermentation stage, corroborating the involvement medium, mainly due to the prolonged effect of glucose of tryptophan in response to ethanol stress. repression and the stress induced by the accumulated Genes involved in histidine biosynthesis (HIS1, HIS7, ethanol. The enriched pathway terms (p < 0.02) are listed HIS2), which were previously reported to be significantly in Table 1. downregulated in the presence of ethanol (Li et al. 2010), Similarly to C2, the KEGG term “biosynthesis of sec- also had a decreased expression level in C3. ondary metabolites” was enriched with the lowest p value in C3. GX2 vs. X Heme is a cofactor consisting of a ferrous ion con- Since the metabolic activity during the xylose fermenta- tained in the center of a large heterocyclic organic tion phase in mixed sugar medium was similar to that ring called a porphyrin. Heme is the component of in xylose alone medium, it is interesting to observe that cytochrome which is primarily responsible for the gen- when the transcription profile of GX2 was compared to eration of ATP via electron transport. Although the that of X, 682 genes were differently expressed, with 375 expression level of genes encoding respirative enzymes upregulated and 307 downregulated genes (Table S5, in increased during the xylose fermentation stage relative the supplementary material). Significantly enriched path - to the glucose fermentation stage in the mixed sugar way terms (p < 0.01) are listed in Table 1. medium (Table  4), several genes involved in the heme Sulfur amino acid biosynthesis can be divided into biosynthesis from uroporphyrinogen (HEM12, HEM13, three parts: sulfate assimilation, cysteine and methio- HEM15) showed a decreased expression level. Whereas nine biosynthesis, and S-adenosyl-l -methionine (SAM) the genes involved in the siroheme biosynthesis from biosynthesis (Thomas and SurdinKerjan 1997). SAM is Zeng et al. AMB Expr (2016) 6:51 Page 6 of 13 probably second only to ATP in the variety of reactions additional broad substrate sugar transporters (Hamacher for which it serves as a cofactor, and is also a precursor et al. 2002). The genes encoding transporters showed dif - of glutathione, which is a major cellular anti-oxidant ferent transcriptional patterns on different sugar sources (Caro and Cederbaum 2004). Compared to their expres- (Fig.  4, Table  2). The transcriptional levels of the high- sion levels in the glucose alone medium as a control, affinity glucose transporter genes HXT2 and HXT4, as the expression of genes involved in sulfur amino bio- well as the low-affinity glucose transporter gene HXT3 synthesis was higher in GX2 and was lower in X. The were downregulated with xylose, in both the xylose alone comparative expression ratios are presented in Fig.  3. medium and during the xylose fermentation phase in It is reasonable to speculate that the demand for SAM the mixed sugar medium. On the other hand, the non- might be higher during the xylose fermentation stage in fermentable carbon source-inducible transporter genes the mixed sugar medium than in xylose alone medium. HXT5 and HXT13 were upregulated with xylose, indi- Considering that SAM is protective against a variety of cating that xylose was not recognized by S. cerevisiae toxic oxidative agents, this difference might be due to as being fermentable. HXT15 and HXT16, repressed by the metabolites accumulated in the process of glucose high levels of glucose, were induced nearly 40 times in fermentation. the xylose alone medium, but not during the xylose fer- The expression of genes involved in arginine biosynthe - mentation phase in the mixed sugar medium, suggesting sis from glutamine (CPA2, ARG3, ARG1, and ARG4) also that the effect of glucose repression could last after glu - increased in C4. cose had been depleted. HXT6//7 were expressed at high levels under all four conditions. Other genes in the HXTs Carbon source sensing and signaling family did not show obvious differences in expression Uptake of xylose by S. cerevisiae has been proposed to be level, and all of them were negligibly expressed, with the mediated unspecifically by its hexose-transport system, exception of HXT9. As deduced from mRNA level of dif- which is composed of 18 genes from the HXTs family and ferent transporters (Table  2), HXT3 and HXT4, together Fig. 3 The expression ratio of the genes involved in sulfur amino acid biosynthesis. The fold change of expression is presented for GX2 (left panel) and X (right panel), both compared to the transcriptional state in the glucose alone medium (G) as a control. Averages of biological duplicates were compared. The red panel indicates that the transcription level of gene increased over twofold; the green panel indicates the transcription level of gene decreased over twofold Zeng et al. AMB Expr (2016) 6:51 Page 7 of 13 Fig. 4 Expression profiles of the hexose transporter genes and comparison of the transcriptional changes in the genes involved in glucose sens‑ ing and repression network. The relative expression of hexose transporter genes is indicated with different colors. For the regulator genes, the fold change of expression is presented for GX2 (left panel) and X (right panel), both compared to G as a control. Averages of biological duplicates were compared. The red panel indicates the transcription level of gene increased over twofold; the green panel indicates the transcription level of gene decreased over twofold Table 2 The hybridization signal (MAS5.0 signal intensity) of  the genes encoding hexose-transport system in  different fermentation states Gene G GX1 GX2 X HXT1 25.9 ± 2.9 29.7 ± 11.5 38.8 ± 15.8 24.3 ± 1.9 HXT2 700.8 ± 76.9 500.1 ± 236.3 125.4 ± 13.0 114.8 ± 47.5 HXT3 3937.3 ± 299.5 3039.1 ± 828.4 643.8 ± 235.8 555.2 ± 106.5 HXT4 3542.4 ± 626.0 3868.6 ± 17.7 1389.3 ± 118.5 1136.8 ± 207.0 HXT5 34.6 ± 1.2 48.9 ± 30.0 220.1 ± 80.3 142.9 ± 59.3 HXT6//7 1756.3 ± 302.6 1801.2 ± 181.3 2031.2 ± 157.8 1949.0 ± 7.5 HXT8 24.9 ± 0.2 23.2 ± 0.7 22.3 ± 0.9 22.6 ± 0.6 HXT9 263.5 ± 56.7 256.0 ± 50.2 242.6 ± 9.4 190.3 ± 8.7 HXT10 7.9 ± 0.2 7.6 ± 0.2 7.0 ± 0.1 8.0 ± 0.2 HXT11 8.1 ± 0.2 5.8 ± 0.8 7.9 ± 1.5 8.4 ± 1.1 HXT13 17.0 ± 6.9 13.6 ± 0.6 39.8 ± 1.6 131.3 ± 19.3 HXT14 7.9 ± 0.1 7.9 ± 0.3 11.7 ± 0.8 7.7 ± 0.3 HXT15//16 12.0 ± 1.1 13.2 ± 0.9 14.5 ± 0.9 536.4 ± 110.5 GAL2 38.6 ± 0.3 50.1 ± 6.9 79.2 ± 5.1 58.3 ± 2.7 Values are given as the average and standard deviation of two biological duplicates “//” indicates genes detected by the same probe set because of their sequence similarity with HXT6 and HXT7 might serve as the main transport- It has been previously reported that genes encod- ers in the glucose fermentation state, while HXT6, HXT7 ing the glucose sensor proteins Snf3p and Rgt2p, which as well as HXT4 might be the main transporters in the mediate signal for the presence of glucose at low or xylose fermentation state. high concentration respectively, had higher expression Zeng et al. AMB Expr (2016) 6:51 Page 8 of 13 levels with xylose (Salusjärvi et  al. 2008). In our study, both have been previously reported to be poor aldo–keto SNF3 and RGT2 had stable expression levels across all reductases (Chang et  al. 2007). The expression levels of four conditions. Glucose binds to Snf3 and Rgt2, induc- the endogenous genes SOR1 and SOR2 increased during ing them to bind to Mth1 and Std1, and finally leading xylose fermentation, and to a much stronger extent in the to the degradation of Mth1 and Std1, which are two Rgt- xylose alone medium than in the mixed medium, which 1corepressors (Zaman et  al. 2008). The expression levels might have accounted for the lower observed xylitol yield of the corepressor genes MTH1 and STD1 decreased in the xylose alone medium than in the mixed medium. with xylose, and that of RGT1 slightly increased (Fig.  4). For SOR1//2, the microarray data was validated by RT- Rgt1 is a DNA-binding protein that represses the hexose qPCR (Table  3). Even though the SOR1//2 expression transporter genes HXT2-4, as well as hexokinase gene level increased in magnitudes in X as compared to glu- HXK2. The expression levels of these genes were corre - cose fermentation state (G and GX1), the transcript spondingly downregulated. The xylose-dependent upreg - abundance of SOR1//2 was still lower than XYL2 under ulation of RGT1, irrespective of oxygen availability, has xylose fermentation state. been previously observed (Alff-Tuomala et al. 2016). The The major enzymes in the non-oxidative pentose phos - expression levels of the transcription factor Adr1, which phate pathway, TAL1 and TKL1 encoding transaldo- is negatively regulated by PKA in glucose growing cells lase and transketolase (Matsushika et  al. 2012), did not (Zaman et  al. 2008), were upregulated with xylose. Glu- express significantly differently across all conditions, cose metabolism activates the transcriptional repressor but the expression of their minor isoenzyme NQM1 Mig1 to move into the nucleus and repress its many tar- and TKL2 was induced during xylose fermentation. The gets (Kuttykrishnan et  al. 2010). The expression level of expression level of NQM1 is usually induced during Mig1 decreased with xylose, while that of Cat8, which is diauxic shift and TKL2 is regulated in a Msn2/4p manner. repressed by Mig1, increased with xylose. Transcription Another non-oxidative PPP gene, RKI1, was expressed at factor Cat8 activates gene expression required for gluco- a lower level on xylose. In the oxidative pentose phos- neogenesis during growth in the absence of glucose, the phate pathway, the expression levels of ZWF1 encoding transcriptional changes of which will be discussed in the glucose-6-phosphate dehydrogenase and GND2 encod- following section. ing 6-phosphogluconate dehydrogenase increased on xylose. Central carbon metabolism In three structural genes encoding enzymes that cata- Xylose was channeled to ethanol via the heterologous lyze the phosphorylation of glucose to glucose 6-phos- xylose assimilating pathway, pentose phosphate pathway, phate, HXK2 was repressed with xylose, especially glycolysis pathway and alcohol fermentation pathway. To during the xylose fermentation phase in the mixed sugar pinpoint the effect of xylose on the fermentation state, medium. Hxk2p is known to provide the main sugar- the genes expression in the central carbon metabolism phosphorylating capability during growth on fermentable was investigated in greater detail, using the transcrip- carbon source, and becomes repressed on non-fermenta- tional state when cultivated in glucose alone medium as ble carbon source by transcription factor Rgt1 (Palomino a control (Fig. 5). et al. 2005). Since the heterologous XYL1 and XYL2 genes from S. Most glycolysis genes showed stable expression levels stipitis were not included in the S. cerevisiae genechips, across all four conditions, except for two genes encod- their expression levels were determined by RT-qPCR ing homolog of Gpm1p phosphoglycerate mutase, GPM2 (Table 3). The heterologous XYL1 and XYL2 did not show and GPM3, were slightly downregulated on xylose, and significant difference across all the tested conditions, but these two genes may be non-functional. the S. cerevisiae genes encoding enzymes with xylose In the alcohol fermentation and by-product path- reductase and xylitol dehydrogenase function manifested way, two genes encoding the minor isoform of pyruvate different expression patterns (Fig.  5). The expression decarboxylase, PDC5 and PDC6, respectively, showed level of GCY1 increased as long as the medium con- decreased and increased expression levels with xylose. tained xylose. Another transcription study using mixed PDC5 is repressed by thiamine, and PDC6 is induced sugar medium also indicated expression level of GCY1 during sulfur limitation. Yet several genes involved in increased even in glucose fermentation phase (Alff- thiamine biosynthesis (THI genes) showed decreased Tuomala et  al. 2016). During the xylose fermentation expression levels with xylose. Most ALD genes involved phase in the mixed sugar medium, the expression level in acetate formation showed increased expression of YJR096W increased and that of YDL241W decreased, level during the xylose fermentation stage in the mixed while both genes did not express differently in xylose sugar medium, including cytosolic aldehyde dehydroge- alone medium, compared to glucose alone medium, and nase ALD2, ALD3, ALD6, and mitochondrial aldehyde Zeng et al. AMB Expr (2016) 6:51 Page 9 of 13 Fig. 5 Expression profiles of genes involved in the central carbon metabolism. The fold change of expression is presented for GX1 (left panel), GX2 (middle panel) and X (right panel), compared to G as a control. Averages of biological duplicates were compared. The red panel indicates the tran‑ scription level of gene increased over twofold; the green panel indicates the transcription level of gene decreased over twofold dehydrogenase ALD4. This upregulation was likely to be is sensed as a non-fermentable carbon source by S. a response to the higher ethanol level at the sampling cerevisiae. time of GX2. The expression level of ACS1 , encoding The expression levels of several genes in the tricarbox - a synthetase to form acetyl coenzyme A from acetate, ylic acid cycles were upregulated with xylose. Transcripts increased with xylose, once again indicating that xylose of CIT2 and CIT3, encoding enzymes that catalyze the Zeng et al. AMB Expr (2016) 6:51 Page 10 of 13 Table 3 Expression level of  xylose metabolizing genes genes in the glyoxylate shunt, ICL1 and MLS1 encoding by RT-qPCR isocitrate lyase and malate synthase respectively, together with CIT2 encoding peroxisomal citrate synthase were Sugar XYL1/ACT1 XYL2/ACT1 SOR1//2/ACT1 also induced on xylose. G 1.9 ± 0.25 2.53 ± 0.52 7.35E−5 ± 1.2E−6 In the glycerol catabolism pathway, the expression X 1.9 ± 0.42 2.36 ± 0.58 1.13E−1 ± 2.1E−2 level of the glycerol-producing gene RHR2 encoding GX1 2.0 ± 0.22 2.17 ± 0.34 6.84E−5 ± 3.3E−6 glycerol-1-phosphatase increased with xylose, which GX2 2.19 ± 0.45 1.28 ± 0.28 1.08E−3 ± 5.7E−5 is contrary to an earlier finding with anaerobic batch Values are given as the average and standard deviation of two duplicates fermentation (Matsushika et  al. 2014). The expression levels of glycerol-consuming genes, GUT2 encoding mitochondrial glycerol-3-phosphate dehydrogenase condensation of acetyl coenzyme A and oxaloacetate and DAK2 encoding dihydroxyacetone kinase, also to form citrate, increased with xylose, especially in the increased with xylose. DAK2 was strongly induced dur- xylose alone medium. The levels of ACO2 , encoding a ing the xylose fermentation stage in the mixed sugar putative mitochondrial aconitase and repressed by eth- medium, likely due to its involvement in stress adap- anol, decreased during the xylose fermentation stage tation. Similar to DAK2, STL1 encoding the glycerol in the mixed sugar medium. As for the IDP genes that proton symporter of the plasma membrane was also encode isocitrate dehydrogenase, the expression level strongly induced during the xylose fermentation stage of IDP1 encoding mitochondrial isocitrate dehydroge- in the mixed sugar medium, but not in the xylose alone nase did not change, but that of IDP2 and IDP3, encod- medium. STL1 was the gene with the largest upregula- ing NADP specific cytosolic and peroxisomal isocitrate tion ratio (70.90) in C4, reflecting the different osmotic dehydrogenase respectively, increased with xylose. The state between GX2 and X, because STL1 is documented KGD genes (KGD1 and KGD2) encoding components of to be strongly but transiently induced when cells are the mitochondrial alpha-ketoglutarate dehydrogenase subjected to osmotic shock. complex, as well as the SDH genes (SDH1, SDH2, SDH3, and SDH4) that encode subunits of succinate dehydro- genase, had slightly increased expression levels with Table 4 The expressional fold changes of  the genes xylose. The expression level of MDH2, encoding cyto - responsible for  respiratory metabolism and  ATP synthesis plasmic malate dehydrogenase, which is also involved for GX1, GX2, and X, compared to G as a control (averages in the gluconeogenesis and glyoxylate cycles, increased of biological duplicates were compared) with xylose. Because 2 carbons lost as CO in one TCA Genes GX1 GX2 X Description cycle, the strong activity of the TCA cycle could be the reason for the lowered ethanol yield from xylose. Cor- HAP4 1.11 1.55 2.13 Transcriptional activator of respiratory genes roborating with the upregulated TCA cycle pathway, NDE1 0.82 5.54 5.65 Mitochondrial external NADH dehydrogenase the expression levels of a number of genes encoding NDE2 1.29 2.92 3.39 respiratory enzymes increased with xylose (Table  4). NDI1 1.21 2.55 2.27 Ubiquinone oxidoreductase The expression level of HAP4, a transcriptional activa- QCR2 1.14 1.99 2.12 Subunits of ubiquinol cytochrome c reduc‑ tor and global regulator of respiratory gene expression, tase complex QCR9 1.18 2.38 1.67 also slightly increased with xylose. As a result, a sig- QCR8 1.10 2.07 1.49 nificant amount of ATP was assumed to be produced QCR10 1.24 2.06 1.87 in xylose metabolism, indicating high requirements for CYC3 0.88 1.80 2.01 Cytochrome c heme lyase maintenance energy to utilize xylose by recombinant S. CYC7 1.18 3.49 1.63 Cytochrome c isoform 2 cerevisiae. COX5A 0.99 2.56 1.96 Subunits of cytochrome c oxidase The gluconeogenic gene FBP1 , encoding fructose- COX7 1.20 2.78 2.30 1,6-bisphosphatase, was strongly induced with xylose. COX12 1.00 1.92 2.09 The transcription levels of other genes involved in glu - COX13 0.96 2.81 1.96 coneogenesis, such as PCK1 encoding phosphoenol- ATP2 1.05 1.15 1.91 Subunits of ATP synthase pyruvate carboxykinase and MDH2, also increased with ATP3 1.06 1.38 1.63 xylose. The upregulation ratio was higher in the xylose ATP4 1.15 1.39 1.78 alone medium than during the xylose fermentation phase ATP5 1.05 0.93 1.62 in the mixed sugar medium. The gluconeogenesis path - ATP14 1.06 1.27 1.62 way is repressed by glucose, and our results indicated ATP16 1.10 1.34 1.67 that this pathway was not fully derepressed after glucose ATP18 1.10 1.20 1.78 had been depleted. The expression levels of the specific Zeng et al. AMB Expr (2016) 6:51 Page 11 of 13 second step of gluconeogenesis were increased, and to Discussion a significantly larger extent in the xylose alone medium The transcription profiles of an industrial recombinant than in the xylose fermentation phase in the mixed-sugar S. cerevisiae strain was compared between different fer - medium. The generation of phosphoenolpyruvate from mentation states of glucose and xylose both in single oxaloacetate is accompanied by a loss of 1 carbon as CO , sugar and mixed sugar media. The presence of glucose and consumption of 1 ATP. The upregulation ratio of repressed the utilization of xylose, and it is found that FBP1 was consistent to that of PCK1. the transcriptome of the glucose fermentation phase As to the regeneration of NAD , since alcoholic fer- in the mixed-sugar medium was very similar to that in mentation itself is a redox-neutral process (the NADH the glucose alone medium. A recently published gene reduced in ethanol formation accounts for the NADH expression study using mixed sugar medium found that generated in the glyceraldehyde-3-phosphate dehy- respiratory genes were not fully repressed when xylose drogenase reaction), the accumulated NADH must was present with abundant glucose (Alff-Tuomala et  al. be otherwise reoxidized. Despite the impermeability 2016), which was not observed in our study. On the other of the inner mitochondrial membrane for NADH and hand, we found that although the transcriptome highly NAD , both cytosolic and mitochondrial NADH can differed between the xylose metabolizing cells cultured be reoxidized by the respiratory chain in S. cerevisiae. in the xylose alone medium and those in the xylose fer- The cytosolic NAD can be generated via mitochon- mentation phase of the mixed sugar medium, the xylose drial external NADH hydrogenases (Bakker et al. 2001), consumption rate was nearly identical under these two encoded by NDE1 and NDE2. Alternatively, cytosolic conditions. It appears that S. cerevisiae has adequately NADH can be reoxidized by the respiratory chain via evolved mechanisms to respond to metabolites accumu- the glycerol- 3-phosphate shuttle consisting of cytosolic lated in the process of glucose fermentation. NADH-linked glycerol-3-phosphate dehydrogenase and The main challenge in the recombinant strains express - mitochondrial FAD linked glycerol-3-phosphate dehy- ing S. stipitis XR-XDH pathway is redox imbalance, drogenase encoded by GUT2 that transfers electrons because xylose reductase prefers NADPH, whereas from cytosolic glycerol- 3-phosphate to ubiquinone. The xylitol dehydrogenase strictly utilizes NAD , leading expression level of NDE1, NDE2 and GUT2 increased to the accumulation of NADP and NADH. NADPH is significantly with xylose in our study, both in the xylose regenerated in S. cerevisiae mainly through the oxida- alone medium and during the xylose fermentation phase tive branch of the pentose phosphate pathway and the in the mixed sugar medium (Table  4 and Fig.  5). The cytosolic and peroxisomal isocitrate dehydrogenases. We expression levels of many others genes encoding respira- found that the expression levels of ZWF1 and GND2 in tory enzymes such as cytochrome c reductase that trans- oxidative PPP, and that of IDP2 and IDP3 were upregu- fers electrons from ubiquinone to cytochrome c and lated on xylose (Fig.  5), which corroborated with earlier cytochrome c oxidase, which catalyzes the oxidation of reports either in aerobic or anaerobic conditions (Mat- cytochrome c by molecular oxygen, also increased with sushika et  al. 2014; Salusjärvi et  al. 2008). It is known xylose (Table 4). that the ratio of NADP /NADPH directly influence the It is assumed that xylose is poorly metabolized to eth- activity of glucose-6-phosphate dehydrogenase, while anol because it does not repress respiration in the same the dehydrogenation of glucose-6-phosphate catalyzed manner as glucose ( Jin et al. 2004). S. cerevisiae growing by ZWF1 is the rate-limiting reaction in PPP. When one in glucose exhibits the Crabtree effect, where respira - mole of ribulose-5-P is produced from glucose-6-phos- tory growth can only be achieved with a limited sugar phate through oxidative PPP generating two moles of supply and at low specific growth rate. It is possible NADPH, one carbon is lost as CO , leading to the poor that the lack of ability to repress respiration by xylose- use of carbon substrate. metabolizing cells is the consequence of a low level of Meanwhile, the gluconeogenesis pathway was upregu- xylose uptake by S. cerevisiae. However, it has been lated to feed glucose-6-phosphate to oxidative PPP, con- previously reported that the overexpression of xylose- sistent with an earlier report (Runquist et  al. 2009). In transporting protein did not enhance xylose fermenta- our study, differences in the expression level were slight tion (Hamacher et  al. 2002). It is then possible that the for PYC1 and PYC2, which encode cytoplasmic pyru- upregulation of respiration is driven by the demand for vate carboxylase that converts pyruvate to oxaloacetate ATP and NAD . However, the metabolic flux analy - in the first step of gluconeogenesis. However, cytoplas - sis of a xylose isomerase-expressing strain, which does mic oxaloacetate can also be produced from acetyl-coA not face the NADH accumulation challenge during through the glyoxylate cycle, which was upregulated on xylose assimilation, also indicated that xylose failed to xylose (Fig.  4). The expression levels of PCK1 encoding elicit the full carbon catabolite repression response phosphoenolpyruvate carboxykinase that catalyzes the Zeng et al. AMB Expr (2016) 6:51 Page 12 of 13 References (Wasylenko and Stephanopoulos 2015). As for ATP Alff‑ Tuomala S, Salusjärvi L, Barth D, Oja M, Penttilä M, Pitkänen JP, Ruohonen metabolism, since ADP activates the Snf1 regulator by L, Jouhten P. Xylose‑induced dynamic effects on metabolism and gene protecting it against dephosphorylation (Mayer et  al. expression in engineered Saccharomyces cerevisiae in anaerobic glucose‑ xylose cultures. Appl Microbiol Biotechnol. 2016;100(2):969–85. 2011), it is tempting to infer, from the increased Snf1- Bakker BM, Overkamp KM, Maris AJAV, Kötter P, Luttik MAH, Dijken JPV, Pronk activated transcription factors Adr1 and Cat8 expres- JT. Stoichiometry and compartmentation of NADH metabolism in Sac- sion levels as well as the decreased Snf1-deactivated charomyces cerevisiae. FEMS Microbiol Rev. 2001;25:15–37. Batt CA, Carvallo S, Easson DD, Akedo M, Sinskey AJ. Direct evidence for Mig1 expression levels (Fig.  4), that the intracellular a xylose metabolic pathway in Saccharomyces cerevisiae. Biotechnol ADP level increased in xylose metabolizing cells. Coin- Bioeng. 1986;28:549–53. cidentally, high requirements for maintenance energy Caro AA, Cederbaum AI. Antioxidant properties of S‑adenosyl‑ l ‑methionine in 2+ Fe ‑initiated oxidations. Free Radical Biol Med. 2004;36:1303–16. by the recombinant xylose metabolizing S. cerevisiae Chang Q, Griest TA, Harter TM, Petrash JM. Functional studies of aldok ‑ eto reduc‑ was found in a metabolic analysis using XR-XDH strains tases in Saccharomyces cerevisiae. BBAM ‑ ol Cell Res. 2007;1773:321–9. (Feng and Zhao 2013b). Meanwhile, the TCA cycle and Feng X, Zhao H. Investigating host dependence of xylose utilization in recombinant Saccharomyces cerevisiae strains using RNA‑seq analysis. respiratory enzymes were found to be upregulated even Biotechnol Biofuels. 2013a;6:96. under anaerobic xylose fermenting conditions where no Feng X, Zhao H. Investigating xylose metabolism in recombinant Saccharo- ATP can be generated by oxidative phosphorylation due myces cerevisiae via C metabolic flux analysis. Microb Cell Factories. 2013b;12(1):1. to the absence of oxygen (Matsushika et al. 2014). Taken Gancedo JM. Yeast carbon catabolite repression. Microbiol Mol Biol Rev. together, we propose that it is crucial to reduce the 1998;62:334–61. maintenance energy of the xylose metabolizing strains Hamacher T, Becker J, Gardonyi M, Hahn‑Hägerdal B, Boles E. Characterization of the xylose‑transporting properties of yeast hexose transporters and to improve xylose utilization. Moreover, the regulatory their influence on xylose utilization. Microbiology. 2002;148:2783–8. network for xylose fermentation appeared sub-optimal, Hirasawa T, Yoshikawa K, Nakakura Y, Nagahisa K, Furusawa C, Katakura Y, as futile induction could occur. The regulatory network Shimizu H, Shioya S. Identification of target genes conferring ethanol stress tolerance to Saccharomyces cerevisiae based on DNA microarray could alternatively be refigured by transcription factor data analysis. J Biotechnol. 2007;131:34–44. engineering or adaptive evolution. Jin YS, Laplaza JM, Jeffries TW. Saccharomyces cerevisiae engineered for xylose metabolism exhibits a respiratory response. Appl Environ Microbiol. 2004;70:6816–25. Additional file Kida K, Kume K, Morimura S, Sonoda Y. Repeated‑batch fermentation process using a thermotolerant flocculating yeast constructed by protoplast fusion. J Ferment Bioeng. 1992;74:169–73. Additional file 1: Supplementary material. Kuttykrishnan S, Sabina J, Langton LL, Johnston M, Brent MR. A quantitative model of glucose signaling in yeast reveals an incoherent feed forward loop leading to a specific, transient pulse of transcription. Proc Natl Acad Abbreviations Sci USA. 2010;107:16743–8. XR: xylose reductase; XDH: xylitol dehydrogenase; XI: xylose isomerase; YPX: Li BZ, Cheng JS, Ding MZ, Yuan YJ. Transcriptome analysis of differential yeast extract peptone xylose; YPDX: yeast extract peptone dextrose xylose; responses of diploid and haploid yeast to ethanol stress. J Biotechnol. SAM: S‑adenosylmethionine; TCA: tricarboxylic acid; GEO: gene expression 2010;148:194–203. omnibus. Li YC, Mitsumasu K, Gou ZX, Gou M, Tang YQ, Li GY, Wu XL, Akamatsu T, Taguchi H, Kida K. Xylose fermentation efficiency and inhibitor tolerance of the Authors’ contributions recombinant industrial Saccharomyces cerevisiae strain NAPX37. Appl WYZ carried out the experiment, analyzed the batch fermentation and Microbiol Biotechnol. 2015;100:1531–42. transcriptome data, and drafted the manuscript. YQT designed the research, Matsushika A, Inoue H, Kodaki T, Sawayama S. Ethanol production from xylose helped with the data analysis, and revised the manuscript. MG and ZYX in engineered Saccharomyces cerevisiae strains, current state and per‑ helped with the experiments. KK participated in the design of the research. All spectives. Appl Microbiol Biotechnol. 2009a;84(1):37–53. authors read and approved the final manuscript. Matsushika A, Inoue H, Murakami K, Takimura O, Sawayama S. Bioethanol production performance of five recombinant strains of laboratory and industrial xylose‑fermenting Saccharomyces cerevisiae. Bioresour Technol. Acknowledgements 2009b;100:2392–8. We thank CapitalBio Technology Inc. Beijing for the help in DNA microarray Matsushika A, Goshima T, Fujii T, Inoue H, Sawayama S, Yano S. Characterization analysis. of non‑ oxidative transaldolase and transketolase enzymes in the pentose phosphate pathway with regard to xylose utilization by recombinant Sac- Competing interests charomyces cerevisiae. Enzyme Microb Technol. 2012;51:16–25. The authors declare that they have no competing interests. Matsushika A, Goshima T, Hoshino T. Transcription analysis of recombinant industrial and laboratory Saccharomyces cerevisiae strains reveals the Availability of supporting data molecular basis for fermentation of glucose and xylose. Microb Cell The microarray datasets analyzed in this article are available in Gene Expres‑ Factories. 2014;13(1):1. sion Omnibus, accession number GSE80748. Mayer FV, Heath R, Underwood E, Sanders MJ, Carmena D, McCartney RR, Leiper FC, Xiao B, Jing C, Walker PA, Haire LF, Ogrodowicz R, Martin SR, Funding Schmidt MC, Gamblin SJ, Carling D. ADP regulates SNF1, the Saccharo- This work was supported by the National Natural Science Foundation of China myces cerevisiae homolog of AMP‑activated protein kinase. Cell Metab. (31170093). 2011;14:707–14. Palomino A, Herrero P, Moreno F. Rgt1, a glucose sensing transcription factor, is Received: 21 July 2016 Accepted: 21 July 2016 required for transcriptional repression of the HXK2 gene in Saccharomyces cerevisiae. Biochem J. 2005;388:697–703. Zeng et al. AMB Expr (2016) 6:51 Page 13 of 13 Peng B, Shen Y, Li X, Chen X, Hou J, Bao X. Improvement of xylose fermenta‑ Salusjärvi L, Kankainen M, Soliymani R, Pitkänen JP, Penttilä M, Ruohonen L. tion in respiratory‑ deficient xylose ‑fermenting Saccharomyces cerevisiae. Regulation of xylose metabolism in recombinant Saccharomyces cerevi- Metab Eng. 2012;14:9–18. siae. Microb Cell Factories. 2008;7:18. Regenberg B, Grotkjaer T, Winther O, Fausboll A, Akesson M, Bro C, Hansen LK, Tang Y, An M, Liu K, Nagai S, Shigematsu T, Morimura S, Kida K. Ethanol produc‑ Brunak S, Nielsen J. Growth‑rate regulated genes have profound impact tion from acid hydrolysate of wood biomass using the flocculating yeast on interpretation of transcriptome profiling in Saccharomyces cerevisiae. Saccharomyces cerevisiae strain KF‑7. Process Biochem. 2006;41:909–14. Genome Biol. 2006;7:R107. Thomas D, SurdinKerjan Y. Metabolism of sulfur amino acids in Saccharomyces Runquist D, Hahn‑Hägerdal B, Bettiga M. Increased expression of the oxidative cerevisiae. Microbiol Mol Biol Rev. 1997;61:503. pentose phosphate pathway and gluconeogenesis in anaerobically Wasylenko TM, Stephanopoulos G. Metabolomic and C‑metabolic flux growing xylose‑utilizing Saccharomyces cerevisiae. Microb Cell Factories. analysis of a xylose‑ consuming Saccharomyces cerevisiae strain expressing 2009;8:49. xylose isomerase. Biotechnol Bioeng. 2015;112:470–83. Salusjärvi L, Pitkänen JP, Aristidou A, Ruohonen L, Penttilä M. Transcription Zaman S, Lippman SI, Zhao X, Broach JR. How Saccharomyces responds to analysis of recombinant Saccharomyces cerevisiae reveals novel responses nutrients. Annu Rev Genet. 2008;42:27–81. to xylose. Appl Biochem Biotechnol. 2006;128:237–73. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png AMB Express Springer Journals

Transcriptomes of a xylose-utilizing industrial flocculating Saccharomyces cerevisiae strain cultured in media containing different sugar sources

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/transcriptomes-of-a-xylose-utilizing-industrial-flocculating-nZGtvLMZ20

References (59)

Publisher
Springer Journals
Copyright
Copyright © 2016 by The Author(s)
Subject
Life Sciences; Microbiology; Microbial Genetics and Genomics; Biotechnology
eISSN
2191-0855
DOI
10.1186/s13568-016-0223-y
pmid
27485516
Publisher site
See Article on Publisher Site

Abstract

Lignocellulosic hydrolysates used for bioethanol production contain a mixture of sugars, with xylose being the second most abundant after glucose. Since xylose is not a natural substrate for Saccharomyces cerevisiae, recombi‑ nant S. cerevisiae strongly prefers glucose over xylose, and the fermentation rate and ethanol yield with xylose are both lower than those with glucose. To determine the molecular basis for glucose and xylose fermentation, we used microarrays to investigate the transcriptional difference of a xylose ‑ utilizing industrial strain cultured in both single sugar media and a mixed sugar medium of glucose and xylose. The transcriptomes were nearly identical between glucose metabolizing cells in the glucose alone medium and those in the glucose fermentation phase in the mixed‑ sugar medium. Whereas the transcriptomes highly differed between the xylose metabolizing cells in the xylose alone medium and those in the xylose fermentation phase in the mixed sugar medium, and the differences mainly involved sulfur metabolism. When the transcriptional profiles were compared between glucose fermentation state and xylose fermentation state, we found the expression patterns of hexose transporters and glucose signaling pathway dif‑ fered in response to different sugar sources, and the expression levels of the genes involved in gluconeogenesis, the glyoxylate and tricarboxylic acid cycles and respiration increased with xylose, indicating that the xylose‑ metabolizing cells had high requirements for maintenance energy and lacked the carbon catabolite repression capability. The effect of carbon catabolite repression by glucose lasted after glucose depletion for specific genes to different extents. Keywords: Transcriptome, Saccharomyces cerevisiae, Xylose fermentation, Bioethanol, Glucose and xylose cofermentation In the past two decades, fermentation of xylose to eth- Introduction anol has been achieved in S. cerevisiae by genetic engi- Lignocellulosic biomass has been recognized as a sustain- neering. Through expression of the heterogeneous xylose able source for fuel ethanol production without affecting metabolic pathway—either xylose reductase-xylitol dehy- the food and feed markets. It is of economic interest to drogenase (XR-XDH) or xylose isomerase (XI)—S. cer- convert all lignocellulosic sugar fractions, predominantly evisiae can convert xylose to xylulose, which can then glucose and xylose, into ethanol at sufficiently high rates be natively catabolized (Matsushika et  al. 2009a). The and yields. However, Saccharomyces cerevisiae, which is xylose-utilizing capacity of the recombinant strains can widely used in bioethanol plants due to its high fermen- be further optimized by enhancing the downstream met- tation efficiency and process robustness, cannot ferment abolic pathway rationally or through evolutionary engi- xylose (Batt et al. 1986). neering (Peng et al. 2012). However, recombinant strains strongly prefer glucose over xylose, and therefore the *Correspondence: tangyq@scu.edu.cn co-consumption remains a challenge. What’s more, the College of Architecture and Environment, Sichuan University, No. 24, specific ethanol productivity from xylose was an order of South Section 1, First Ring Road, Chengdu 610065, Sichuan, China © 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Zeng et al. AMB Expr (2016) 6:51 Page 2 of 13 magnitude lower than that from glucose, despite tremen- attenuate the possible impact of different growth rates on dous efforts, and the ethanol yield from xylose was lower transcriptome analysis. This is the first study investigat - than that from glucose as well (Matsushika et al. 2009b). ing transcriptional difference of an industrial S. cerevi - To reveal the major reasons for the suboptimal fermen- siae strain in both single sugar media and mixed sugar tation of xylose by recombinant S. cerevisiae strains, the medium. difference in transcriptional response between xylose and glucose fermentation has been examined in the past Materials and methods decade. It has been recognized that S. cerevisiae does not Strains and batch fermentation conditions sense xylose as a fermentable carbon source (Jin et  al. The recombinant xylose-utilizing S. cerevisiae used in this 2004). Early transcriptional analysis on xylose was con- study was KF7M-16 derived from KF-7 (Kida et al. 1992). ducted during aerobic growth, and revealed that xylose Two integrative plasmids were transformed into KF-7: was neither recognized as a fermentable carbon source pIUX1X2XK (contains XYL1, XYL2 from Scheffersomyces nor as a respirative carbon source (Salusjärvi et al. 2006). stipitis and XKS1 from S. cerevisiae) and pIWBGL1 (con- Using transcriptome and proteome, the difference in car - tains BGL1 from Aspergillus aculeatus) (Li et al. 2015). bon source signaling and catabolite repression was stud- The single and mixed sugar media used in the batch ied in the aerobic batch fermentation of either glucose or fermentation were as follows: 6  % YPD (20  g/L peptone, xylose, and it has been suggested that cells metabolizing 10 g/L yeast extract, and 60 g/L glucose), 4 % YPX (20 g/L xylose were neither in a completely repressed nor in a peptone, 10  g/L yeast extract, and 40  g/L xylose), and derepressed state (Salusjärvi et  al. 2008). The transcrip - 10 % YPDX (20 g/L peptone, 10 g/L yeast extract, 60 g/L tional difference in S. cerevisiae growing anaerobically glucose, and 40 g/L xylose). in either glucose or xylose was subsequently analyzed Batch fermentations were conducted in 300-ml cotton- (Matsushika et  al. 2014; Runquist et  al. 2009), indicat- plugged shake-flasks containing 100 ml medium, at a stir - ing xylose was recognized as a non-fermentable carbon ring speed of 200 rpm controlled by a HS-6DN magnetic source and induced the expression of stress-responsive stirrer (As One, Japan). The temperature was maintained genes. More recently, the specific regulatory response of at 35 °C in a thermostatic water bath. S. cerevisiae to xylose was quantified at a range of culti - Working cultures of yeasts were obtained following an vation times in anaerobic glucose-xylose mixed medium activation period at 30  °C for 24  h on a 2  % YPD plate (Alff-Tuomala et  al. 2016), and xylose was observed to (20 g/L peptone, 10 g/L yeast extract, 20 g/L glucose with delay the glucose-dependent repression of specific genes 2 % agar) and then pre-cultivating the yeast cells at 30 °C in the mixed culture. for 16 h in 5 % YPD (20 g/L peptone, 10 g/L yeast extract, To the best of our knowledge, the transcription and and 50  g/L glucose). The harvested cells were inocu - regulatory responses induced by xylose have been ana- lated into batch fermentation at 5  g-dry cell weight/L lyzed either in single sugar or mixed sugar cultures. As (g-DCW/L). the transcriptional profile is dependent on different host For the pre-experiment, the culture was sampled every strain background (Feng and Zhao 2013a), a systematic 2  h in the fast fermenting stage to monitor cell growth transcriptional analysis of a particular S. cerevisiae strain and metabolites concentration. The experiments for under different fermentation conditions, including in microarray analysis were conducted in biological dupli- media with a single sugar and mixed sugars, would be cate, and samples were collected at the time points indi- useful to identify the genetic factors responsible for the cated in the results section. discrepancy in sugar source utilization efficiency. Indus - trial S. cerevisiae strains generally have a superior ethanol RNA extraction production efficiency and inhibitor tolerance compared After cell collection, the total RNA was extracted using to laboratory strains. In this study, we therefore carried the Takara Yeast RNAiso Kit according to the manufac- out batch fermentations of KF7M-16, an XR- and XDH- turer’s protocol. RNA quality and concentration were expressing industrial flocculating strain, in both sin - measured by agarose gel electrophoresis and NanoDrop gle sugar medium (either glucose or xylose) and mixed 2000/2000C (Thermo Scientific, USA). sugar medium (including the glucose fermentation phase and the xylose fermentation phase), and analyzed the Microarray analysis global transcriptomes based on microarrays to identify Microarray analysis was performed using the 7G Affy - the molecular response to different fermentation states. metrix GeneChip Yeast Genome 2.0 Array (CapitalBio It has been reported that cellular growth rate has great Tec., Beijing). The isolated total RNA was cleaned up influence on transcriptional regulation (Regenberg et  al. with RNeasy Kit (Qiagen, Germany). 100 ng of total RNA 2006), we therefore adopt high-density inoculations to was used for cDNA synthesis and produce biotin-tagged Zeng et al. AMB Expr (2016) 6:51 Page 3 of 13 cRNA with GeneChip IVT Labeling kit (Affymetrix). 15  µg fragmented cRNA, together with contol oligo B2 and eukaryotic hybridization controls, was hybrid- ized to each GeneChip array at 45  °C for 16  h (Affym - etrix GeneChipHybridization Oven 640) according to manufacturer’s instructions. After hybridization, the GeneChip arrays were washed, stained with streptavidin phycoerythrinonan (SAPE) with Affymetrix Fluidics Sta - tion 450, and then scanned with the Affymetrix Gene - Chip Scanner 3000 7G. The data extraction and analysis were carried out using Affymetrix GeneChip Command Console Software. The microarray data can be accessed through GEO accession through GSE80748. To identify altered gene expression, the averages of biological dupli- cates were compared, and fold changes in gene expres- sion ≥2 were considered significant. The gene annotation information was based on the Saccharomyces genome database. The pathway terms were enriched using the KEGG orthology based annotation system (KOBAS). Quantitative real‑time PCR The cDNA was reverse-transcribed from total RNA using the Takara PrimeScript RT reagent Kit with gDNA Eraser (Perfect Real Time). qPCR analysis was performed ® ™ using the Takara SYBR Premix Ex Taq II (Tli RNaseH Plus) following manufacture’s manual. ACT1 was used as the normalization standard. The primers used are listed in Additional file 1: Table S1. Fig. 1 Time‑ dependent fermentation profile of KF7M ‑16 in a YPD medium containing 60 g/L glucose (6 % YPD), b YPX medium Analytical methods for substrate and products containing 40 g/L xylose (4 % YPX), and c YPDX medium containing Metabolite analysis was conducted as previously 60 g/L glucose and 40 g/L xylose (10 %YPDX). Black diamond OD ; described (Tang et  al. 2006). Glucose and xylose in black square glucose; empty square xylose; empty up‑pointing triangle ethanol; and error marks xylitol. The arrows (denoted by G, X, GX1, and the fermentation medium was determined by a HPLC GX2) indicate the times at which samples were taken for transcription equipped with a fluorescence detector (RF-10A ). Etha- XL analysis nol was measured by GC with a FID detector and 2-pro- panol was used as the internal standard. Xylitol was assayed by HPLC equipped with an Aminex HPX-87H column (300 × 7.8 mm) (Bio-Rad, USA) and a refractive YPDX. The utilization rate of xylose was approximately a index detector. magnitude slower than that of glucose (2.88 g/L/h in 4 % YPX in the first 8  h, 17.27  g/L/h in 6  % YPD in the first Results 2  h and 16.80  g/L/h in 10  % YPDX in the first 2  h), and Experimental design and batch fermentation after 24  h of metabolism, KF7  M-16 almost consumed Batch fermentations were carried out in YP medium con- all of the xylose in both the 4  % YPX and 10  % YPDX. taining 60  g/L glucose (6  % YPD) or 40  g/L xylose (4  % It is worth noting that the presence of glucose repressed YPX) as single sugar medium, as well as in YP medium the utilization of xylose, as in 10  % YPDX the xylose containing 60  g/L glucose and 40  g/L xylose as mixed consumption rate was 1.38  g/L/h in the first 4  h and sugar medium (10  % YPDX). To eliminate the effect of 3.03 g/L/h in 4–8 h. Therefore the fermentation profile in different growth rates between glucose and xylose uti - 10 % YPDX can be divided into the glucose fermentation lization on the transcriptional profile, a relatively high phase and the xylose fermentation phase. inoculum density of 5  g-DCW/L was chosen. A pre- For transcription analysis, experiments were con- experiment was conducted to grasp the fermentation ducted in duplicate. Samples of glucose-grown cells were profile of KF7M-16 (Fig.  1). Glucose was depleted in harvested at 2  h while samples of xylose-grown cells the first 4  h of fermentation, both in 6  % YPD and 10  % were harvested at 7  h. The samples taken at 2  h in 6  % Zeng et al. AMB Expr (2016) 6:51 Page 4 of 13 YPD were denoted as G, and the residual glucose was 16.4  ±  1.2  g/L, with 16.4  ±  0.3  g/L ethanol produced at a yield of 0.46  ±  0.01  g/g at that time point. The sam - ples taken at 7  h in 4  % YPX were denoted as X, and at that time point the residual xylose was 20.4  ±  0.1  g/L, with 6.2  g/L ethanol and 1.7  ±  0.1  g/L xylitol produced at yields of 0.35  ±  0.1 and 0.10  ±  0.003  g/g respec- tively. The samples taken at 2  h in 10  % YPDX (during glucose fermentation stage) were denoted as GX1, and during this first 2  h of fermentation, 37.1  ±  5.1  g/L glu- Fig. 2 Overview of differently expressed gene numbers between cose and 2.3  ±  1.3  g/L xylose were consumed, produc- different fermentation conditions. G glucose fermentation state in ing 17.2  ±  1.8  g/L ethanol at a yield of 0.44  ±  0.02  g/g. the glucose alone medium; X xylose fermentation state in the xylose The samples taken at 7  h in 10  % YPDX (during xylose alone medium; GX1 glucose fermentation phase in the mixed sugar; fermentation stage) were denoted as GX2, and the glu- GX2 xylose fermentation phase in the mixed sugar. C1–C4 indicate the specific pairwise comparisons. Averages of biological duplicates cose had been depleted for at least 3  h by then, and the were compared. The numbers indicate differently expressed genes in residual xylose was 19.5  ±  1.0  g/L, with 27.8  ±  1.8  g/L each comparison ethanol as well as 4.6 ± 0.3 g/L xylitol produced at yields of 0.38 and 0.26 ± 0.04 g/g respectively. It’s notable that the ethanol yield from xylose was 24  % lower than that from glucose, and that the xylitol yield from xylose was X vs. G 160 % higher in the mixed sugar medium than in the sin- To reveal the molecular basis for the fermentation abil- gle sugar medium. ity discrepancy between glucose and xylose, the tran- scriptional difference in S. cerevisiae growing in either Overview of transcriptional differences across four the glucose alone or the xylose alone medium has long conditions been analyzed, both aerobically and anaerobically. In To obtain the general information of transcriptional our study with the low aeration condition, a total of 487 responses to the different sugar sources, the microarray genes showed significantly different expression levels data was organized into four relevant pairwise compar- (fold change ≥2) in C2, with 281 upregulated and 206 isons: GX1 vs. G (comparison 1  =  C1); X vs. G (com- downregulated genes (Additional file  1: Table S3). The parison 2  =  C2); GX2 vs. GX1 (comparison 3  =  C3); significantly enriched pathway terms (p value  <  0.1) are and GX2 vs. X (comparison 4  =  C4). The averages of listed in Table 1. biological duplicates were compared. The numbers of Secondary metabolites are organic compounds that are genes with significantly changed expression levels (fold not directly involved in the normal growth or reproduc- change ≥2) between the corresponding conditions are tion of the cells, and often have a role in stress defense. shown in Fig.  2. Differentially expressed genes can be A secondary metabolite forms frequently near or in the identified and enriched for pathway terms using the stationary phase of growth. The KEGG term of “Biosyn - KEGG Orthology Based Annotation System (KOBAS). thesis of secondary metabolites” (ID: sce01110) involves a number of pathways, including carbon metabolism and GX1 vs. G biosynthesis of amino acids. Only 36 genes stood out as being differently expressed in The expressional changes in carbon metabolism C1 (Additional file  1: Table S2), with 16 upregulated and directly reflect the different metabolic states in differ - 20 downregulated genes. This suggested that the transcrip - ent sugar sources. Genes involved in the citrate and gly- tional profile of cells during glucose fermentation phase oxylate cycles, which are repressible by glucose (Gancedo in the mixed sugar medium was nearly identical to those 1998), were upregulated in the xylose alone medium, growing in glucose alone medium. Since xylose is not a indicating that xylose lacked the carbon catabolite natural carbon source for S. cerevisiae, recombinant strains repression capability. Genes involved in the central car- have a stronger preference for glucose rather than xylose. bon metabolism across the four conditions will be further KF7M-16 utilized glucose first in the mixed sugar medium. discussed later. Although a meager amount of xylose was consumed in the The expression levels of the genes involved in sulfur first 2 h of fermentation, the added xylose seemed power - metabolism were downregulated in C2. The detailed less to affect the glucose dominated transcriptional profile. transcriptional changes will be further discussed in C4. Zeng et al. AMB Expr (2016) 6:51 Page 5 of 13 Table 1 Enriched pathways of the differently expressed genes Comparison Term Database ID p value C2 Biosynthesis of secondary metabolites KEGG PATHWAY sce01110 2.21E−05 Carbon metabolism KEGG PATHWAY sce01200 2.46E−04 Histidine biosynthesis PANTHER P02747 5.18E−04 Glyoxylate and dicarboxylate metabolism KEGG PATHWAY sce00630 6.25E−04 Biosynthesis of amino acids KEGG PATHWAY sce01230 1.09E−03 Metabolic pathways KEGG PATHWAY sce01100 2.96E−03 Sulfur metabolism KEGG PATHWAY sce00920 3.70E−03 Citrate cycle ( TCA cycle) KEGG PATHWAY sce00020 5.86E−03 C3 Biosynthesis of secondary metabolites KEGG PATHWAY sce01110 4.19E−03 Heme biosynthesis PANTHER P02746 9.02E−03 Histidine metabolism KEGG PATHWAY sce00340 1.20E−02 Tryptophan metabolism KEGG PATHWAY sce00380 1.42E−02 Porphyrin and chlorophyll metabolism KEGG PATHWAY sce00860 1.42E−02 C4 Sulfur metabolism KEGG PATHWAY sce00920 7.11E−04 superpathway of sulfur amino acid biosynthesis BioCyc PWY‑821 1.44E−03 Arginine biosynthesis PANTHER P02728 1.58E−03 Sulfate reduction I (assimilatory) BioCyc SO4ASSIM‑PWY 9.14E−03 http://bioinfo.capitalbio.com/mas3/ GX2 vs. GX1 uroporphyrinogen (MET1, MET8) showed an increased The transcription profiles in the mixed sugar medium expression level. were compared between the xylose fermentation stage It has been previously suggested that the increased and the glucose fermentation stage. A total of 499 and expression level of the genes related to tryptophan bio- 377 genes were upregulated and downregulated respec- synthesis might confer ethanol stress tolerance to yeast tively in C3 (Additional file  1: Table S4). The number of cells (Hirasawa et  al. 2007). In our study, the expression differently expressed genes in C3 was 80 % more than in level of the genes related to tryptophan degradation C2, although both comparisons were carried out between (BNA1, BNA4, BNA7, ARO8) decreased when the tran- glucose fermentation state and xylose fermentation scriptional profile of xylose fermentation stage in the state. The transcriptional shift induced by different sugar mixed sugar medium was compared to that of the glu- source utilization was more complex in the mixed sugar cose fermentation stage, corroborating the involvement medium, mainly due to the prolonged effect of glucose of tryptophan in response to ethanol stress. repression and the stress induced by the accumulated Genes involved in histidine biosynthesis (HIS1, HIS7, ethanol. The enriched pathway terms (p < 0.02) are listed HIS2), which were previously reported to be significantly in Table 1. downregulated in the presence of ethanol (Li et al. 2010), Similarly to C2, the KEGG term “biosynthesis of sec- also had a decreased expression level in C3. ondary metabolites” was enriched with the lowest p value in C3. GX2 vs. X Heme is a cofactor consisting of a ferrous ion con- Since the metabolic activity during the xylose fermenta- tained in the center of a large heterocyclic organic tion phase in mixed sugar medium was similar to that ring called a porphyrin. Heme is the component of in xylose alone medium, it is interesting to observe that cytochrome which is primarily responsible for the gen- when the transcription profile of GX2 was compared to eration of ATP via electron transport. Although the that of X, 682 genes were differently expressed, with 375 expression level of genes encoding respirative enzymes upregulated and 307 downregulated genes (Table S5, in increased during the xylose fermentation stage relative the supplementary material). Significantly enriched path - to the glucose fermentation stage in the mixed sugar way terms (p < 0.01) are listed in Table 1. medium (Table  4), several genes involved in the heme Sulfur amino acid biosynthesis can be divided into biosynthesis from uroporphyrinogen (HEM12, HEM13, three parts: sulfate assimilation, cysteine and methio- HEM15) showed a decreased expression level. Whereas nine biosynthesis, and S-adenosyl-l -methionine (SAM) the genes involved in the siroheme biosynthesis from biosynthesis (Thomas and SurdinKerjan 1997). SAM is Zeng et al. AMB Expr (2016) 6:51 Page 6 of 13 probably second only to ATP in the variety of reactions additional broad substrate sugar transporters (Hamacher for which it serves as a cofactor, and is also a precursor et al. 2002). The genes encoding transporters showed dif - of glutathione, which is a major cellular anti-oxidant ferent transcriptional patterns on different sugar sources (Caro and Cederbaum 2004). Compared to their expres- (Fig.  4, Table  2). The transcriptional levels of the high- sion levels in the glucose alone medium as a control, affinity glucose transporter genes HXT2 and HXT4, as the expression of genes involved in sulfur amino bio- well as the low-affinity glucose transporter gene HXT3 synthesis was higher in GX2 and was lower in X. The were downregulated with xylose, in both the xylose alone comparative expression ratios are presented in Fig.  3. medium and during the xylose fermentation phase in It is reasonable to speculate that the demand for SAM the mixed sugar medium. On the other hand, the non- might be higher during the xylose fermentation stage in fermentable carbon source-inducible transporter genes the mixed sugar medium than in xylose alone medium. HXT5 and HXT13 were upregulated with xylose, indi- Considering that SAM is protective against a variety of cating that xylose was not recognized by S. cerevisiae toxic oxidative agents, this difference might be due to as being fermentable. HXT15 and HXT16, repressed by the metabolites accumulated in the process of glucose high levels of glucose, were induced nearly 40 times in fermentation. the xylose alone medium, but not during the xylose fer- The expression of genes involved in arginine biosynthe - mentation phase in the mixed sugar medium, suggesting sis from glutamine (CPA2, ARG3, ARG1, and ARG4) also that the effect of glucose repression could last after glu - increased in C4. cose had been depleted. HXT6//7 were expressed at high levels under all four conditions. Other genes in the HXTs Carbon source sensing and signaling family did not show obvious differences in expression Uptake of xylose by S. cerevisiae has been proposed to be level, and all of them were negligibly expressed, with the mediated unspecifically by its hexose-transport system, exception of HXT9. As deduced from mRNA level of dif- which is composed of 18 genes from the HXTs family and ferent transporters (Table  2), HXT3 and HXT4, together Fig. 3 The expression ratio of the genes involved in sulfur amino acid biosynthesis. The fold change of expression is presented for GX2 (left panel) and X (right panel), both compared to the transcriptional state in the glucose alone medium (G) as a control. Averages of biological duplicates were compared. The red panel indicates that the transcription level of gene increased over twofold; the green panel indicates the transcription level of gene decreased over twofold Zeng et al. AMB Expr (2016) 6:51 Page 7 of 13 Fig. 4 Expression profiles of the hexose transporter genes and comparison of the transcriptional changes in the genes involved in glucose sens‑ ing and repression network. The relative expression of hexose transporter genes is indicated with different colors. For the regulator genes, the fold change of expression is presented for GX2 (left panel) and X (right panel), both compared to G as a control. Averages of biological duplicates were compared. The red panel indicates the transcription level of gene increased over twofold; the green panel indicates the transcription level of gene decreased over twofold Table 2 The hybridization signal (MAS5.0 signal intensity) of  the genes encoding hexose-transport system in  different fermentation states Gene G GX1 GX2 X HXT1 25.9 ± 2.9 29.7 ± 11.5 38.8 ± 15.8 24.3 ± 1.9 HXT2 700.8 ± 76.9 500.1 ± 236.3 125.4 ± 13.0 114.8 ± 47.5 HXT3 3937.3 ± 299.5 3039.1 ± 828.4 643.8 ± 235.8 555.2 ± 106.5 HXT4 3542.4 ± 626.0 3868.6 ± 17.7 1389.3 ± 118.5 1136.8 ± 207.0 HXT5 34.6 ± 1.2 48.9 ± 30.0 220.1 ± 80.3 142.9 ± 59.3 HXT6//7 1756.3 ± 302.6 1801.2 ± 181.3 2031.2 ± 157.8 1949.0 ± 7.5 HXT8 24.9 ± 0.2 23.2 ± 0.7 22.3 ± 0.9 22.6 ± 0.6 HXT9 263.5 ± 56.7 256.0 ± 50.2 242.6 ± 9.4 190.3 ± 8.7 HXT10 7.9 ± 0.2 7.6 ± 0.2 7.0 ± 0.1 8.0 ± 0.2 HXT11 8.1 ± 0.2 5.8 ± 0.8 7.9 ± 1.5 8.4 ± 1.1 HXT13 17.0 ± 6.9 13.6 ± 0.6 39.8 ± 1.6 131.3 ± 19.3 HXT14 7.9 ± 0.1 7.9 ± 0.3 11.7 ± 0.8 7.7 ± 0.3 HXT15//16 12.0 ± 1.1 13.2 ± 0.9 14.5 ± 0.9 536.4 ± 110.5 GAL2 38.6 ± 0.3 50.1 ± 6.9 79.2 ± 5.1 58.3 ± 2.7 Values are given as the average and standard deviation of two biological duplicates “//” indicates genes detected by the same probe set because of their sequence similarity with HXT6 and HXT7 might serve as the main transport- It has been previously reported that genes encod- ers in the glucose fermentation state, while HXT6, HXT7 ing the glucose sensor proteins Snf3p and Rgt2p, which as well as HXT4 might be the main transporters in the mediate signal for the presence of glucose at low or xylose fermentation state. high concentration respectively, had higher expression Zeng et al. AMB Expr (2016) 6:51 Page 8 of 13 levels with xylose (Salusjärvi et  al. 2008). In our study, both have been previously reported to be poor aldo–keto SNF3 and RGT2 had stable expression levels across all reductases (Chang et  al. 2007). The expression levels of four conditions. Glucose binds to Snf3 and Rgt2, induc- the endogenous genes SOR1 and SOR2 increased during ing them to bind to Mth1 and Std1, and finally leading xylose fermentation, and to a much stronger extent in the to the degradation of Mth1 and Std1, which are two Rgt- xylose alone medium than in the mixed medium, which 1corepressors (Zaman et  al. 2008). The expression levels might have accounted for the lower observed xylitol yield of the corepressor genes MTH1 and STD1 decreased in the xylose alone medium than in the mixed medium. with xylose, and that of RGT1 slightly increased (Fig.  4). For SOR1//2, the microarray data was validated by RT- Rgt1 is a DNA-binding protein that represses the hexose qPCR (Table  3). Even though the SOR1//2 expression transporter genes HXT2-4, as well as hexokinase gene level increased in magnitudes in X as compared to glu- HXK2. The expression levels of these genes were corre - cose fermentation state (G and GX1), the transcript spondingly downregulated. The xylose-dependent upreg - abundance of SOR1//2 was still lower than XYL2 under ulation of RGT1, irrespective of oxygen availability, has xylose fermentation state. been previously observed (Alff-Tuomala et al. 2016). The The major enzymes in the non-oxidative pentose phos - expression levels of the transcription factor Adr1, which phate pathway, TAL1 and TKL1 encoding transaldo- is negatively regulated by PKA in glucose growing cells lase and transketolase (Matsushika et  al. 2012), did not (Zaman et  al. 2008), were upregulated with xylose. Glu- express significantly differently across all conditions, cose metabolism activates the transcriptional repressor but the expression of their minor isoenzyme NQM1 Mig1 to move into the nucleus and repress its many tar- and TKL2 was induced during xylose fermentation. The gets (Kuttykrishnan et  al. 2010). The expression level of expression level of NQM1 is usually induced during Mig1 decreased with xylose, while that of Cat8, which is diauxic shift and TKL2 is regulated in a Msn2/4p manner. repressed by Mig1, increased with xylose. Transcription Another non-oxidative PPP gene, RKI1, was expressed at factor Cat8 activates gene expression required for gluco- a lower level on xylose. In the oxidative pentose phos- neogenesis during growth in the absence of glucose, the phate pathway, the expression levels of ZWF1 encoding transcriptional changes of which will be discussed in the glucose-6-phosphate dehydrogenase and GND2 encod- following section. ing 6-phosphogluconate dehydrogenase increased on xylose. Central carbon metabolism In three structural genes encoding enzymes that cata- Xylose was channeled to ethanol via the heterologous lyze the phosphorylation of glucose to glucose 6-phos- xylose assimilating pathway, pentose phosphate pathway, phate, HXK2 was repressed with xylose, especially glycolysis pathway and alcohol fermentation pathway. To during the xylose fermentation phase in the mixed sugar pinpoint the effect of xylose on the fermentation state, medium. Hxk2p is known to provide the main sugar- the genes expression in the central carbon metabolism phosphorylating capability during growth on fermentable was investigated in greater detail, using the transcrip- carbon source, and becomes repressed on non-fermenta- tional state when cultivated in glucose alone medium as ble carbon source by transcription factor Rgt1 (Palomino a control (Fig. 5). et al. 2005). Since the heterologous XYL1 and XYL2 genes from S. Most glycolysis genes showed stable expression levels stipitis were not included in the S. cerevisiae genechips, across all four conditions, except for two genes encod- their expression levels were determined by RT-qPCR ing homolog of Gpm1p phosphoglycerate mutase, GPM2 (Table 3). The heterologous XYL1 and XYL2 did not show and GPM3, were slightly downregulated on xylose, and significant difference across all the tested conditions, but these two genes may be non-functional. the S. cerevisiae genes encoding enzymes with xylose In the alcohol fermentation and by-product path- reductase and xylitol dehydrogenase function manifested way, two genes encoding the minor isoform of pyruvate different expression patterns (Fig.  5). The expression decarboxylase, PDC5 and PDC6, respectively, showed level of GCY1 increased as long as the medium con- decreased and increased expression levels with xylose. tained xylose. Another transcription study using mixed PDC5 is repressed by thiamine, and PDC6 is induced sugar medium also indicated expression level of GCY1 during sulfur limitation. Yet several genes involved in increased even in glucose fermentation phase (Alff- thiamine biosynthesis (THI genes) showed decreased Tuomala et  al. 2016). During the xylose fermentation expression levels with xylose. Most ALD genes involved phase in the mixed sugar medium, the expression level in acetate formation showed increased expression of YJR096W increased and that of YDL241W decreased, level during the xylose fermentation stage in the mixed while both genes did not express differently in xylose sugar medium, including cytosolic aldehyde dehydroge- alone medium, compared to glucose alone medium, and nase ALD2, ALD3, ALD6, and mitochondrial aldehyde Zeng et al. AMB Expr (2016) 6:51 Page 9 of 13 Fig. 5 Expression profiles of genes involved in the central carbon metabolism. The fold change of expression is presented for GX1 (left panel), GX2 (middle panel) and X (right panel), compared to G as a control. Averages of biological duplicates were compared. The red panel indicates the tran‑ scription level of gene increased over twofold; the green panel indicates the transcription level of gene decreased over twofold dehydrogenase ALD4. This upregulation was likely to be is sensed as a non-fermentable carbon source by S. a response to the higher ethanol level at the sampling cerevisiae. time of GX2. The expression level of ACS1 , encoding The expression levels of several genes in the tricarbox - a synthetase to form acetyl coenzyme A from acetate, ylic acid cycles were upregulated with xylose. Transcripts increased with xylose, once again indicating that xylose of CIT2 and CIT3, encoding enzymes that catalyze the Zeng et al. AMB Expr (2016) 6:51 Page 10 of 13 Table 3 Expression level of  xylose metabolizing genes genes in the glyoxylate shunt, ICL1 and MLS1 encoding by RT-qPCR isocitrate lyase and malate synthase respectively, together with CIT2 encoding peroxisomal citrate synthase were Sugar XYL1/ACT1 XYL2/ACT1 SOR1//2/ACT1 also induced on xylose. G 1.9 ± 0.25 2.53 ± 0.52 7.35E−5 ± 1.2E−6 In the glycerol catabolism pathway, the expression X 1.9 ± 0.42 2.36 ± 0.58 1.13E−1 ± 2.1E−2 level of the glycerol-producing gene RHR2 encoding GX1 2.0 ± 0.22 2.17 ± 0.34 6.84E−5 ± 3.3E−6 glycerol-1-phosphatase increased with xylose, which GX2 2.19 ± 0.45 1.28 ± 0.28 1.08E−3 ± 5.7E−5 is contrary to an earlier finding with anaerobic batch Values are given as the average and standard deviation of two duplicates fermentation (Matsushika et  al. 2014). The expression levels of glycerol-consuming genes, GUT2 encoding mitochondrial glycerol-3-phosphate dehydrogenase condensation of acetyl coenzyme A and oxaloacetate and DAK2 encoding dihydroxyacetone kinase, also to form citrate, increased with xylose, especially in the increased with xylose. DAK2 was strongly induced dur- xylose alone medium. The levels of ACO2 , encoding a ing the xylose fermentation stage in the mixed sugar putative mitochondrial aconitase and repressed by eth- medium, likely due to its involvement in stress adap- anol, decreased during the xylose fermentation stage tation. Similar to DAK2, STL1 encoding the glycerol in the mixed sugar medium. As for the IDP genes that proton symporter of the plasma membrane was also encode isocitrate dehydrogenase, the expression level strongly induced during the xylose fermentation stage of IDP1 encoding mitochondrial isocitrate dehydroge- in the mixed sugar medium, but not in the xylose alone nase did not change, but that of IDP2 and IDP3, encod- medium. STL1 was the gene with the largest upregula- ing NADP specific cytosolic and peroxisomal isocitrate tion ratio (70.90) in C4, reflecting the different osmotic dehydrogenase respectively, increased with xylose. The state between GX2 and X, because STL1 is documented KGD genes (KGD1 and KGD2) encoding components of to be strongly but transiently induced when cells are the mitochondrial alpha-ketoglutarate dehydrogenase subjected to osmotic shock. complex, as well as the SDH genes (SDH1, SDH2, SDH3, and SDH4) that encode subunits of succinate dehydro- genase, had slightly increased expression levels with Table 4 The expressional fold changes of  the genes xylose. The expression level of MDH2, encoding cyto - responsible for  respiratory metabolism and  ATP synthesis plasmic malate dehydrogenase, which is also involved for GX1, GX2, and X, compared to G as a control (averages in the gluconeogenesis and glyoxylate cycles, increased of biological duplicates were compared) with xylose. Because 2 carbons lost as CO in one TCA Genes GX1 GX2 X Description cycle, the strong activity of the TCA cycle could be the reason for the lowered ethanol yield from xylose. Cor- HAP4 1.11 1.55 2.13 Transcriptional activator of respiratory genes roborating with the upregulated TCA cycle pathway, NDE1 0.82 5.54 5.65 Mitochondrial external NADH dehydrogenase the expression levels of a number of genes encoding NDE2 1.29 2.92 3.39 respiratory enzymes increased with xylose (Table  4). NDI1 1.21 2.55 2.27 Ubiquinone oxidoreductase The expression level of HAP4, a transcriptional activa- QCR2 1.14 1.99 2.12 Subunits of ubiquinol cytochrome c reduc‑ tor and global regulator of respiratory gene expression, tase complex QCR9 1.18 2.38 1.67 also slightly increased with xylose. As a result, a sig- QCR8 1.10 2.07 1.49 nificant amount of ATP was assumed to be produced QCR10 1.24 2.06 1.87 in xylose metabolism, indicating high requirements for CYC3 0.88 1.80 2.01 Cytochrome c heme lyase maintenance energy to utilize xylose by recombinant S. CYC7 1.18 3.49 1.63 Cytochrome c isoform 2 cerevisiae. COX5A 0.99 2.56 1.96 Subunits of cytochrome c oxidase The gluconeogenic gene FBP1 , encoding fructose- COX7 1.20 2.78 2.30 1,6-bisphosphatase, was strongly induced with xylose. COX12 1.00 1.92 2.09 The transcription levels of other genes involved in glu - COX13 0.96 2.81 1.96 coneogenesis, such as PCK1 encoding phosphoenol- ATP2 1.05 1.15 1.91 Subunits of ATP synthase pyruvate carboxykinase and MDH2, also increased with ATP3 1.06 1.38 1.63 xylose. The upregulation ratio was higher in the xylose ATP4 1.15 1.39 1.78 alone medium than during the xylose fermentation phase ATP5 1.05 0.93 1.62 in the mixed sugar medium. The gluconeogenesis path - ATP14 1.06 1.27 1.62 way is repressed by glucose, and our results indicated ATP16 1.10 1.34 1.67 that this pathway was not fully derepressed after glucose ATP18 1.10 1.20 1.78 had been depleted. The expression levels of the specific Zeng et al. AMB Expr (2016) 6:51 Page 11 of 13 second step of gluconeogenesis were increased, and to Discussion a significantly larger extent in the xylose alone medium The transcription profiles of an industrial recombinant than in the xylose fermentation phase in the mixed-sugar S. cerevisiae strain was compared between different fer - medium. The generation of phosphoenolpyruvate from mentation states of glucose and xylose both in single oxaloacetate is accompanied by a loss of 1 carbon as CO , sugar and mixed sugar media. The presence of glucose and consumption of 1 ATP. The upregulation ratio of repressed the utilization of xylose, and it is found that FBP1 was consistent to that of PCK1. the transcriptome of the glucose fermentation phase As to the regeneration of NAD , since alcoholic fer- in the mixed-sugar medium was very similar to that in mentation itself is a redox-neutral process (the NADH the glucose alone medium. A recently published gene reduced in ethanol formation accounts for the NADH expression study using mixed sugar medium found that generated in the glyceraldehyde-3-phosphate dehy- respiratory genes were not fully repressed when xylose drogenase reaction), the accumulated NADH must was present with abundant glucose (Alff-Tuomala et  al. be otherwise reoxidized. Despite the impermeability 2016), which was not observed in our study. On the other of the inner mitochondrial membrane for NADH and hand, we found that although the transcriptome highly NAD , both cytosolic and mitochondrial NADH can differed between the xylose metabolizing cells cultured be reoxidized by the respiratory chain in S. cerevisiae. in the xylose alone medium and those in the xylose fer- The cytosolic NAD can be generated via mitochon- mentation phase of the mixed sugar medium, the xylose drial external NADH hydrogenases (Bakker et al. 2001), consumption rate was nearly identical under these two encoded by NDE1 and NDE2. Alternatively, cytosolic conditions. It appears that S. cerevisiae has adequately NADH can be reoxidized by the respiratory chain via evolved mechanisms to respond to metabolites accumu- the glycerol- 3-phosphate shuttle consisting of cytosolic lated in the process of glucose fermentation. NADH-linked glycerol-3-phosphate dehydrogenase and The main challenge in the recombinant strains express - mitochondrial FAD linked glycerol-3-phosphate dehy- ing S. stipitis XR-XDH pathway is redox imbalance, drogenase encoded by GUT2 that transfers electrons because xylose reductase prefers NADPH, whereas from cytosolic glycerol- 3-phosphate to ubiquinone. The xylitol dehydrogenase strictly utilizes NAD , leading expression level of NDE1, NDE2 and GUT2 increased to the accumulation of NADP and NADH. NADPH is significantly with xylose in our study, both in the xylose regenerated in S. cerevisiae mainly through the oxida- alone medium and during the xylose fermentation phase tive branch of the pentose phosphate pathway and the in the mixed sugar medium (Table  4 and Fig.  5). The cytosolic and peroxisomal isocitrate dehydrogenases. We expression levels of many others genes encoding respira- found that the expression levels of ZWF1 and GND2 in tory enzymes such as cytochrome c reductase that trans- oxidative PPP, and that of IDP2 and IDP3 were upregu- fers electrons from ubiquinone to cytochrome c and lated on xylose (Fig.  5), which corroborated with earlier cytochrome c oxidase, which catalyzes the oxidation of reports either in aerobic or anaerobic conditions (Mat- cytochrome c by molecular oxygen, also increased with sushika et  al. 2014; Salusjärvi et  al. 2008). It is known xylose (Table 4). that the ratio of NADP /NADPH directly influence the It is assumed that xylose is poorly metabolized to eth- activity of glucose-6-phosphate dehydrogenase, while anol because it does not repress respiration in the same the dehydrogenation of glucose-6-phosphate catalyzed manner as glucose ( Jin et al. 2004). S. cerevisiae growing by ZWF1 is the rate-limiting reaction in PPP. When one in glucose exhibits the Crabtree effect, where respira - mole of ribulose-5-P is produced from glucose-6-phos- tory growth can only be achieved with a limited sugar phate through oxidative PPP generating two moles of supply and at low specific growth rate. It is possible NADPH, one carbon is lost as CO , leading to the poor that the lack of ability to repress respiration by xylose- use of carbon substrate. metabolizing cells is the consequence of a low level of Meanwhile, the gluconeogenesis pathway was upregu- xylose uptake by S. cerevisiae. However, it has been lated to feed glucose-6-phosphate to oxidative PPP, con- previously reported that the overexpression of xylose- sistent with an earlier report (Runquist et  al. 2009). In transporting protein did not enhance xylose fermenta- our study, differences in the expression level were slight tion (Hamacher et  al. 2002). It is then possible that the for PYC1 and PYC2, which encode cytoplasmic pyru- upregulation of respiration is driven by the demand for vate carboxylase that converts pyruvate to oxaloacetate ATP and NAD . However, the metabolic flux analy - in the first step of gluconeogenesis. However, cytoplas - sis of a xylose isomerase-expressing strain, which does mic oxaloacetate can also be produced from acetyl-coA not face the NADH accumulation challenge during through the glyoxylate cycle, which was upregulated on xylose assimilation, also indicated that xylose failed to xylose (Fig.  4). The expression levels of PCK1 encoding elicit the full carbon catabolite repression response phosphoenolpyruvate carboxykinase that catalyzes the Zeng et al. AMB Expr (2016) 6:51 Page 12 of 13 References (Wasylenko and Stephanopoulos 2015). As for ATP Alff‑ Tuomala S, Salusjärvi L, Barth D, Oja M, Penttilä M, Pitkänen JP, Ruohonen metabolism, since ADP activates the Snf1 regulator by L, Jouhten P. Xylose‑induced dynamic effects on metabolism and gene protecting it against dephosphorylation (Mayer et  al. expression in engineered Saccharomyces cerevisiae in anaerobic glucose‑ xylose cultures. Appl Microbiol Biotechnol. 2016;100(2):969–85. 2011), it is tempting to infer, from the increased Snf1- Bakker BM, Overkamp KM, Maris AJAV, Kötter P, Luttik MAH, Dijken JPV, Pronk activated transcription factors Adr1 and Cat8 expres- JT. Stoichiometry and compartmentation of NADH metabolism in Sac- sion levels as well as the decreased Snf1-deactivated charomyces cerevisiae. FEMS Microbiol Rev. 2001;25:15–37. Batt CA, Carvallo S, Easson DD, Akedo M, Sinskey AJ. Direct evidence for Mig1 expression levels (Fig.  4), that the intracellular a xylose metabolic pathway in Saccharomyces cerevisiae. Biotechnol ADP level increased in xylose metabolizing cells. Coin- Bioeng. 1986;28:549–53. cidentally, high requirements for maintenance energy Caro AA, Cederbaum AI. Antioxidant properties of S‑adenosyl‑ l ‑methionine in 2+ Fe ‑initiated oxidations. Free Radical Biol Med. 2004;36:1303–16. by the recombinant xylose metabolizing S. cerevisiae Chang Q, Griest TA, Harter TM, Petrash JM. Functional studies of aldok ‑ eto reduc‑ was found in a metabolic analysis using XR-XDH strains tases in Saccharomyces cerevisiae. BBAM ‑ ol Cell Res. 2007;1773:321–9. (Feng and Zhao 2013b). Meanwhile, the TCA cycle and Feng X, Zhao H. Investigating host dependence of xylose utilization in recombinant Saccharomyces cerevisiae strains using RNA‑seq analysis. respiratory enzymes were found to be upregulated even Biotechnol Biofuels. 2013a;6:96. under anaerobic xylose fermenting conditions where no Feng X, Zhao H. Investigating xylose metabolism in recombinant Saccharo- ATP can be generated by oxidative phosphorylation due myces cerevisiae via C metabolic flux analysis. Microb Cell Factories. 2013b;12(1):1. to the absence of oxygen (Matsushika et al. 2014). Taken Gancedo JM. Yeast carbon catabolite repression. Microbiol Mol Biol Rev. together, we propose that it is crucial to reduce the 1998;62:334–61. maintenance energy of the xylose metabolizing strains Hamacher T, Becker J, Gardonyi M, Hahn‑Hägerdal B, Boles E. Characterization of the xylose‑transporting properties of yeast hexose transporters and to improve xylose utilization. Moreover, the regulatory their influence on xylose utilization. Microbiology. 2002;148:2783–8. network for xylose fermentation appeared sub-optimal, Hirasawa T, Yoshikawa K, Nakakura Y, Nagahisa K, Furusawa C, Katakura Y, as futile induction could occur. The regulatory network Shimizu H, Shioya S. Identification of target genes conferring ethanol stress tolerance to Saccharomyces cerevisiae based on DNA microarray could alternatively be refigured by transcription factor data analysis. J Biotechnol. 2007;131:34–44. engineering or adaptive evolution. Jin YS, Laplaza JM, Jeffries TW. Saccharomyces cerevisiae engineered for xylose metabolism exhibits a respiratory response. Appl Environ Microbiol. 2004;70:6816–25. Additional file Kida K, Kume K, Morimura S, Sonoda Y. Repeated‑batch fermentation process using a thermotolerant flocculating yeast constructed by protoplast fusion. J Ferment Bioeng. 1992;74:169–73. Additional file 1: Supplementary material. Kuttykrishnan S, Sabina J, Langton LL, Johnston M, Brent MR. A quantitative model of glucose signaling in yeast reveals an incoherent feed forward loop leading to a specific, transient pulse of transcription. Proc Natl Acad Abbreviations Sci USA. 2010;107:16743–8. XR: xylose reductase; XDH: xylitol dehydrogenase; XI: xylose isomerase; YPX: Li BZ, Cheng JS, Ding MZ, Yuan YJ. Transcriptome analysis of differential yeast extract peptone xylose; YPDX: yeast extract peptone dextrose xylose; responses of diploid and haploid yeast to ethanol stress. J Biotechnol. SAM: S‑adenosylmethionine; TCA: tricarboxylic acid; GEO: gene expression 2010;148:194–203. omnibus. Li YC, Mitsumasu K, Gou ZX, Gou M, Tang YQ, Li GY, Wu XL, Akamatsu T, Taguchi H, Kida K. Xylose fermentation efficiency and inhibitor tolerance of the Authors’ contributions recombinant industrial Saccharomyces cerevisiae strain NAPX37. Appl WYZ carried out the experiment, analyzed the batch fermentation and Microbiol Biotechnol. 2015;100:1531–42. transcriptome data, and drafted the manuscript. YQT designed the research, Matsushika A, Inoue H, Kodaki T, Sawayama S. Ethanol production from xylose helped with the data analysis, and revised the manuscript. MG and ZYX in engineered Saccharomyces cerevisiae strains, current state and per‑ helped with the experiments. KK participated in the design of the research. All spectives. Appl Microbiol Biotechnol. 2009a;84(1):37–53. authors read and approved the final manuscript. Matsushika A, Inoue H, Murakami K, Takimura O, Sawayama S. Bioethanol production performance of five recombinant strains of laboratory and industrial xylose‑fermenting Saccharomyces cerevisiae. Bioresour Technol. Acknowledgements 2009b;100:2392–8. We thank CapitalBio Technology Inc. Beijing for the help in DNA microarray Matsushika A, Goshima T, Fujii T, Inoue H, Sawayama S, Yano S. Characterization analysis. of non‑ oxidative transaldolase and transketolase enzymes in the pentose phosphate pathway with regard to xylose utilization by recombinant Sac- Competing interests charomyces cerevisiae. Enzyme Microb Technol. 2012;51:16–25. The authors declare that they have no competing interests. Matsushika A, Goshima T, Hoshino T. Transcription analysis of recombinant industrial and laboratory Saccharomyces cerevisiae strains reveals the Availability of supporting data molecular basis for fermentation of glucose and xylose. Microb Cell The microarray datasets analyzed in this article are available in Gene Expres‑ Factories. 2014;13(1):1. sion Omnibus, accession number GSE80748. Mayer FV, Heath R, Underwood E, Sanders MJ, Carmena D, McCartney RR, Leiper FC, Xiao B, Jing C, Walker PA, Haire LF, Ogrodowicz R, Martin SR, Funding Schmidt MC, Gamblin SJ, Carling D. ADP regulates SNF1, the Saccharo- This work was supported by the National Natural Science Foundation of China myces cerevisiae homolog of AMP‑activated protein kinase. Cell Metab. (31170093). 2011;14:707–14. Palomino A, Herrero P, Moreno F. Rgt1, a glucose sensing transcription factor, is Received: 21 July 2016 Accepted: 21 July 2016 required for transcriptional repression of the HXK2 gene in Saccharomyces cerevisiae. Biochem J. 2005;388:697–703. Zeng et al. AMB Expr (2016) 6:51 Page 13 of 13 Peng B, Shen Y, Li X, Chen X, Hou J, Bao X. Improvement of xylose fermenta‑ Salusjärvi L, Kankainen M, Soliymani R, Pitkänen JP, Penttilä M, Ruohonen L. tion in respiratory‑ deficient xylose ‑fermenting Saccharomyces cerevisiae. Regulation of xylose metabolism in recombinant Saccharomyces cerevi- Metab Eng. 2012;14:9–18. siae. Microb Cell Factories. 2008;7:18. Regenberg B, Grotkjaer T, Winther O, Fausboll A, Akesson M, Bro C, Hansen LK, Tang Y, An M, Liu K, Nagai S, Shigematsu T, Morimura S, Kida K. Ethanol produc‑ Brunak S, Nielsen J. Growth‑rate regulated genes have profound impact tion from acid hydrolysate of wood biomass using the flocculating yeast on interpretation of transcriptome profiling in Saccharomyces cerevisiae. Saccharomyces cerevisiae strain KF‑7. Process Biochem. 2006;41:909–14. Genome Biol. 2006;7:R107. Thomas D, SurdinKerjan Y. Metabolism of sulfur amino acids in Saccharomyces Runquist D, Hahn‑Hägerdal B, Bettiga M. Increased expression of the oxidative cerevisiae. Microbiol Mol Biol Rev. 1997;61:503. pentose phosphate pathway and gluconeogenesis in anaerobically Wasylenko TM, Stephanopoulos G. Metabolomic and C‑metabolic flux growing xylose‑utilizing Saccharomyces cerevisiae. Microb Cell Factories. analysis of a xylose‑ consuming Saccharomyces cerevisiae strain expressing 2009;8:49. xylose isomerase. Biotechnol Bioeng. 2015;112:470–83. Salusjärvi L, Pitkänen JP, Aristidou A, Ruohonen L, Penttilä M. Transcription Zaman S, Lippman SI, Zhao X, Broach JR. How Saccharomyces responds to analysis of recombinant Saccharomyces cerevisiae reveals novel responses nutrients. Annu Rev Genet. 2008;42:27–81. to xylose. Appl Biochem Biotechnol. 2006;128:237–73.

Journal

AMB Express – Springer Journals

Published: Aug 2, 2016

There are no references for this article.