Extra metabolic burden by displaying over secreting: Growth, fermentation and enzymatic activity in cellobiose of recombinant yeast expressing β-glucosidase

β-glucosidases from Saccharomyces cerevisiae: production, protein precipitation, characterization, and application in the enzymatic hydrolysis of delignified sugarcane bagasse

β-glucosidase is an essential enzyme for the enzymatic hydrolysis of lignocellulosic biomass, as it catalyzes the final stage of cellulose breakdown, releasing glucose. This paper aims to produce β-glucosidase from Saccharomyces cerevisiae and evaluate the enzymatic degradation of delignified sugarcane bagasse. S. cerevisiae was grown in yeast peptone dextrose medium. Partial purification of the enzyme was achieved through precipitating proteins with ethanol, and the optimal activity was measured by optimizing pH and temperature. The effects of ions, glucose tolerance, and heat treatment were evaluated. Delignified sugarcane bagasse was hydrolyzed by the enzyme. β-glucosidase showed a specific activity of 14.0712 ± 0.0207 U mg-1. Partial purification showed 1.22-fold purification. The optimum pH and temperature were 6.24 and 54 °C, respectively. β-glucosidase showed tolerance to glucose, with a relative activity of 71.27 ± 0.16%. Thermostability showed a relative activity of 58.84 ± 0.91% at 90 °C. The hydrolysis of delignified sugarcane bagasse showed a conversion rate of 87.97 ± 0.10% in the presence of Zn2+, an ion that promoted the highest increase in enzymatic activity. S. cerevisiae produced an extracellular β-glucosidase with good stability at pH and temperatures conventionally applied in the hydrolysis of lignocellulosic biomass, showing viability for industrial application.

Overexpression of LAS21 in Cellulase-Displaying Saccharomyces cerevisiae for High-Yield Ethanol Production from Pretreated Sugarcane Bagasse

The valorization of lignocellulosic feedstocks into biofuels and biochemicals has received much attention due to its environmental friendliness and sustainability. However, engineering an ideal microorganism that can both produce sufficient cellulases and ferment ethanol is highly challenging. In this study, we have tested seven different genes that are involved in glycosylphosphatidylinositol (GPI) biosynthesis and remodeling for the improvement of cellulase activity tethered on the S. cerevisiae cell surface. It was found that the overexpression of LAS21 can improve β-glucosidase activity by 48.8% compared to the original strain. Then, the three cellulase genes (cellobiohydrolase, endoglucanase, and β-glucosidase) and the LAS21 gene were co-introduced into a diploid thermotolerant S. cerevisiae strain by a multiple-round transformation approach, resulting in the cellulolytic ECBLCCE5 strain. Further optimization of the bioprocess parameters was found to enhance the ethanol yield of the ECBLCCE5 strain. Scaling up the valorization of pretreated sugarcane bagasses in a 1 L bioreactor resulted in a maximum ethanol concentration of 28.0 g/L (86.5% of theoretical yield). Our study provides a promising way to improve the economic viability of second-generation ethanol production. Moreover, the engineering of genes involved in GPI biosynthesis and remodeling can be applied to other yeast cell surface display applications.

Rapid evolution and mechanism elucidation for efficient cellobiose-utilizing Saccharomyces cerevisiae through Synthetic Chromosome Rearrangement and Modification by LoxPsym-mediated Evolution

Lack of cellobiose utilization capability for many microorganisms results in carbon source waste in lignocellulosic biorefinery. In this study, genes for cellobiose transport and hydrolysis were introduced to Saccharomyces cerevisiae synV, a semi-synthetic yeast with an inducible SCRaMbLE (Synthetic Chromosome Rearrangement and Modification by LoxPsym-mediated Evolution) system incorporated into its chromosome V, endowing cellobiose utilization capability to this strain. Thereafter, two evolved strains with 98.1% and 79.2% improvement, respectively, in cellobiose utilization rate were obtained through induced SCRaMbLE. Further studies suggested that the enhanced cellobiose utilization capability directly correlated with copy number increases of introduced genes and some chromosome structural variations. In particular, it was experimentally demonstrated for the first time that deletion of redox stress related gene MXR1 and ATP conversion related gene ADK2 contributed to enhanced cellobiose conversion. Thereafter, the effectiveness of MXR1 and ADK2 deletions was demonstrated in artificial hydrolysate and rice straw hydrolysate, respectively.

Comparison of the Unfolded Protein Response in Cellobiose Utilization of Recombinant Angel- and W303-1A-Derived Yeast Expressing β-Glucosidase

The unfolded protein response (UPR) is one of the most important protein quality control mechanisms in cells. At least, three factors are predicted to activate the UPR in yeast cells during fermentation. Using UPRE-lacZ as a reporter, we constructed two indicator strains, KZ and WZ, based on Angel-derived K-a and W303-1A strains, respectively, and investigated their UPR response to tunicamycin, ethanol, and acetic acid. Then, four strains carrying plasmids BG-cwp2 and BG were obtained to realize the displaying and secretion of β-glucosidase, respectively. The results of cellobiose utilization assays indicated interactions between the UPR and the metabolic burden between the strain source, anchoring moiety, oxygen supply, and cellobiose concentration. Meanwhile, as expected, growth (OD600), β-glucosidase, and β-galactosidase activities were shown to have a positive inter-relationship, in which the values of the KZ-derived strains were far lower than those of the WZ-derived strains. Additionally, extra metabolic burden by displaying over secreting was also much more serious in strain KZ than in strain WZ. The maximum ethanol titer of the four strains (KZ (BG-cwp2), KZ (BG), WZ (BG-cwp2), and WZ (BG)) in oxygen-limited 10% cellobiose fermentation was 3.173, 5.307, 5.495, and 5.486% (v/v), respectively, and the acetic acid titer ranged from 0.038 to 0.060% (v/v). The corresponding maximum values of the ratio of β-galactosidase activity to that of the control were 3.30, 5.29, 6.45, and 8.72, respectively. Under aerobic conditions with 2% cellobiose, those values were 3.79, 4.97, 6.99, and 7.67, respectively. A comparison of the results implied that β-glucosidase expression durably induced the UPR, and the effect of ethanol and acetic acid depended on the titer produced. Further study is necessary to identify ethanol- or acid-specific target gene expression. Taken together, our results indicated that the host strain W303-1A is a better secretory protein producer, and the first step to modify strain K-a for cellulosic ethanol fermentation would be to relieve the bottleneck of UPR capacity. The results of the present study will help to identify candidate host strains and optimize expression and fermentation by quantifying UPR induction.

Burden Imposed by Heterologous Protein Production in Two Major Industrial Yeast Cell Factories: Identifying Sources and Mitigation Strategies

Production of heterologous proteins, especially biopharmaceuticals and industrial enzymes, in living cell factories consumes cellular resources. Such resources are reallocated from normal cellular processes toward production of the heterologous protein that is often of no benefit to the host cell. This competition for resources is a burden to host cells, has a negative impact on cell fitness, and may consequently trigger stress responses. Importantly, this often causes a reduction in final protein titers. Engineering strategies to generate more burden resilient production strains offer sustainable opportunities to increase production and profitability for this growing billion-dollar global industry. We review recently reported impacts of burden derived from resource competition in two commonly used protein-producing yeast cell factories: Saccharomyces cerevisiae and Komagataella phaffii (syn. Pichia pastoris). We dissect possible sources of burden in these organisms, from aspects related to genetic engineering to protein translation and export of soluble protein. We also summarize advances as well as challenges for cell factory design to mitigate burden and increase overall heterologous protein production from metabolic engineering, systems biology, and synthetic biology perspectives. Lastly, future profiling and engineering strategies are highlighted that may lead to constructing robust burden-resistant cell factories. This includes incorporation of systems-level data into mathematical models for rational design and engineering dynamical regulation circuits in production strains.

Efficient, Flexible Autoinduction Expression Systems with Broad Initiation in Bacillus subtilis

Low expression levels and inflexible induction initiation have been the main obstacles to produce proteins using bacterial quorum sensing (QS). The typical QS system in Bacillus subtilis, ComQXPA, activates the promoter P_srfA using ComX and ComA as an auto-inducer and a promoter activator, respectively. Here, we developed a series of flexible autoinduction expression systems in B. subtilis WB600 based on ComQXPA using a super-folder green fluorescent protein as the reporter. The -35 region of P_srfA was replaced with corresponding conserved sequences of σA-dependent promoters, yielding P₁ with 85% enhanced strength. We then applied a semi-rational design within the spacer between the -35 and -15 regions of P₁ to generate the QS promoter P_S1E, which generated 8.22-fold more expression than P_srfA. Based on P_S1E, we finally obtained three types of autoinduction expression systems with initiation ranging from 1.5-9.5 h by optimizing the combination of the promoters for ComX and ComA. The yield of Bacillus deramificans pullulanase generated using autoinduction expression systems in B. subtilis reached 80.2 U/mL, which was 36% more than that of the most powerful constitutive promoter P₅₆₆. Flexible autoinduction expression systems with diverse dynamic features have considerable potential for improving protein expression and metabolite production in B. subtilis.

Stress modulation as a means to improve yeasts for lignocellulose bioconversion

The second-generation (2G) fermentation environment for lignocellulose conversion presents unique challenges to the fermentative organism that do not necessarily exist in other industrial fermentations. While extreme osmotic, heat, and nutrient starvation stresses are observed in sugar- and starch-based fermentation environments, additional pre-treatment-derived inhibitor stress, potentially exacerbated by stresses such as pH and product tolerance, exist in the 2G environment. Furthermore, in a consolidated bioprocessing (CBP) context, the organism is also challenged to secrete enzymes that may themselves lead to unfolded protein response and other stresses. This review will discuss responses of the yeast Saccharomyces cerevisiae to 2G-specific stresses and stress modulation strategies that can be followed to improve yeasts for this application. We also explore published -omics data and discuss relevant rational engineering, reverse engineering, and adaptation strategies, with the view of identifying genes or alleles that will make positive contributions to the overall robustness of 2G industrial strains. KEYPOINTS: • Stress tolerance is a key driver to successful application of yeast strains in biorefineries. • A wealth of data regarding stress responses has been gained through omics studies. • Integration of this knowledge could inform engineering of fit for purpose strains.

Rational Design of the N-Terminal Coding Sequence for Regulating Enzyme Expression in Bacillus subtilis

Synonymous mutation of the N-terminal coding sequence (NCS) has been used to regulate gene expression. We here developed a statistical model to predict the effect of the NCSs on protein expression in Bacillus subtilis WB600. First, a synonymous mutation was performed within the first 10 residues of a superfolder green fluorescent protein to generate a library of 172 NCS synonymous mutants with different expression levels. A prediction model was then developed, which adopted G/C frequency at the third position of each codon and minimum free energy of mRNA as the independent variables, using multiple regression analysis between the 11 sequence parameters of the NCS and their fluorescence intensities. By designing the NCS of the 10 signal peptides de novo according to the model, the extracellular yield of B. subtilis pullulanase fused to each signal peptide was up-regulated by up to 515% or down-regulated by at most 79%. This work provided a candidate tool for fine-tuning gene expression or enzyme production in B. subtilis.

Repetitive δ-integration of a cellulase-encoding gene into the chromosome of an industrial Angel Yeast-derived strain by URA3 recycling

Genetic modification of industrial yeast strains often faces more difficulties than that of laboratory strains. Thus, new approaches are still required. In this research, the Angel Yeast-derived haploid strain Kα was genetically modified by multiple rounds of δ-integration, which was achieved via URA3 recycling. Three δ-integrative plasmids, pGδRU, pGδRU-BGL, and pGδRU-EG, were first constructed with two 167 bp δ sequences and a repeat-URA3-repeat fragment. Then, the δ-integrative strains containing the bgl1 or egl2 gene were successfully obtained by one-time transformation of the linearized pGδRU-BGL or pGδRU-EG fragment, respectively. Their counterparts in which the URA3 gene was looped out were also easily isolated by selection for growth on 5´-fluoroorotic acid plates, although the ratio of colonies lacking URA3 to the total number of colonies decreased with increasing copy number of the corresponding integrated cellulase-encoding gene. Similar results were observed during the second round of δ-integration, in which the δ-integration strain Kα(δ::bgl1-repeat) obtained from the first round was transformed with a linearized pGδRU-EG fragment. After 10 rounds of cell growth and transfer to fresh medium, the doubling times and enzyme activities of Kα(δ::bgl1-repeat), Kα(δ::egl2-repeat), and Kα(δ::bgl1-repeat)(δ::egl2-repeat) showed no significant change and were stable. Further, their maximum ethanol concentrations during simultaneous saccharification and fermentation of pretreated corncob over a 7-day period were 46.35, 33.13, and 51.77 g/L, respectively, which were all substantially higher than the parent Kα strain. Thus, repetitive δ-integration with URA3 recycling can be a feasible and valuable method for genetic engineering of Angel Yeast. These results also provide clues about some important issues related to δ-integration, such as the structural stability of δ-integrated genes and the effects of individual integration-site locations on gene expression. Further be elucidation of these issues should help to fully realize the potential of δ-integration-based methods in industrial yeast breeding.

Delta-Integration of Single Gene Shapes the Whole Metabolomic Short-Term Response to Ethanol of Recombinant Saccharomyces cerevisiae Strains

In yeast engineering, metabolic burden is often linked to the reprogramming of resources from regular cellular activities to guarantee recombinant protein(s) production. Therefore, growth parameters can be significantly influenced. Two recombinant strains, previously developed by the multiple δ-integration of a glucoamylase in the industrial Saccharomyces cerevisiae 27P, did not display any detectable metabolic burden. In this study, a Fourier Transform InfraRed Spectroscopy (FTIR)-based assay was employed to investigate the effect of δ-integration on yeast strains’ tolerance to the increasing ethanol levels typical of the starch-to-ethanol industry. FTIR fingerprint, indeed, offers a holistic view of the metabolome and is a well-established method to assess the stress response of microorganisms. Cell viability and metabolomic fingerprints have been considered as parameters to detecting any physiological and/or metabolomic perturbations. Quite surprisingly, the three strains did not show any difference in cell viability but metabolomic profiles were significantly altered and different when the strains were incubated both with and without ethanol. A LC/MS untargeted workflow was applied to assess the metabolites and pathways mostly involved in these strain-specific ethanol responses, further confirming the FTIR fingerprinting of the parental and recombinant strains. These results indicated that the multiple δ-integration prompted huge metabolomic changes in response to short-term ethanol exposure, calling for deeper metabolomic and genomic insights to understand how and, to what extent, genetic engineering could affect the yeast metabolome.

Metabolomic Alterations Do Not Induce Metabolic Burden in the Industrial Yeast M2n[pBKD2-Pccbgl1]-C1 Engineered by Multiple δ-Integration of a Fungal β-Glucosidase Gene

In the lignocellulosic yeast development, metabolic burden relates to redirection of resources from regular cellular activities toward the needs created by recombinant protein production. As a result, growth parameters may be greatly affected. Noteworthy, Saccharomyces cerevisiae M2n[pBKD2-Pccbgl1]-C1, previously developed by multiple δ-integration of the β-glucosidase BGL3, did not show any detectable metabolic burden. This work aims to test the hypothesis that the metabolic burden and the metabolomic perturbation induced by the δ-integration of a yeast strain, could differ significantly. The engineered strain was evaluated in terms of metabolic performances and metabolomic alterations in different conditions typical of the bioethanol industry. Results indicate that the multiple δ-integration did not affect the ability of the engineered strain to grow on different carbon sources and to tolerate increasing concentrations of ethanol and inhibitory compounds. Conversely, metabolomic profiles were significantly altered both under growing and stressing conditions, indicating a large extent of metabolic reshuffling involved in the maintenance of the metabolic homeostasis. Considering that four copies of BGL3 gene have been integrated without affecting any parental genes or promoter sequences, deeper studies are needed to unveil the mechanisms implied in these metabolomic changes, thus supporting the optimization of protein production in engineered strains.

Ameliorating the Metabolic Burden of the Co-expression of Secreted Fungal Cellulases in a High Lipid-Accumulating Yarrowia lipolytica Strain by Medium C/N Ratio and a Chemical Chaperone

Yarrowia lipolytica, known to accumulate lipids intracellularly, lacks the cellulolytic enzymes needed to break down solid biomass directly. This study aimed to evaluate the potential metabolic burden of expressing core cellulolytic enzymes in an engineered high lipid-accumulating strain of Y. lipolytica. Three fungal cellulases, Talaromyces emersonii-Trichoderma reesei chimeric cellobiohydrolase I (chimeric-CBH I), T. reesei cellobiohydrolase II (CBH II), and T. reesei endoglucanase II (EG II) were expressed using three constitutive strong promoters as a single integrative expression block in a recently engineered lipid hyper-accumulating strain of Y. lipolytica (HA1). In yeast extract-peptone-dextrose (YPD) medium, the resulting cellulase co-expressing transformant YL165-1 had the chimeric-CBH I, CBH II, and EG II secretion titers being 26, 17, and 132 mg L^-1, respectively. Cellulase co-expression in YL165-1 in culture media with a moderate C/N ratio of ∼4.5 unexpectedly resulted in a nearly two-fold reduction in cellular lipid accumulation compared to the parental control strain, a sign of cellular metabolic drain. Such metabolic drain was ameliorated when grown in media with a high C/N ratio of 59 having a higher glucose utilization rate that led to approximately twofold more cell mass and threefold more lipid production per liter culture compared to parental control strain, suggesting cross-talk between cellulase and lipid production, both of which involve the endoplasmic reticulum (ER). Most importantly, we found that the chemical chaperone, trimethylamine N-oxide dihydride increased glucose utilization, cell mass and total lipid titer in the transformants, suggesting further amelioration of the metabolic drain. This is the first study examining lipid production in cellulase-expressing Y. lipolytica strains under various C/N ratio media and with a chemical chaperone highlighting the metabolic complexity for developing robust, cellulolytic and lipogenic yeast strains.

The metabolic switch can be activated in a recombinant strain of Streptomyces lividans by a low oxygen transfer rate in shake flasks

Background

In Streptomyces, understanding the switch from primary to secondary metabolism is important for maximizing the production of secondary metabolites such as antibiotics, as well as for optimizing recombinant glycoprotein production. Differences in Streptomyces lividans bacterial aggregation as well as recombinant glycoprotein production and O-mannosylation have been reported due to modifications in the shake flask design. We hypothetized that such differences are related to the metabolic switch that occurs under oxygen-limiting conditions in the cultures.

Results

Shake flask design was found to affect undecylprodigiosin (RED, a marker of secondary metabolism) production; the RED yield was 12 and 385 times greater in conventional normal Erlenmeyer flasks (NF) than in baffled flasks (BF) and coiled flasks (CF), respectively. In addition, oxygen transfer rates (OTR) and carbon dioxide transfer rates were almost 15 times greater in cultures in CF and BF as compared with those in NF. Based on these data, we obtained respiration quotients (RQ) consistent with aerobic metabolism for CF and BF, but an RQ suggestive of anaerobic metabolism for NF.

Conclusion

Although the metabolic switch is usually related to limitations in phosphate and nitrogen in Streptomyces sp., our results reveal that it can also be activated by low OTR, dramatically affecting recombinant glycoprotein production and O-mannosylation and increasing RED synthesis in the process.

Electronic supplementary material

The online version of this article (10.1186/s12934-018-1035-3) contains supplementary material, which is available to authorized users.