Plasmids and their construction
The two heterologous genes, miox4 gene encoding MIOX4 enzyme in A. thaliana [16] and the udh gene encoding UDH in P. syringae [17], were codon-optimized according to the codon preference of S. cerevisiae and synthesized by Sangon Biotech (Shanghai, China). Then these two genes were ligated to the plasmid of pY26-GPD-TEF, purchased from Miaoling Bioscience & Technology Co., Ltd. (Wuhan, China), using the combinations of restriction enzymes, BglII/NotI and EcoRI/XhoI, to generate pY26-miox4-udh (Table S1). This recombinant vector carrying miox4 and udh under the control of the promoters Pgpd and Ptef, respectively, were used for the subsequent transformation of S. cerevisiae.
pUG6 from Miaoling Bioscience & Technology Co., Ltd. (Wuhan, China) with kanamycin resistance gene was used as the template for the construction of the knock out cassette loxP-Kan-loxP by polymerase chain reaction (PCR) using the primers knock-OPI1F/R (Table S2). The plasmid pSH47 from Miaoling Bioscience & Technology Co., Ltd. (Wuhan, China) provided Cre recombinase for the self-recombination of the knock out cassette.
The plasmid pCAMBIA1300 [18, 19] was used as the backbone for constructing the recombinant vector pCA-Pcbh1-ips-Tcbh1 (Table S1), where the gene ips (GenBank: L23520.1) encoding myo-inositol-1-phosphate synthase from S. cerevisiae [2] was codon-optimized according to the codon preference of T. reesei and expressed heterologously under the control of the strong promoter Pcbh1 to improve the myo-inositol production.
Strains and media
All strains used in this work, including the starting strains and the engineered strains, are listed in Table S1.
LB medium, containing 1 g/L tryptone, 0.5 g/L yeast extract, and 1 g/L NaCl, was used to culture E. coli cells and Agrobacterium tumefaciens cells after being autoclaved at 121℃ for 20 min. YPD medium was used to culture S. cerevisiae cells, which had a following composition (g/L): 10 yeast extract, 20 peptone, and 20 glucose. When selective YPD medium was prepared, geneticin G418 was added to a specific concentration after being autoclaved and cooled down. Solid media were prepared by adding 2 g/L agar before autoclave.
The seed medium for T. reesei strains was composed of 10 g/L glucose, 1 g/L peptone, 5 mL Mandels nutrient salts solution [20], 2.5 mL citrate buffer (1 mol/L), 0.05 mL Mandels trace elements solution [20], and 0.1 g/L Tween 80. The seed medium was autoclaved at 121℃ for 20 min. The fermentation medium for cellulase production by T. reesei comprised of 15 g/L Avicel or 30 g/L pretreated lignocellulose (dry biomass), 1 g/L glucose, 6 g/L (NH4)2SO4, 2.0 g/L KH2PO4, 0.3 g/L CaCl2, 0.3 g/L MgSO4, 0.005 g/L FeSO4, 0.0016 g/L MnSO4, 0.0014 g/L ZnSO4 and 0.0037 g/L CoCl2. The initial pH was adjusted to 4.8 with citrate buffer. This fermentation medium was autoclaved at 121℃ for 30 min.
All chemicals except lignocellulosic materials were purchased from Sinopharm Chemical Reagent Co. Ltd., Shanghai, China. The media for SSF and CBP were detailed in the Methods sections about them.
Genetic engineering of S. cerevisiae
The cassette loxP-Kan-loxP amplified from pUG6 by PCR using the primers knock-OPI1F/R (Table S2) was transformed into the competent cells of S. cerevisiae INVSc1 prepared with the Li-Ac method [21] to knock out opi1. The KanMX gene disruption cassette was cured by the homologous recombination between the loxP sites mediated by Cre recombinase expressed by the plasmid pSH47. This plasmid pSH47 was lost as the host cells were cultivated continuously, resulting in the S. cerevisiae INVSc1 opi1Δ strain which was used as the host for constructing the biosynthetic pathway for D-glucaric acid production.
For this target, there were two ways to express foreign genes, episomal or integrative expression. In the scenario of the episomal expression, the recombinant vector pY26-miox4-udh was directly transformed into the competent cells of S. cerevisiae INVSc1 opi1Δ and the high glucaric acid-producing strain was screened. For the integrative expression, the overlap PCR was conducted to splice the expression cassettes with delta1 and delta2, targeting the integrations into Ty loci [3, 22]. First, delta1 and delta2 were amplified from S. cerevisiae genome using the primers delta1-F/delta1-R and delta2-F/delta2-R (Table S2) respectively, and the miox4 and udh expression cassettes were amplified from the plasmid pY26-miox4-udh using the primers MIXO4-F/R and UDH-F/R respectively. Second, the fragments d1-M-F and L-U-d2 were generated by the overlap PCR from delta1 and MIXO4 expression cassette using the primers delta1-F and FURA3-R and from UDH expression cassette and delta2 using the primers LURA-F/delta2-R, respectively. Finally, the whole fragment was gained from the overlap PCR from the fragments d1-M-F and L-U-d2 using the primers delta1-F and delta2-R, which was purified with Cyle-Pure Kit 200 (Omega Bio-tek, Georgia, USA) and used for the transformation of the competent cells of S. cerevisiae INVSc1 opi1Δ.
Genetic engineering of T. reesei
The recombinant plasmid pCA-Pcbh1-ips-Tcbh1 was transformed into T. reesei Rut-C30 by the method of A. tumefaciens mediated transformation (AMT) [23]. Then the potential T. reesei transformants were selected by the two rounds of screening, the first PDA (potato dextrose agar) plates added with hygromycin B and the second Avicel plates as described above [18, 19]. The fast-growing T. reesei transformants selected by the two rounds of screening were tested in the fermentation for myo-inositol production and used in the artificial microbial consortium to increase myo-inositol availability.
Fermentation for D-glucaric acid production by the engineered S. cerevisiae
Fermentation was carried out in 250 mL Erlenmeyer flasks with a working volume 50 mL of YPD medium with or without 10.8 g/L (60 mM) myo-inositol. Before inoculation into fermentation medium, S. cerevisiae strains were precultured in 5 mL YPD medium with 10.8 g/L (60 mM) myo-inositol in 50 mL Erlenmeyer flasks at 30℃ with a shaking of 250 rpm to an optical density at 600 nm (OD600) of ~5. Then, the cells were collected and inoculated into fermentation medium with or without 10.8 g/L (60 mM) myo-inositol to an OD600 of 0.1. Fermentation was implemented at 30℃ with a shaking of 250 rpm.
Fed-batch fermentation was conducted under the same conditions as the batch fermentation mentioned above, except that 5 g/L glucose was supplemented twice during the fermentation process, 24 h and 48 h respectively.
Separated hydrolysis and fermentation (SHF)
In the scenario of SHF (Fig. 1), the cellulase was produced by T. reesei Rut-C30 from Avicel or steam exploded corn stover (SECS) [14, 24, 25] was applied to the enzymatic hydrolysis of themselves, so called “on-site cellulase production” [13-15, 25]. T. reesei Rut-C30 was precultured in the seed medium for 36 h and then the seed collected and inoculated into the fermentation medium for cellulase production. The fermentation broth containing crude cellulase was directly used in the enzymatic hydrolysis of Avicel or SECS [14, 19], which was operated in 250 mL Erlenmeyer flasks with a working volume of 50 mL 2.5 mL 1 M citrate buffer solution (for final pH 4.8), 50 g/L Avicel or 100 g/L SECS (dry material), 25 FPIU/g glucan the cellulase harvested after 5 d fermentation, and a supplementary amount of water to make up 50 mL. The enzymatic hydrolysis was conducted in a rotary shaker at 50℃ with a shaking of 140 rpm for 48 h. The resulted enzymatic hydrolysates containing fermentable sugars were used to comprise of fermentation medium supplemented with the same nutrients as YPD medium except glucose. The following operations were the same as those described in the section of fermentation for D-glucaric acid production by the engineered S. cerevisiae.
Simultaneous saccharification and fermentation (SSF)
In the scenario of SSF (Fig. 1), the cellulase prepared by the same method in SHF was used in the enzymatic prehydrolysis of Avicel or SECS for 12 h. The prehydrolysis of SSF was performed in 250 mL Erlenmeyer flasks with 50 mL SSF reaction mixture containing 50 g/L Avicel or 100 g/L SECS (dry material), 25 FPIU/g glucan the cellulase harvested after 5 d fermentation, 6 g/L (NH4)2SO4, 2.0 g/L KH2PO4, 0.3 g/L MgSO4∙7H2O, 0.3 g/L CaCl2∙2H2O, 0.1 g/L Tween 80, 10 g/L peptone, 5 g/L yeast extract,, and a supplementary amount of water to make up 50 mL. The initial pH of the reaction mixture was adjusted to 4.8 with citrate buffer. The reaction mixture without cellulase was autoclaved at 121℃ for 30 min. After cellulase addition, the reaction mixture was incubated at 50℃ with a shaking of 140 rpm for 12 h. Then, the temperature was decreased to 33℃ (or specified in the main text when studying its effect on SSF) and the shaking was increased to 250 rpm. The precultured S. cerevisiae with an OD600 of ~5 was inoculated into the reaction mixture for the SSF to produce D-glucaric acid.
Consolidated Bioprocessing (CBP)
The medium for CBP had a following composition: 15 g/L (or specified in the main text when investigating its effect on CBP) Avicel or SECS (dry material), 1 g/L peptone, 1 g/L yeast extract, 10%(v/v) Mandels nutrient salts solution [20], 0.1%(v/v) trace elements solution [20], 5%(v/v) citrate buffer (1 mol/L), 0.1 g/L Tween 80. CBP medium was autoclaved at 121℃ for 30 min. T. reesei was precultured in the seed medium at 30℃ for 36 h and S. cerevisiae was precultured in YPD medium at 30℃ till an OD600 of ~5. Then T. reesei was inoculated into CBP medium and the inoculation of S. cerevisiae was implemented immediately or delayed (or specified in the main text when studying its effect on CBP). The inoculum ratio of T. reesei to S. cerevisiae was 1:1 (or specified in the main text when studying its effect on CBP). If other strain or species was inoculated, the detailed information would be specified in the main text. The total inocula were 10%(v/v) of the fermentation medium. CBP was carried out in 250 mL Erlenmeyer flasks with 50 mL medium at 30℃ with a shaking of 180 rpm. A. niger was precultured for 48 h by the same method as T. reesei if needed by CBP.
Analytical methods
Filter paper activity (FPA) of cellulase, representing the total enzymatic activity, was assayed by the method standardized by the International Union of Pure and Applied Chemistry (IUPAC) [26], which quantifies the total amount of the reducing sugars produced from 50 mg Whatman No.1 filter paper (1×6 cm strip) by cellulase within 60 min. One International Unit of FPA (FPIU) was defined as the amount of cellulase needed for producing 1 μmol reducing sugars in 1 min.
β-Glucosidase activity (BGA) was determined using the standard method [26] with the tiny modification, i.e. the substrate ρNPG (ρ-nitrophenyl-β-d-1,4-glucopiranoside) (Sigma-Aldrich, St. Louis, MO, USA). The amount of ρ-nitrophenol produced from ρNPG by β-glucosidase within 10 min was assayed using spectrophotometer at a wavelength of 400 nm. One International Unit of BGA (IU) was defined as the amount of β-glucosidase required for producing 1 μmol of ρ-nitrophenol from ρNPG in 1 min.
Cellobiohydrolase activity (BGA) was assayed according to the method modified from FPA measurement method [26]. Microcrystalline cellulose PH101 purchased from Sigma-Aldrich (St. Louis, MO, USA) in the form of 1% (w/v) suspension was used as the substrate for the reaction with the duration of 30 min. One Unit (1 U) of CBA was defined as the amount of enzyme required for producing 1 mg reducing sugars in 1 h.
High performance liquid chromatography (HPLC) was adopted to analyze and quantify D-glucaric acid, myo-inositol and sugars, where Shimadzu Prominence LC-20A system equipped with Bio-Rad Aminex HPX-87H (300 mm×7.8 mm) column was used. When D-glucaric acid was determined, an ultraviolet (UV) detector was employed to detect the eluate. While myo-inositol and sugars were determined, a refractive index (RI) detector was employed. Sulfuric acid (5 mmol/L) was used as the mobile phase and set at a flow rate of 0.6 mL/min. The column temperature was maintained at 50℃. Sample loading for each injection was 10 μL.
Liquid Chromatography-Mass Spectrometry (LC-MS) was used to characterize and identify D-glucaric acid, where Shimadzu L-30A and AB Sciex Triple TOF 5600 were employed. The column was Shimadzu Shim-pack XR-ODS 100L×2.0. The mobile phases, set at a flow rate of 0.15 mL/min, were the aqueous solution containing 1 mmol/L ammonium formate and 1%(v/v) formic acid (Mobile phase A) and the acetonitrile solution containing 1 mmol/L ammonium formate (Mobile phase B), respectively. Mobile phase B, B for short and the same below, was used as the eluent and the elution procedure was as follows: 0-12 min 30% B, 12-30 min 30%-65% B, 30-31 min 65%-95% B, 31-35 min 95% B, 35-36 min 95%-30% B, 36-40 min 30% B. Sample loading for each injection was 5 μL. Negative electrospray ionization mode was chosen to ionize samples. The flow rate of atomizing gas (N2) was 1.5 L/min. The temperature for CDL and HB was 200℃. Scanning scope (m/z) ranged from 150 to 300.