Corn fiber was obtained from Juneng Golden Corn Co. Ltd., (Shouguang, China). It was ground using an MF 10 basic Microfine grinder (IKA, Germany) with a 2.0-mm-hole diameter sieve, then stored in a sealed polyethylene bag at −20°C before use.

The liquid chromatography standards, including glucose, galactose, arabinose, xylose, mannose, galacturonic acid, glucuronic acid, furfural, HMF, and acetic acid, were purchased from Sigma-Aldrich Co. LLC (St. Louis, MO, United States). Reagent grade sulfuric acid (98%), anhydrous sodium acetate, sodium hydroxide, and ethanol were obtained from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China).

Penicillium oxalicum MCAX, an engineered strain in our laboratory (Gao et al., 2021), was used to produce cellulases.

Saccharomyces cerevisiae LF2 is capable of metabolizing xylose and glucose was maintained in our group. It was provided by Professor Xiaoming Bao from Qilu University of Technology, Shandong.

The chemical compositions of corn fiber were determined according to the National Renewable Energy Laboratory (NREL) protocols. The extractives were quantified according to the analytical method NREL/TP-510-42619 (Sluiter et al., 2005a), and the ash was determined by the incineration method described by Sluiter et al. (2005b). The contents of cellulose, hemicellulose (xylose, arabinose, and galactose), lignin, and acetyl group were determined by two-step hydrolyzing corn fiber with H₂SO₄ of 72% (w/w) and 4% (w/w), respectively, according to the analytical method NREL/TP-510-42618 (Sluiter et al., 2008a). The starch was quantified according to the analytical method NREL/TP-510-42619 (Sluiter and Sluiter, 2005). The protein was quantified according to the analytical method NREL/TP-510-42625 (Hames et al., 2008). The sample collected during the determination of acetyl group content was filtered by 0.22 μm Millipore Syringe Filters (Jinteng, China), and the filtrate was used to analyze glucuronic acid by high-performance liquid chromatography (HPLC) (Shimadzu, Japan), which was equipped with Shimadzu’s 20A refractive index detector. An Aminex HPX-87H column (300 × 7.8 mm) (Bio-Rad, Richmond, CA, United States) was used to separate the compounds at 60°C. The mobile phase was 0.005 M H₂SO₄ with a 0.5 ml/min flow rate.

The cellulase used in this study was produced by P. oxalicum MCAX according to the method suggested by Gao et al. (2021). The strain MCAX was cultivated in 50 ml of seed medium [g/L, wheat bran: 20, peptone: 10, glucose: 10, (NH₄)₂SO₄: 2, KH₂PO₄: 3, and MgSO₄: 0.5] in 300 ml Erlenmeyer flasks for 24 h at 30°C and 200 rpm on a rotary shaker. Then, the culture was inoculated into a fermentation medium [g/L, wheat bran: 30, bean cake powder: 15, microcrystalline cellulose: 30, (NH₄)₂SO₄: 2, KH₂PO₄: 5, and MgSO₄: 0.5] with an inoculation ratio of 10% (v/v) and incubated at 30°C and 200 rpm for 6 days. After fermentation, the supernatant was obtained by centrifugation at 10,000 rpm for 10 min and used as crude cellulase. The hydrolytic activities of the crude cellulase were as follows: filter paper activity (FPA) 6 stocktickerFPU/ml, xylanase 783 IU/ml, arabinofuranosidase 5 IU/ml, and β-xylosidase 1 IU/ml.

Pretreatment of corn fiber with dilute sulfuric acid was performed in a laboratory-scale vertical pressure steam sterilizer (Deqiang, China). A total of 4 g of corn fiber and dilute acid solution was thoroughly mixed in 50 ml flasks with 12 ml of reaction volume. Based on preliminary experiments, the following experimental conditions were tested: temperature of 105–125°C, H₂SO₄ concentration of 0.2–0.5% (w/v), and time of 10–60 min. The experiments were designed by a Box-Behnken RSM using Design-Expert version 8.0.6.1 software and are shown in Table 2.

For the significance analysis of different factors in the pretreatment, p-value was obtained by ANOVA and used for determining statistically significant. p > 0.05 indicates that the factor is not significant; p ≤ 0.05 indicates that it is significant; and p ≤ 0.01 indicates that it is very significant.

After pretreatment, the flasks were cooled down to room temperature. A part of the pretreated corn fiber slurries was collected to analyze the contents of furfural, HMF, formic acid, and acetic acid.

The saccharification was performed at 5% (w/v) substrate concentration in 50 ml flasks with a reaction volume of 20 ml. The pretreated corn fiber slurries were first adjusted to pH 5.0 with 10 M NaOH solution, mixed well with crude cellulase solution and acetate buffer of pH 4.8 in flasks, and incubated at 48°C in a thermostat air bath shaker set at 150 rpm. The samples were taken at reaction times of 0, 12, 24, and 48 h, respectively, and 72 h to analyze the concentrations of monomeric sugars in the enzymatic hydrolysates.

All experiments were performed in triplicate, and the average values are given in this study.

The hydrolysate of corn fiber was treated with dilute acid under the optimum pretreatment conditions, which contains 4.60 g/L glucose, 3.92 g/L xylose, 0.44 g/L furfural, 1.04 g/L formic acid, and 1.62 g/L acetic acid. For acclimation of the yeast, four different acclimation media were used. One of the media consisted of hydrolysate that was not diluted. The three other media were prepared by diluting hydrolysate to a concentration of 25, 50, and 75% (v/v), respectively. In each acclimation medium, 15.0 g/L glucose, 0.40 g/L (NH₄)₂SO₄, and 1.00 g/L yeast extract were added. S. cerevisiae LF2 was first inoculated into YPD medium for activation of 24 h at 30°C, then take 2 ml of inoculum, centrifuge, and transfer the yeast cells to a 50 ml Erlenmeyer flask containing 20 ml of the acclimation medium with the lowest inhibitors concentration and incubated at 30°C to the glucose content in the medium less than 90% of the initial glucose content. After that, the strain was further transferred to the next acclimation medium with a higher inhibitors’ concentration and cultured according to the procedure above until the highest inhibitors concentration. To make the yeast better adapt to the change of inhibitor concentration, the acclimation of yeast in each inhibitor concentration was repeated more than four times.

The inoculum was prepared by growing S. cerevisiae LF2 in a YPD medium containing 20 g/L glucose, 20 g/L peptone, and 10 g/L yeast extract at 30°C, 200 rpm for 24 h in a thermostat air bath shaker. The inoculum with an OD of 2.0 at 600 nm was used as seed culture, and the inoculation volume was 5% (v/v).

The semi-SSF experiment was performed in a 50 ml sealed Erlenmeyer flask by a rubber plug with a needle. First, the pretreated corn fiber slurries were adjusted to pH 5.0 with 10 M NaOH, and the pre-hydrolysis was conducted at 20% (w/v) solid consistency, 48°C, pH 5.0 for 12 h using cellulase of 10 FPU/g (DM). After pre-hydrolysis, the reaction system temperature was adjusted to 30°C, and 5% (v/v) of seed was inoculated to the reaction system. Fermentation was carried out in a thermostat air bath shaker with a rotation rate of 200 rpm. The samples were taken periodically to determine the concentrations of glucose, xylose, and ethanol.

The FPA and the activities of xylanase, β-xylosidase (pNPXase), and α-arabinofuranosidase (pNPAase) of culture supernatants were measured using a Whatman No.1 filter paper, insoluble xylan (beech), p-nitrophenyl-β-D-xylopyranoside, and p-nitrophenyl-α-L-arabinofuranoside as the substrates, respectively, according to the methods presented by Gao et al. (2021). One unit (U) of enzyme activity was defined as the amount of enzyme that liberates 1 μmol of reducing sugars (for FPA and xylanase) equivalent or p-nitrophenol (for pNPAase and pNPXase) per minute under the assay conditions.

The samples collected in the pretreatment, enzymatic hydrolysis, and fermentation stages, respectively, were centrifuged at 10,000 rpm for 10 min. The supernatants were used for analyzing the contents of glucose, xylose, formic Acid, acetic acid, furfural, HMF, and ethanol using HPLC (LC-20AT, refractive index detector RID-20A, Shimadzu, Kyoto, Japan) With an Aminex HPX-87H column (Bio-Rad, Hercules, CA, United States) at 60°C using a mobile phase of 5 mM H₂SO₄ at a rate of 0.5 ml/min (Sluiter et al., 2008b). For the samples from the enzymatic hydrolysis and fermentation stage, the supernatants were first boiled for 10 min for inactivating enzyme. The sugar yields (%) were calculated according to the following formula:

where C_HPLC is the concentration of sugar as determined by HPLC, mg/ml; V is the volume of the reaction system, ml; CF (conversion factor) is 0.9 for C-6 sugars and 0.88 for C-5 sugars; W is the weight of the substrate in the reaction system, g; and Content_Sugar is the percentage of the polymeric sugar in corn fiber.

Table 1 provides the chemical compositions of corn fiber. The corn fiber contains starch, cellulose, and hemicellulose with a total carbohydrate content of 71.22 ± 1.27%. The major components were hemicellulosic sugars (xylose, arabinose, and galactose), and their total contents reached about 40%. These data were consistent with the previously reported results (Kim et al., 2008; Eylen et al., 2011). Compared with the chemical components of corn stover and corn cob (Eylen et al., 2011), the corn fiber had similar total glucan content to corn stover (30.6%) and corn cob (34.8%) and similar xylan content to corn stover (19%). But unlike corn cob and corn stover, a considerable part of the glucose in corn fiber was derived from starch, not just cellulose. In contrast, the contents of other hemicellulose sugars in corn fiber were significantly higher than that in corn stover and corn cob, which means that the degradation of corn fiber may require a more complex enzyme system with more abundant hemicellulase components.

From batch kinetics studies, it was found that the factors affecting acid pretreatment included the type of biomass, the type of acid, acid concentration, reaction time, and reaction temperature (Rahman et al., 2006). Among them, the operating conditions, including reaction time, temperature, and acid concentration, would significantly affect the efficiency of dilute acid pretreatment (López-Arenas et al., 2010). In this study, RSM incorporating variables of temperature, time, and acid concentration was used to optimize the pretreatment process parameters for maximizing the enzymatic conversion of carbohydrate in corn fiber to monosaccharides and minimizing the formation of inhibitors. The complete list of runs and responses is provided in Table 2. Five experimental runs were replicated at the center point of the design to confirm any differences in the estimation procedure as a measure of precision. Table 2 shows the yields of formic acid, acetic acid, and furfural after pretreatment and the yields of glucose, xylose, and arabinose after pretreatment and following enzymatic hydrolysis. Figure 1 shows the response surface plots of glucose, xylose, and arabinose yields under different pretreatment conditions. ANOVA of the models for glucose, xylose, arabinose, furfural, formic acid, and acetic acid is given in Table 3. The regression equations of different products were as follows, respectively.

Glucose = + 334.95 – 0.49 × Temp. + 4.69 × Time + 8.45 × Acid conc. – 9.53 × Temp. × Time – 6.38 × Temp. × Acid conc. – 2.60 × Time × Acid conc.

Xylose = + 132.21 + 19.61 × Temp. + 14.86 × Time + 19.06 × Acid conc. – 0.64 × Temp. × Time + 1.19 × Temp. × Acid conc. – 0.66 × Time × Acid conc. + 1.56 × Temp.2 – 7.19 × Time 2 – 3.91 × Acid conc.2

Arabinose = + 110.82 + 5.20 × Temp. + 4.97 × Time + 6.71 × Acid conc. – 4.23 × Temp. × Time −3.00 × Temp. × Acid conc. – 2.97 × Time × Acid conc. – 0.62 × Temp.2 – 4.26 × Time 2 – 2.29 × Acid conc.2

Furfural = + 3.07 + 0.57 × Temp. + 0.19 × Time – 0.14 × Acid conc. – 0.24 × Temp. × Time – 7.500E-003 × Temp. × Acid conc. + 0.018 × Time × Acid conc. + 0.061 × Temp.2 – 0.014 × Time 2 – 0.39 × Acid conc.2

Formic acid = + 5.30 + 0.33 × Temp. + 0.14 × Time + 0.23 × Acid conc. + 0.16 × Temp. × Time + 0.17 × Temp. × Acid conc. + 0.047 × Time × Acid conc.

Acetic acid = + 8.25 + 2.90 × Temp. + 1.69 × Time + 2.65 × Acid conc. + 0.73 × Temp. × Time + 0.93 × Temp. × Acid conc. + 1.10 × Time × Acid conc.

where sugars and inhibitors appear as yield (mg/g) and the variables are Temp. (temperature, °C), Time (in min), and Acid conc. (acid concentration, %, w/v).

It is shown in Figure 1 and p-values in ANOVA that the yield of glucose obtained from the enzymatic hydrolysis of pretreated corn fiber was sensitive to acid concentration (p = 0.0372 ≤ 0.05) but not to pretreatment time (p = 0.2119 > 0.05) and temperature (p = 0.8922 > 0.05), but the yield of xylose was significantly very sensitive to reaction time (p < 0.0001), temperature (p < 0.0001), and acid concentration (p < 0.0001). Initially, the increase in pretreatment temperature is in favor of the conversion of xylan to xylose in subsequent enzymatic hydrolysis, but a high pretreatment temperature led to the dehydration reaction of xylose accelerated, thus decreased xylose production but increased furfural content (Table 2), which is similar to the results reported in some literature (Kabel et al., 2007; Noureddini and Byun, 2010).

In contrast, it was found that the yields of glucose and xylose (based on the original content in corn fiber) were 10.5 and 38.3%, respectively, in the pretreatment stage, and the yields of glucose and xylose were 95.5 and 72.4% after enzymatic hydrolysis, indicating that xylan was more dissolved out in the pretreatment stage. Avci et al. (2013) also reported the difference in production behavior between glucose and xylose. They found that after being treated with dilute sulfuric acid a considerable part of xylan was directly hydrolyzed into xylose, while glucose was mostly produced by enzymatic hydrolysis. This is similar to our results. In the study, maximum xylose of 171 mg/g corn fiber was obtained by pretreatment at 125°C for 35 min using 0.5% (m/v) H₂SO₄. Similar to xylose, as shown in Figure 1I, the yield of arabinose was also sensitive to pretreatment temperature (p = 0.0011 ≤ 0.01), time (p = 0.0014), and acid concentration (p = 0.0002).

Overall, for three main sugars in corn fiber (glucose, xylose, and arabinose), the highest yield of total sugars was produced by pretreatment at 125°C for 35 min using 0.5% (w/v) H₂SO₄ and enzyme hydrolysis, including glucose of 340 mg, xylose of 171 mg, and arabinose of 117 mg, respectively, on the basis of per gram corn fiber. This amount was approximately 85.4% of the three sugars present in the corn fiber.

Furfural and HMF were generated from xylose and glucose in the pretreatment process through a further dehydration reaction, respectively (Jönsson and Martín, 2016). It was found that furfural formation was also most sensitive to pretreatment temperature (p < 0.0001), but not to acid concentration (p = 0.949 > 0.05), maybe because sulfuric acid only acts as a catalyst to provide an acidic environment. The highest furfural content of 3.93 mg/g corn fiber was found after pretreatment at 125°C for 10 min with 0.35% sulfuric acid (Table 2). Extending pretreatment time from 10 min to 60 min under conditions of 0.35% sulfuric acid and 125°C, furfural content in pretreatment liquid decreased, maybe due to further decomposition of furfural. It was also found that the yield of formic acid, a degradation product of furfural (Yang et al., 2012), increased (Table 2). In addition, Yemiş and Mazza (2011) also reported that the too-long reaction time could cause the condensation reaction of furfural to form other products. However, no HMF was detected in any hydrolysate from the acid pretreatment process, possibly because little glucose was degraded during the acid pretreatment under pretreatment conditions used in the paper. Acetic acid was also detected in hydrolysates after the acidic pretreatment (Table 2), which comes from the shedding of acetyl groups in hemicellulose (Ibbett et al., 2011), and its content was also sensitive to reaction time (p < 0.0001), temperature (p < 0.0001), and acid concentration (p < 0.0001). Longer reaction time, higher reaction temperature, and acid concentration can promote acetic acid formation.

It was found that, although the pretreatment can obtain the highest amounts of sugars at 125°C for 35 min using 0.5% H₂SO₄, there was also the higher production of formic acid (5.97 mg/g corn fiber), acetic acid (14.5 mg/g corn fiber), and furfural (3.11 mg/g corn fiber), which may result in the inhibitory effect on microbial growth in subsequent ethanol fermentation. To obtain the maximum sugars while minimizing the formation of inhibitory compounds as much as possible, based on the RSM analysis (Figure 1 and Tables 2, 3), it was determined that the suitable pretreatment conditions for the corn fiber were the acid concentration of 0.5%, reaction temperature of 105°C, and reaction time of 43 min. Under this condition, the yield of glucose, xylose, and arabinose reached 354 mg/g corn fiber, 133 mg/g corn fiber, and 114 mg/g corn fiber, respectively, which is equivalent to 81.8% of the total amount of the three sugars present in the corn fiber. Although the yield (81.8%) was slightly lower than the highest sugar yield (85.4%) obtained at 125°C for 35 min using 0.5% H₂SO₄, the amounts of inhibitors produced in the mild pretreatment significantly decreased, for example, decreased by 25.7% for furfural, 45.6% for acetic acid, and 14.2% for formic acid, respectively.