Efficient Production of Succinic Acid from Sugarcane Bagasse Hydrolysate by Actinobacillus succinogenes GXAS137

Abstract

Sugarcane bagasse (SCB) is an abundant agricultural waste, rich in cellulose and hemicellulose, that could be used as an ideal raw material for succinic acid (SA) production. A two-step chemical pretreatment, involving alkali extraction and alkaline hydrogen peroxide treatment, was utilized to treat SCB, followed by multi-enzyme hydrolysis to obtain a reducing sugar hydrolysate mainly composed of glucose and xylose. Optimization of the multi-enzyme hydrolysis of pretreated SCB resulted in a final reducing sugar concentration of 78.34 g/L. In order to enhance the bioconversion of SCB to SA and to reduce the production costs, the initial reducing sugar concentration, nitrogen source, and MgCO₃ content were further optimized. The results demonstrated that the inexpensive corn steep liquor powder (CSLP) could be utilized as an alternative nitrogen source to yeast extract for the production of SA; and the optimal concentrations of initial reducing sugar, CSLP, and MgCO₃ were 70 g/L, 18 g/L, and 60 g/L, respectively. When fed-batch fermentation was conducted in a 2 L stirred bioreactor, approximately 72.9 g/L of SA was produced, with a yield of 83.2% and a productivity of 1.40 g/L/h. The high SA concentration, yield, and productivity achieved in this study demonstrate the potential of SCB, an agricultural waste, as a viable alternative substrate for Actinobacillus succinogenes GXAS137 to produce SA. This lays a solid foundation for the resource utilization of agricultural waste and cost-effective industrial-scale production of SA in the future.

1. Introduction

Succinic acid (SA) is an important platform compound with wide applications in food, medicine, the chemical industry, spices, and other fields [1]. In 2004 and 2010, the U.S. Department of Energy identified SA as one of the top five promising bio-based platform compounds for the future [2]. In 2015, the global market size of SA was USD 157.2 million, which was equivalent to 58,500 tons. It is expected to reach USD 1.8 billion by 2025, with a compound annual growth rate of 27.4% [3]. SA is traditionally produced via chemical synthesis, but this process heavily relies on non-renewable petroleum substrates, resulting in extremely volatile prices and serious environmental pollution [4]. With the depletion of oil reserves and increasing concerns about environmental issues, there has been growing attention on SA production from renewable biomass resources in recent years. Compared to petrochemical processes, microbial fermentation for SA production is cost-competitive due to the utilization of inexpensive renewable raw materials as substrates, environmentally friendly production processes, and higher yields and productivity [5]. Among various SA-producing microorganisms, A. succinogenes is a facultative anaerobic Gram-negative bacterium with excellent osmotic tolerance; moreover, it has the ability to metabolize diverse carbon sources, including glucose, xylose, arabinose, fructose, lactose, disaccharides, and mannose, under anaerobic conditions, thereby possessing potential advantages in the conversion of lignocellulosic biomass to SA [4].

Currently, the large-scale production of bio-based SA primarily relies on starch derived from grains (such as wheat and corn) or refined sugar (glucose) as substrates [6]. The resulting higher production cost (USD 2.86 to USD 3.00 per kg) hinders its full competitiveness against petrochemical-derived chemicals (USD 2.40 to USD 2.60 per kg) [7]. Therefore, in order to circumvent competition with food demand and reduce production costs, it is essential to utilize inexpensive non-food biomass for SA production. Lignocellulosic biomass, a highly abundant and renewable resource, can produce significant amounts of hexoses and pentoses through hydrolysis [8], making it a promising future feedstock for SA production. The utilization of it as a substrate has been evaluated to potentially reduce the production cost of SA by over 50% [9]. Sugarcane, a vital sugar-producing crop extensively cultivated in tropical and subtropical regions, yields sugarcane bagasse (SCB) as its primary by-product, with an estimated global production of 540 million tons annually [3]. SCB consists of approximately 32–45% cellulose, 20–32% hemicellulose, and 17–32% lignin; and the abundant carbohydrate content and low cost render it a promising substrate widely used for the production of biofuels, enzymes, sugars, and other high-value-added products [10]. However, the presence of lignin in the cell wall confers structural refractoriness, impeding the degradation of its structural polysaccharides by microorganisms and enzymes [11], and therefore pretreatment is required. A range of pretreatment methods, encompassing physical, chemical, physicochemical, and biological approaches and combinations thereof, have been developed to enhance the digestibility of lignocellulosic biomass; however, each method functions differently in decomposing lignocellulosic biomass, resulting in distinct efficiency in subsequent conversion processes [12]. SCB usually exhibits a high SiO₂ content [13]. According to the literature, the enzymatic hydrolysis of carbohydrates was interfered with during lignocellulosic ethanol production due to the interaction between SiO₂ and cellulose [14]. In this context, the removal of SiO₂ and lignin in the pretreatment stage holds significant importance for enhancing the bioconversion of SCB to SA. Among various pretreatments, the two-step chemical pretreatment that involves alkali extraction and alkaline hydrogen peroxide (AHP) treatment is acknowledged as an efficient approach for SCB due to the following advantages: Firstly, the acid-insoluble SiO₂ in SCB can be converted into soluble silicate through alkali extraction, mitigating the adverse effects of SiO₂ on the subsequent enzymatic hydrolysis [14]. Secondly, alkali extraction removes some readily extractable lignin, acetyl groups, and various uronic acid substitutes on hemicellulose and induces cellulose swelling, thereby significantly augmenting the surface area of the fibers [15]. Thirdly, AHP treatment selectively oxidizes and eliminates the more recalcitrant lignin within the cell wall, significantly enhancing biomass digestibility [16]. Finally, this pretreatment method generates little or no furfural inhibitors compared to others [17]. The hydrolysis of polysaccharides in pretreated lignocellulosic biomass into fermentable sugars constitutes a crucial step for A. succinogenes to produce SA utilizing lignocellulosic materials. The polysaccharides obtained after pretreatment should be subjected to acid or enzymatic hydrolysis to yield fermentable sugars [18]. Enzymatic hydrolysis is widely regarded as the most efficient approach for liberating monosaccharides from lignocellulosic materials, owing to its gentle reaction conditions and absence of glucose degradation by-products. Cellulases, comprising endoglucanase, exoglucanase, and β-glucosidase, are the predominant enzymes employed for hydrolyzing pretreated lignocellulosic materials to glucose [19]. Furthermore, numerous studies have demonstrated that the synergistic action of cellulase with other enzymes can significantly increase the rate of enzymatic hydrolysis. Tabka et al. investigated the effect of adding auxiliary enzymes, including xylanase, ferulic acid esterase, and laccase, on the enzymatic hydrolysis of pretreated wheat straw. The results showed that a synergistic effect of cellulase, xylanase, and ferulic acid esterase resulted in a glucose yield accounting for 81% of the maximum glucose recovery rate at 50 °C [20]. By optimizing the proportion of cellulase, xylanase, β-glucanase, and pectinase in the multi-enzyme mixture, Chen et al. achieved a cellulose hydrolysis efficiency of 88.5% from NaOH-pretreated SCB [21].

In the present study, we employed SCB as a substrate for SA production by A. succinogenes GXAS137. Initially, chemical pretreatment of SCB was conducted, followed by optimization of the multi-enzyme hydrolysis process using an orthogonal design experiment. Simultaneously, single-factor experiments were carried out to optimize fermentation conditions, including initial reducing sugar concentration, nitrogen source, and MgCO₃ concentration. Ultimately, lab-scale fermentation was performed in a 2 L stirred bioreactor to validate the economic feasibility of efficient SA production utilizing SCB hydrolysate by A. succinogenes GXAS137.

2. Materials and Methods

2.1. Materials and Chemicals

SCB was provided by Nanning Mingyang Sugar Co., Ltd. (Nanning, China). Fresh SCB was washed with tap water, dried at 60 °C using a convection oven (UN130, GreenPrima, London, UK) until reaching a constant weight, ground into particles smaller than 1 mm in size, and stored in a dry environment for future use. Cellulase (Celluclast 1.5 L, 700 EGU/g), xylanase (Shearzyme 500 L, 250 FXU-S/g), and pectinase (Pectinex BE XXL, 13,600 PECTU/g) were purchased from Novozymes (Wuxi, China); yeast extract (YE) and tryptone were purchased from OXOID (Cambridge, UK); beef extra (BE) was purchased from Sangon Biotech Co., Ltd. (Shanghai, China); corn steep liquor powder (CSLP) was purchased from Beijing Hongrun Baoshun Technology Co., Ltd. (Beijing, China); high purity CO2 gas (>99%) was obtained from Guangxi Ruida Chemical Technology Co., Ltd. (Nanning, China); MgCO₃ was purchased from Chron Chemicals Co., Ltd. (Chengdu, China). The remaining chemicals, including Glucose, H₂SO₄, (NH₄)₂SO₄, Urea, NaOH, 30% hydrogen peroxide solution, NaH₂PO₄·2H₂O, K₂HPO₄·3H₂O, NaHCO₃, KH₂PO₄, MgCl₂·6H₂O, CaCl₂·2H₂O, Na₂S, and NaCl were all analytical grade and obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

2.2. Pretreatment

SCB was pretreated using alkali (NaOH), AHP, and their combination. Briefly, for NaOH pretreatment, milled SCB was immersed in a 1% (w/v) NaOH aqueous solution at a liquid-to-solid ratio of 20:1 (v/w), followed by reaction in an autoclave at 115 °C for 1 h. For AHP pretreatment, ground SCB was soaked in a 2% (v/v) H₂O₂ aqueous solution with a liquid-solid ratio of 20:1 (v/w), while adjusting the pH to 11.5 using 6 M NaOH, and then reacted at 60 °C for 18 h. For the two-step pretreatment (NaOH-AHP) consisting of a NaOH extraction step and an AHP treatment step, the SCB was initially subjected to extraction under identical conditions as the NaOH pretreatment, followed by filtration and rinsing of the remaining insoluble solids with excess tap water to eliminate all soluble substances; the resulting insoluble solids were then treated with AHP in a second step, using the same conditions as the AHP pretreatment. Subsequently, solid residues obtained from different pretreated SCBs were filtered using a sand core funnel (G3) and washed with tap water until reaching neutral pH. The washed residues were dried at 60 °C until a constant weight was achieved and stored for future use.

2.3. Enzymatic Hydrolysis

The pretreated SCB was immersed in a citrate buffer solution (pH 4.8) with a solid-to-liquid ratio of 1:10 (w/v), and then a mixture of enzyme solutions was added. After mixing, the mixture was transferred into a 100-mL shaking flask with a liquid volume of 30 mL, which was placed in a constant-temperature water bath shaker. The pretreated SCB was hydrolyzed for 30 h at a temperature of 50 °C and a speed of 120 rpm. The solid residue was removed by centrifugation, and the supernatant obtained was used for the subsequent SA production.

2.4. Orthogonal Design

Orthogonal design is one of the most effective and time-saving methods in multivariate studies for identifying the most significant factors affecting production in research [22]. To optimize the dosage of each enzyme and enhance the efficiency of enzymatic hydrolysis for pretreated SCB, an orthogonal design L9(3⁴) based on SPSS Statistics 26.0 (IBM, Armonk, NY, USA) was adopted to analyze the effects of cellulase (A), xylanase (B), and pectinase (C) on the enzymatic hydrolysis of pretreated SCB. Each factor was varied at three levels, resulting in a total of nine experiments. The experiment order was randomized to minimize errors and biases.

2.5. Microorganism and Culture Conditions

A. succinogenes GXAS137, which was isolated from bovine rumen by our laboratory using the method reported by Guettler et al. [23], has been stored in the Chinese Typical Culture Preservation Center (CCTCC M 2016396). This strain demonstrates a remarkable capacity for SA production [24]. The cultures were grown in 100 mL sealed anaerobic bottles with 50 mL of medium. The inoculum culture medium contained 20 g/L glucose, 10 g/L YE, 10 g/L CSLP, 9.0 g/L NaH₂PO₄·2H₂O, 15.5 g/L K₂HPO₄·3H₂O, and 2 g/L NaHCO₃, at neutral pH. Glucose and other components were sterilized individually at 115 °C for 20 min. Glucose was added to the medium after cooling under sterile conditions. All seed cultures were incubated at 37 °C for 12 h in a constant-temperature incubator (PYL-70, Labotery, Tianjin, China).

2.6. Fermentation in Anaerobic Bottles

Cells in the exponential growth phase were inoculated into a 250 mL sealed anaerobic bottle containing 150 mL of culture medium for anaerobic fermentation. The basic fermentation medium contained 10 g/L YE, 4 g/L NaHCO₃, 3 g/L KH₂PO₄, 0.3 g/L MgCl₂·6H₂O, 0.3 g/L CaCl₂·2H₂O, and 1 g/L NaCl. MgCO₃ was added to the fermentation culture medium as a pH buffer and an indirect source of CO₂, which was determined based on the total sugar concentration in the hydrolysate. In all cases, about 1 g of MgCO₃ per gram of total sugar content in the hydrolysate was added, except for specific cases. The SCB hydrolysate was sterilized and then added to the medium under aseptic conditions to achieve the desired sugar concentrations. Prior to inoculation and fermentation, CO₂ gas and Na₂S at a final concentration of 0.2 g/L were introduced to the sterile medium to establish a strictly anaerobic environment. All media were autoclaved at 115 °C for 20 min before use. The seed culture was inoculated into a fermentation medium at a ratio of 8% (v/v), followed by fermentation under controlled conditions of 37 °C and 100 rpm.

2.7. Fermentation in a Stirred Bioreactor

To further assess the impact of utilizing SCB hydrolysate as a carbon source on SA production and to achieve a higher SA titer compared to anaerobic bottle fermentation, batch fermentation experiments were conducted in a 2 L stirred bioreactor (Bailun in Technology Co., Ltd. Shanghai, China), with an initial medium volume of 1.4 L. The fermentation temperature was set at 37 °C, the agitation speed was maintained at 150 rpm, and the flow rate of CO₂ gas was kept at 0.2 L/min. The fermentation medium in the bioreactor was identical to that used in anaerobic bottles, and pH during fermentation was controlled using MgCO₃. The biomass, residual sugar, and organic acid content in the fermented medium were measured by taking samples at regular intervals.

Fed-batch fermentation was carried out under the same conditions as the batch fermentation. When the residual total reducing sugar concentration dropped below 20 g/L during fermentation, condensed SCB hydrolysate at a concentration of 150 g/L was introduced into the bioreactor using a peristaltic pump to maintain the reducing sugar concentration in the medium between 15 and 20 g/L.

2.8. Analytical methods

The main components of SCB were measured by the method published by the National Renewable Energy Laboratory (NREL) of the United States [25]. The biomass in the fermentation samples was determined by measuring the optical density at 600 nm (OD₆₀₀) using a DU 800 UV/VIS Spectrophotometer (Beckman, Fullerton, CA, USA). Sugars and organic acids were analyzed with an UltiMate 3000 Standard HPLC System (Dionex, Sunnyvale, CA, USA) equipped with an ROA-Organic Acid H⁺ (8%) column (Phenomenex, Torrance, CA, USA). A mobile phase of 2.5 mM H₂SO₄ solution at a flow rate of 0.6 mL/min was selected, and the column temperature was maintained at 50 °C. The organic acids were analyzed with a tunable UV detector (Dionex, Sunnyvale, CA, USA) set at 210 nm, and monosaccharides were detected using a Shodex RI-101 refractive index detector (Showa Denko KK, Kawasaki, Japan) at 50 °C. Furfural and 5-hydroxymethyl-2-furaldehyde (HMF) were also detected by HPLC.

2.9. Statistical Design

All experiments were repeated at least three times. Data are presented as the mean ± standard deviation (SD). Significant differences between means were tested using one-way analysis of variance (ANOVA), followed by least significant difference (LSD) tests in IBM SPSS Statistics version 26.0 (Armonk, NY, USA) at a significance level of p < 0.05. The SA yield, productivity, and reducing sugar yield were determined by employing the following equations:

$$\mathrm{SA} \mathrm{yield} (\%)=\frac{\mathrm{SA} \mathrm{mass} \left(\mathrm{g}\right) \times 100\% }{ \left(\mathrm{glucose}+\mathrm{xylose}\right)\mathrm{mass} \mathrm{consumed} \left(\mathrm{g}\right)}$$

(1)

$$\mathrm{SA} \mathrm{productivity} (\mathrm{g}/\mathrm{L}/\mathrm{h})=\frac{\mathrm{SA} \mathrm{concentration} (\mathrm{g}/\mathrm{L})}{ \mathrm{fermentation} \mathrm{time} \left(\mathrm{h}\right)}$$

(2)

$$\mathrm{Cellulose} \mathrm{hydrolysis} \mathrm{rate} (\%)=\frac{\mathrm{glucose} \mathrm{mass} \left(\mathrm{g}\right) \times 0.9 \times 100\%}{\mathrm{cellulose} \mathrm{mass} \mathrm{in} \mathrm{residue} \left(\mathrm{g}\right)}$$

(3)

$$\mathrm{Hemicellulose} \mathrm{hydrolysis} \mathrm{rate} (\%)=\frac{\mathrm{xylose} \mathrm{mass} \left(\mathrm{g}\right) \times 0.88 \times 100\%}{\mathrm{hemicellulose} \mathrm{mass} \mathrm{in} \mathrm{residue} \left(\mathrm{g}\right)}$$

(4)

$$\mathrm{total} \mathrm{reducing} \mathrm{sugar} \mathrm{yield} (\%)=\frac{(\mathrm{glucose} \mathrm{mass} (\mathrm{g}) \times 0.9+\mathrm{xylose} \mathrm{mass} \left(\mathrm{g}\right) \times 0.88) \times 100\%}{(\mathrm{hemicellulose}+\mathrm{cellulose}) \mathrm{mass} \mathrm{in} \mathrm{residue} \left(\mathrm{g}\right)}$$

(5)

where 0.9 is the conversion factor from cellulose to glucose and 0.88 is the conversion factor from hemicellulose to xylose. Cellulose hydrolysis rate (%)

3. Results and Discussion

3.1. Composition Analysis of SCB

The composition of SCB was determined using the NREL method [25]. The results are shown in Figure 1. The analysis revealed that cellulose accounted for approximately 41.3%, while hemicellulose constituted around 24.47% of the original SCB. Cellulose primarily consists of homopolysaccharide chains linked by β-(1,4)-glycosidic bonds, comprising anhydrous glucose units, whereas hemicellulose mainly comprises xylose, galactose, and glucose [3]. Cellulose and hemicellulose could be hydrolyzed by enzymes to obtain reducing sugars for SA production by A. succinogenes. SCB is considered a promising substrate for the production of SA because of its low cost and high content of polysaccharides (cellulose and hemicellulose). However, due to the recalcitrant cell wall structure, most microorganisms cannot directly utilize the polysaccharides within it; therefore, an efficient pretreatment method should be developed.

3.2. Effects of Different Pretreatment Methods on the Composition and Enzymatic Hydrolysis of SCB

In the process of the biotransformation of lignocellulosic biomass, pretreatment is commonly required to remove lignin, reduce cellulose crystallinity, or increase material porosity in order to improve the accessibility of cellulose to enzymes [26]. In this study, NaOH, AHP, and their combined methods were employed for the pretreatment of SCB. The evaluation of these methods was based on the recovery rates of cellulose and hemicellulose, the lignin removal rate, and the concentration and yield of reducing sugars. The composition of SCB after different pretreatment methods and their enzymatic hydrolysis efficiency are summarized in

Table 1. The main components of SCB and the efficiency of enzymatic hydrolysis with different pretreatment methods.

			Pretreatment
		Raw	NaOH	AHP	NaOH-AHP
Solid remain (%)		—	56.03 ± 0.21 b	63.67 ± 0.33 a	46.97 ± 0.18 c
Cellulose	Content (%)	41.32 ± 0.15 d	57.44 ± 0.63 b	54.23 ± 0.09 c	64.45 ± 0.22 a
	Recovery (%)	—	77.89 ± 0.41 b	83.56 ± 0.23 a	73.26 ± 0.26 c
Hemicellulose	Content (%)	24.47 ± 0.11 b	26.06 ± 0.03 a	23.53 ± 0.43 b	22.53 ± 0.22 c
	Recovery (%)	—	59.68 ± 0.53 b	61.22 ± 0.62 a	43.24 ± 0.18 c
Lignin	Content (%)	26.42 ± 0.08 a	11.76 ± 0.12 c	15.36 ± 0.46 b	9.89 ± 0.37 d
	Removal (%)	—	75.08 ± 0.76 b	62.99 ± 0.59 c	82.42 ± 0.24 a
FU and HMF (mg/L)		None	None	None	None
Glucose (g/L)		0.51 ± 0.08 d	41.65 ± 0.32 b	37.63 ± 0.78 c	51.31 ± 0.32 a
Cellulose hydrolysis rate (%)		1.16 ± 0.02 d	67.87 ± 0.48 b	64.95 ± 0.61 c	74.52 ± 0.38 a
Xylose (g/L)		0.13 ± 0.01 d	20.14 ± 0.06 a	13.66 ± 0.62 c	17.12 ± 0.15 b
Hemicellulose hydrolysis rate (%)		0.49 ± 0.08 c	70.73 ± 0.27 a	53.13 ± 0.75 b	69.54 ± 0.26 a
Total reducing sugar (g/L)		0.64 ± 0.06 d	61.79 ± 0.35 b	51.29 ± 0.82 c	68.43 ± 1.02 a
Total reducing sugar yield (%)		0.9 ± 0.00 d	68.75 ± 0.28 b	61.29 ± 0.75 c	73.20 ± 0.22 a

NaOH pretreatment was conducted at 115 °C for 1 h using a 1% (w/v) NaOH solution; alkali hydrogen peroxide (AHP) pretreatment was conducted at 60 °C for 18 h using a 2% H₂O₂ (v/v, pH 11.5) solution; the two-step chemical pretreatment (NaOH-AHP) consists of NaOH extraction and AHP treatment. Each value is an average of three parallel replicates and is represented as the mean ± SD. In the same row, values with different small letter superscripts indicate significant difference (p < 0.05). FU, furfural; HMF, 5-hydroxymethyl-2-furaldehyde. SCB, sugarcane bagasse.

.

It was shown that all pretreatment methods significantly altered the composition of SCB, particularly resulting in a substantial increase in polysaccharide content and a notable decrease in lignin content. After AHP pretreatment, the recovery rates of cellulose and hemicellulose were 83.56% and 61.22%, respectively. However, about 37% of the lignin remained in the residue, where lignin still accounted for 15.36%. After one-step NaOH pretreatment, most of the lignin was degraded and removed, while the proportion of hemicellulose even increased from 24.47% to 26.06%. When adding another step of AHP treatment, the cellulose content in the residue increased from 57.44% to 64.45%, while the lignin content further decreased from 11.76% to 9.89%. Combined pretreatments removed more lignin than one-step pretreatments, simultaneously resulting in high polysaccharide content in the residues. The NaOH-AHP combined pretreatment adopted in this work exerted a comparable effect on the composition of SCB as reported in a previous study, where NaOH and AHP were combined to pretreat SCB [27].

Multi-enzymatic hydrolysis was performed on different pretreated residues to obtain fermentable sugars, as shown in Table 1. A higher total reducing sugar concentration was obtained from NaOH-pretreated residue than AHP-pretreated residue. Furthermore, the total reducing sugar concentration obtained from combined NaOH-AHP-pretreated residue was also higher than that obtained from one-step pretreated residue (NaOH or AHP). This could be attributed to the fact that compared to AHP treatment, NaOH pretreatment resulted in a higher polysaccharide content but lower lignin content in the residues. Similarly, compared to SCB subjected to one-step pretreatment, SCB subjected to combined NaOH-AHP pretreatment also exhibited a higher polysaccharide content but lower lignin content. Lignin, as the primary physical barrier, not only impedes the hydrolysis of the digestible portion of the substrate but also diminishes the availability of free cellulolytic enzymes through non-productive adsorption, thereby reducing cellulolytic hydrolysis [26]. Consequently, the reducing sugar concentration exhibited a negative correlation with lignin content. The abundance of polysaccharides in pretreated residues often facilitates the achievement of a high content of reducing sugars through subsequent enzymatic hydrolysis [28]. In the present study, the enzymatic hydrolysis of pretreated residues involved the utilization of a mixture of enzyme solutions. This synergistic action of multiple enzymes can adequately break down cellulose and hemicellulose in residues with relatively low lignin content. Therefore, the reducing sugar concentration showed a positive correlation with the polysaccharide content. In conclusion, NaOH-AHP pretreatment was selected as the best pretreatment method in the present study.

3.3. Enzymatic Hydrolysis of Pretreated SCB Using Multiple Enzymes

Enzymatic reactions are known for their high specificity, efficiency, and mild catalytic conditions. Cellulase has been widely utilized in the hydrolysis of lignocellulosic materials to obtain fermentative sugars; furthermore, the utilization of cellulase in combination with other enzymes can significantly enhance the enzymatic hydrolysis rate [21]. Therefore, NaOH-AHP-treated SCB was hydrolyzed by a combination of enzymes to produce the reducing sugars required for SA fermentation. The concentrations of cellulase (A), xylanase (B), and pectinase (C) were optimized using an orthogonal design L9(3⁴) to maximize the concentration of reducing sugars. The results are shown in

Table 2. Orthogonal design, experimental results, and range analysis (R) for enzymatic hydrolysis of pretreated SCB.

Test No.	Factors (Levels)				Reducing Sugar Concentration (g/L)
	(A) Cellulase Concentration (% v/w-Biomass)	(B) Xylanase Concentration (% v/w-biomass)	(C) Pectinase Concentration (% v/w-biomass)	Null Column
1	10.00 (1)	2.00 (1)	10.00 (1)	(1)	60.96 ± 0.15
2	10.00 (1)	3.00 (2)	15.00 (2)	(2)	59.75 ± 0.06
3	10.00 (1)	4.00 (3)	20.00 (3)	(3)	58.05 ± 0.77
4	15.00 (2)	2.00 (1)	15.00 (2)	(3)	61.77 ± 0.93
5	15.00 (2)	3.00 (2)	20.00 (3)	(1)	67.87 ± 0.77
6	15.00 (2)	4.00 (3)	10.00 (1)	(2)	76.90 ± 0.75
7	20.00 (3)	2.00 (1)	20.00 (3)	(2)	70.47 ± 0.52
8	20.00 (3)	3.00 (2)	10.00 (1)	(3)	77.13 ± 0.31
9	20.00 (3)	4.00 (3)	15.00 (2)	(1)	75.10 ± 0.86
k1	59.59	64.40	71.66	67.98
k2	68.85	68.25	65.54	69.04
k3	74.23	70.02	65.46	65.65
R	14.65	5.62	6.20	3.39
Best level	A3	B3	C1		78.34 ± 0.87

Each value is an average of three parallel replicates and is represented as the mean ± SD. SCB, sugarcane bagasse.

.

As shown in the table, the R values of all three factors (A, B, and C) were greater than that of the null column (error column), indicating a significant impact on the concentration of reducing sugars from pretreated SCB by all three factors. The order of the three factors influencing the concentration of reducing sugars was A > C > B, and the optimal combination for enzymatic hydrolysis of pretreated SCB was A3B3C1, namely, 20% cellulase, 4% xylanase, and 10% pectinase. Three parallel experiments were conducted under the optimal conditions, resulting in a reducing sugar concentration of 78.34 g/L, with a glucose concentration of 58.21 g/L and a xylose concentration of 20.13 g/L. The glucose/xylose ratio was about 3:1. The obtained reducing sugar yield was 83.8%, which was 14.48% higher than before optimization. The synergistic action of multiple enzymes can improve the sensitivity of sugar-binding bonds to enzymatic hydrolysis, thereby potentially explaining the increased hydrolysis rate of total celluloses in pretreated SCB. Hernández-Salas et al. optimized an enzyme formulation containing Celluclast, Novozyme, and Viscozyme L. The concentration of reducing sugars obtained by alkali-enzyme hydrolysis of SCB and agave bagasse was higher than that obtained by acid hydrolysis [29]. In the present study, the combined utilization of cellulase, xylanase, and pectinase for enzymatic hydrolysis of NaOH-AHP-pretreated SCB drastically enhanced the efficiency of enzymatic hydrolysis.

3.4. Effect of Initial Reducing Sugar Concentration on SA Production

Many studies have demonstrated that the initial sugar concentration can affect cell growth and metabolite synthesis [4,21]. Therefore, based on the basic fermentation medium, fermentation was conducted in anaerobic bottles using SCB hydrolysate with the initial reducing sugar concentration ranging from 40 to 90 g/L. As shown in

Table 3. Effects of different initial reducing sugar concentrations in SCB hydrolysate on SA production.

Initial Reducing Sugar (g/L)	SA (g/L)	AA (g/L)	Residual Reducing Sugar (g/L)	SA Yield (%)
40	25.28 ± 1.93 d	13.78 ± 1.16 c	6.17 ± 0.12 f	74.72 ± 0.32 a
50	30.90 ± 2.74 c	16.85 ± 1.68 b	8.96 ± 0.11 e	75.29 ± 0.55 a
60	35.58 ± 1.91 b	18.41 ± 0.33 ab	11.94 ± 0.74 d	74.03 ± 0.46 a
70	40.56 ± 0.36 a	19.09 ± 0.57 ab	17.11 ± 0.54 c	76.70 ± 0.27 a
80	38.70 ± 0.95 ab	19.54 ± 0.31 a	28.37 ± 0.42 b	74.95 ± 0.13 a
90	38.39 ± 0.61 ab	18.98 ± 0.19 ab	36.15 ± 1.31 a	71.29 ± 0.26 a

Each value is an average of three parallel replicates and is represented as the mean ± SD. In the same column, values with different small letter superscripts indicate significant difference (p < 0.05). SCB, sugarcane bagasse; SA, succinic acid; AA, acetic acid.

, reducing sugars were not completely consumed, with acetic acid (AA) being the main by-product. The SA concentration significantly increased (p < 0.05), with the initial reducing sugar concentration ranging from 40 to 70 g/L. When the initial sugar concentration was 70 g/L, the maximum SA concentration obtained was 40.56 g/L, with a corresponding yield of 76.70%. When the initial reducing sugar concentration exceeded 70 g/L, the SA concentration and yield decreased, while the residual reducing sugar content increased significantly. Zheng et al. reported the effects of the initial sugar concentration on cell growth and metabolite synthesis in A. succinogenes CGMCC159; when the initial sugar concentration was not less than 80 g/L, the residual sugar concentration increased, while the biomass, SA concentration, and yield decreased [30]. Shen et al. also reported the inhibitory effects of high total sugar concentrations in xylose mother liquor on SA production by A. succinogenes GXAS137; with an increase in the initial total sugar concentration in xylose mother liquor from 73.3 to 80.6 g/L, there was a corresponding decrease in SA concentration from 42.54 to 38.45 g/L, resulting in a decline in yield from 58.04% to 47.69% [7]. Similarly, the effect of high initial reducing sugar concentrations on SA production was also observed in this study. No furan compounds (furfural and HMF) with toxic effects on cells were detected in the SCB hydrolysate used in this study, as shown in Table 1. Therefore, the decrease in SA concentration and yield could be attributed to the inhibition caused by a high initial sugar concentration.

3.5. Effect of Nitrogen Source on SA Production from SCB Hydrolysate

Nitrogen is an integral part of proteins, nucleic acids, and certain nitrogen-containing metabolites, and as such is essential for all living organisms [31]. Based on optimized carbon sources, YE, tryptone, BE, CSLP, (NH₄)₂SO₄, and urea were used as nitrogen sources, with corresponding nitrogen concentrations of 1.2 g/L. The effects of different nitrogen sources on SA production from SCB hydrolysate by A. succinogenes were investigated.

As shown in Figure 2, the SA concentration and the SA/AA ratio varied depending on the nitrogen source used. Even in the absence of nitrogen sources, a certain amount of SA (19.83 g/L) was produced, suggesting that SCB hydrolysate might contain some nitrogen sources capable of facilitating the growth and fermentation of A. succinogenes GXAS137. The SA yields with complex organic nitrogen sources (YE, tryptone, CSLP, and BE) were significantly better than those with inorganic nitrogen sources ((NH₄)₂SO₄, and urea) (p < 0.05). This might be due to the fact that complex organic nitrogen sources could provide biotin, niacin, methionine, and other growth factors required for the growth of A. succinogenes [32]. The highest SA concentration (42.03 g/L) was obtained from YE. However, the use of expensive YE is not cost-effective for large-scale SA production. CSLP, a by-product derived from corn starch processing, can be used as a source of biotin and may replace peptone and YE in rich media [33]. Previous reports investigating fermentation by A. succinogenes [7,21] used CSLP instead of YE in a rich medium; these studies demonstrated the feasibility of replacing expensive YE with inexpensive CSLP. In the present study, the SA concentration (40.68 g/L) obtained from CSLP was only 3% lower than that obtained using YE, and ANOVA showed no significant difference (p > 0.05). Therefore, a more affordable CSLP was chosen as the nitrogen source for subsequent experiments.

The effect of different CSLP concentrations on SA production by A. succinogenes GXAS137 was further investigated, as shown in Figure 3. When using a low concentration (<18 g/L), the SA concentration significantly increased with the increase in CSLP concentration (p < 0.05). The highest SA concentration (41.83 g/L) was obtained when the CSLP concentration was 18 g/L. When the CSLP concentration exceeded 18 g/L, the SA yield decreased slightly, while AA accumulation continued to rise from 16.98 to 21.80 g/L, resulting in a corresponding decrease in the SA/AA ratio, which was also observed in our previous report [24]. This could be explained as follows: SA is synthesized via the C4 route, whereas AA, formic acid, and other by-products are synthesized via the C3 route in A. succinogenes; the synthesis of SA is strongly affected by the availability of NADH and ATP produced in the C3 route [34]. Furthermore, an excessive nitrogen supply will increase carbon metabolic flux toward biomass formation via the C3 route [35], which will lead to the reduction of carbon metabolic flux into the C4 route, thus reducing the yield of SA. In addition, 1 mole of AA produced in A. succinogenes is accompanied by 1 mole of ATP formed and the ATP can be used to supplement the energy required for cell growth under oxygen-limited conditions [36,37]. Therefore, excessive nitrogen supply indirectly promoted AA accumulation.

3.6. Effect of MgCO₃ Supplementation on SA Production

pH is a crucial factor affecting SA production by A. succinogenes. It can regulate enzyme activity related to SA production by adjusting the solubility and availability of CO₂ [38]. A pH close to neutral is beneficial for A. succinogenes growth. However, the continuous production of organic acids will acidify the fermentation broth, requiring the use of neutralization reagents to regulate the pH in the required range of optimal cell growth and fermentation. MgCO₃ is generally regarded as the most effective CO₂ donor and buffer for SA production by A. succinogenes [39]. In the present work, MgCO₃ was selected as the pH regulator, and the effects of different MgCO₃ concentrations (ranging from 10 to 70 g/L) on SA production by A. succinogenes GXAS137 were investigated, as shown in Figure 4. When the MgCO₃ concentration was below 60 g/L, the SA concentration gradually increased as the MgCO₃ concentration increased; this might be because MgCO₃ not only acted as a pH regulator but also provided cofactors (CO₂ and Mg²⁺) for many key enzymes in the SA synthesis pathway [40]. The optimal SA concentration (42.26 g/L) was obtained when the MgCO₃ concentration was 60 g/L. Upon increasing the MgCO₃ concentration beyond 60 g/L, a concomitant decrease in the SA concentration was observed. It might be that the osmotic pressure generated by excessive MgCO₃ (>60 g/L) caused damage to the cells.

3.7. Fermentation in a 2 L Stirred Bioreactor

In order to verify the feasibility of SA production using pretreated SCB hydrolysate as substrate by A. succinogenes GXAS137, batch fermentation of SA was performed in a 2-L stirred bioreactor under identical conditions to anaerobic bottles, as shown in Figure 5. The SA concentration steadily increased with the continuous consumption of reducing sugar, reaching its peak within 36 h. Glucose and xylose were consumed simultaneously, without showing any sign of diauxic growth. The biomass (OD₆₀₀) reached its maximum value of 7.42 at 24 h and then gradually decreased. AA was the main by-product formed in the fermentation process, with a final concentration of about 15.85 g/L. When fermentation was terminated at 36 h, the final SA concentration, yield, and productivity reached 47.06 g/L, 81.3%, and 1.31 g/L/h, respectively. The final SA titer was 11% higher than that achieved in anaerobic bottles (42.26 g/L). This phenomenon can be attributed to the continuous spraying of CO₂ by the bioreactor, which directly penetrates the cell membrane and serves as a substrate for the carboxylation of phosphoenolpyruvate, leading to the conversion of oxaloacetic acid into SA through the reductive TCA cycle [41]. Moreover, the enhanced agitation and mixing capacities of the bioreactor can ensure a better interaction between cells and substrates, thereby facilitating SA production.

Substrate inhibition is an important factor limiting SA production by A. succinogenes. An appropriately high initial substrate concentration can enhance SA production, while an excessively high substrate concentration may lead to inhibition. Previous studies [21,22] have shown that maintaining a moderately low sugar concentration during fermentation could significantly improve the performance of SA production by A. succinogenes. Therefore, fed-batch fermentation was conducted under identical conditions to the batch fermentation. When the residual reducing sugar concentration was below 20 g/L, a condensed SCB hydrolysate containing 150 g/L reducing sugar was fed into the stirred bioreactor, using a peristaltic pump to maintain the residual reducing sugar concentration at 15–20 g/L. Cell growth, reducing sugar consumption, and organic acid production are shown in Figure 6. The cells grew faster in fed-batch fermentation compared to batch fermentation and reached a maximum biomass (OD₆₀₀) of 8.48 at 20 h. Furthermore, there was an obvious longer steady phase at 20–36 h. This might be because a low sugar concentration was maintained, alleviating the cell growth inhibition and thus prolonging the steady phase. When the fermentation was terminated at 52 h, 72.9 g/L of SA was achieved, with a yield of 83.2% and a volumetric productivity of 1.40 g/L/h, respectively; which indicated that fed-batch fermentation was a more efficient strategy for SA production by A. succinogenes.

The cost of carbon sources is one of the main factors limiting the industrial production of chemicals and biofuels. To date, large-scale SA production has primarily focused on starch-based sugars, the production cost of which is prohibitively high for full competition with the traditional petrochemical route. In order to reduce the cost of carbon sources and avoid competition between SA production and food resources, it is necessary to use inexpensive lignocellulose-derived sugars. A. succinogenes can naturally utilize various carbon sources, including glucose, xylose, arabinose, fructose, galactose, cellobiose, and mannose, which provide the possibility of fully utilizing carbon sources in lignocellulosic biomass [4]. Considering the abundance of raw materials and the need to reduce production costs, many efforts have been devoted to exploring SA production by A. succinogenes using inexpensive and renewable lignocellulosic biomass (e.g., cotton stalk, oil palm empty fruit bunches, corn stalk, carob pods, rapeseed meal, coffee husk, etc.), which have yielded positive results (

Table 4. Summary of SA production from different lignocellulosic substrates by A. succinogenes.

Substrate	Nitrogen Resource (g/L)	Fermentation Type	SA Titer (g/L)	SA Yield (%)	SA Productivity (g/L/h)	Reference
Cotton stalk	YE (30)	Batch	63.0	64.0	1.17	2013 Li [44]
Oil palm empty fruit bunches	YE (20) and CSL (20)	Batch	42.9	61.3	0.89	2019 Akhtar [45]
Rapeseed meal	YE (15)	Fed-batch	23.4	11.5	0.33	2011 Chen [46]
Carob Pods	YE (10)	Batch	9.4	54.0	1.32	2016 Carvalho [47]
		Fed-batch	19.0	94.0	1.43
Corn stover	YE (6) and CSL (10)	Continuous	39.6	78.0	1.77	2015 Bradfield [48]
Grape pomace and stalks	YE (5)	Batch	24.6	47.0	0.75	2021 Filippi [49]
		Fed-batch	40.2	67.0	0.79
Sake lees	YE (2.5)	Batch	48.0	75.5	0.94	2010 Chen [42]
Coffee husk	CHH	Batch	19.3	95.0	0.54	2018 Dessie [50]
Sugarcane bagasse	CSLP (20)	Batch	39.9	82.0	1.37	2016 Chen [21]
		Fed-batch	70.8	81.5	1.42
Sugarcane bagasse	SBH	Fed-batch	41.0	32.0	0.30	2021 Chen [8]
Sugarcane bagasse	CSLP (18)	Batch	47.1	81.3	1.31	This study
		Fed-batch	72.9	83.2	1.40

SA, succinic acid; YE, yeast extract; CSL, corn steep liquor; CHH, coffee husk hydrolysate; CSLP, corn steep liquor powder; SBH, sugarcane bagasse hydrolysate.

). The nitrogen source is another key factor affecting industrial SA production. A. succinogenes is a nutrient-fastidious microorganism that requires expensive and complex nitrogen sources, especially YE, for growth and fermentation [33]. However, the extensive use of expensive YE can increase the cost of SA production. Therefore, the use of YE should be reduced or it should be replaced with a cheaper complex nitrogen source to decrease the cost of SA production. Chen et al. conducted batch fermentation using sake lees hydrolysate as a substrate, with only 2.5 g/L YE added, and the resulting SA concentration, yield, and productivity were 48 g/L, 75%, and 0.94 g/L/h, respectively [42]. Shen et al. performed batch fermentation employing duckweed hydrolysate as a substrate, supplemented with 14 g/L CSLP as a nitrogen source. The resulting SA concentration, yield, and productivity were 57.8 g/L, 89.8%, and 1.20 g/L/h, respectively [24]. These results demonstrated the feasibility of reducing the use of YE or substituting it with inexpensive organic nitrogen sources for SA production from lignocellulose hydrolysate by A. succinogenes. However, from the perspective of production cost, the price of CSLP (USD 446 per ton) is approximately 1/13th that of YE (USD 5944 per ton) [43]. Therefore, using CSLP as a nitrogen source is more beneficial for controlling the SA production cost. The results obtained in this study were comparable to or better than those obtained using A. succinogenes and lignocellulosic biomass (Table 4). The comparably high SA concentration, yield, and productivity demonstrated the efficient production of SA from SCB hydrolysate by A. succinogenes GXAS137 with CSLP as a nitrogen source.

4. Conclusions

The present study demonstrated that SCB is a highly promising feedstock for SA production. Among the pretreatment methods used in this study, the two-step chemical pretreatment method using a combination of NaOH and AHP was the most effective in improving the enzymatic hydrolysis efficiency of SCB. Furthermore, the enzymatic hydrolysis process of pretreated SCB was optimized by adjusting the ratio and dosage of enzymes, resulting in a concentration of 78.34 g/L reducing sugars in the hydrolysate and a yield of 83.8%. The optimization results of fermentation conditions demonstrated the feasibility of using inexpensive CSLP as a nitrogen source instead of YE for SA production. The optimal concentrations of initial reducing sugars, CSLP, and MgCO₃ were 70, 18, and 60 g/L, respectively. Fed-batch fermentation was a more efficient cultivation strategy for SA production. When fed-batch fermentation was conducted in a 2 L stirred bioreactor, approximately 72.9 g/L SA was obtained, with a yield of 83.2% and a productivity of 1.40 g/L/h. We utilized enzymatic hydrolysate of SCB as a feedstock for the production of SA by A. succinogenes GXAS137, thereby providing valuable insights for the future resource utilization of agricultural wastes and the economic and efficient industrial-scale production of SA.