Lignocellulose Degrading Weizmannia coagulans Capable of Enantiomeric L-Lactic Acid Production via Consolidated Bioprocessing

: Second-generation lactic acid production requires the development of sustainable and economically feasible processes and renewable lignocellulose biomass as a starting raw material. Weizmannia coagulans MA42 was isolated from a soil sample in Chiang Mai province, Thailand and showed the highest production of L-lactic acid and lignocellulolytic enzymes (cellulase, β -mannanase, xylanase, β -glucosidase, β -mannosidase, and β -xylosidase) compared to other isolates. Weizmannia coagulans MA42 was able to grow, secrete lignocellulolytic enzymes, and directly produce L-lactic acid in the medium containing various lignocellulosic feedstocks as the sole carbon source. Moreover, L-lactic acid production efﬁciency was improved after the substrates were pretreated with diluted sulfuric acid and diluted sodium hydroxide. The highest L-lactic acid production efﬁciency of 553.4 ± 2.9, 325.4 ± 4.1, 326.6 ± 4.4, 528.0 ± 7.2, and 547.0 ± 2.2 mg/g total available carbohydrate was obtained from respective pretreated substrates including sugarcane bagasse, sugarcane trash, corn stover, rice straw, and water hyacinth. It is suggested that structural complexity of the lignocellulosic materials and properties of lignocellulolytic enzymes are the key factors of consolidated bioprocessing (CBP) of lignocellulosic feedstocks to lactic acid. In addition, the results of this study indicated that W. coagulans MA42 is a potent bacterial candidate for CBP of a variety of lignocellulosic feedstocks to L-lactic acid production; however, further bioprocess development and genetic engineering technique would provide higher lactic acid production efﬁciency, and this would lead to sustainable lactic acid production from lignocellulosic feedstocks.


Introduction
Lactic acid (LA) or 2-hydroxypropanoic acid is a naturally occurring hydroxycarboxylic acid that has found applications in medicinal, pharmaceutical, agricultural, food, lignocellulolytic B. coagulans or Weizmannia coagulans strain capable of fermenting xylose to optically pure L-lactic acid as well as secreting lignocellulolytic enzymes from various lignocellulosic substrates and the application of W. coagulans in the production of lactic acid from lignocellulosic feedstocks by CBP.

Isolation and Screening of Presumptive Bacillus coagulans (Weizmannia coagulans) Strains from Soil Samples
Fifteen soil samples were collected from Chiang Mai province, Thailand and were used to isolate Bacilli strains that are able to produce lactic acid. One gram of soil sample was enriched in pasteurized milk of 100 mL and incubated at 55 • C under static conditions for 3 days, using a shaking incubator (Labtech, Daihan Labtech LSI-3016A, Namyangju, Republic of Korea). The suspension was serially diluted using sterile 0.85% NaCl to make ten-fold dilution ranging from 10 0 to 10 7 , and 0.1 mL of each dilution was spread on plate of nutrient agar (NA) containing 10 g/L of xylose and supplemented with 0.1% (w/v) bromocresol purple as an indicator. After incubating at 55 • C for 24 h, colonies that showed yellow clear zones on the plates with purple background were picked for further screening step. A single colony of all bacterial isolates was picked using sterile needles and spotted onto xylose, yeast extract, and peptone (XYP) agar (per liter): xylose 10 g, yeast extract 5 g, peptone 5 g, KH 2 PO 4 250 mg, K 2 HPO 4 250 mg, and salt solution (MgSO 4 400 mg, MnSO 4 20 mg, FeSO 4 20 mg, and NaCl 20 mg in 100 mL) 10 mL, CaCO 3 0.5% (w/v) as indicator and Thermoaciduran agar (per liter): yeast extract 5 g, peptone 5 g, dextrose 5 g, K 2 HPO 4 4 g [19], and incubated at 55 • C for 24 h. The culture aliquot was centrifuged at 14,000 rpm for 3 min and the supernatant was determined for D-or L-lactic acid content. The isolates that produced the highest optically pure D-lactic acid or L-lactic acid were selected for identification and characterization.

Identification and Biochemical Characterization
The 16S rRNA gene was amplified by PCR using primers 27F (5 -AGAGTTTGATCMTG GCTCAG-3 ) and 1525R (5 -AAGGAGGTGWTCCARCC-3 ) and genomic DNA as the template. The PCR was operated using Phusion ® High-Fidelity PCR Master Mix (New England Biolabs, Beverly, MA, USA), while the quality of the PCR product was validated using agarose gel electrophoresis. The nucleotide sequence of the amplified 16S rRNA gene was determined by the service provider (1st BASE Laboratory Company, Singapore). Online resemblance searches were performed using the BLAST algorithm of GenBank. Multiple sequence alignment was done using the BioEdit program [20], and subsequent phylogenetic tree analysis was performed using the MEGA4 program [21]. Characterizations of the isolated bacterial strains including Gram staining and growth in aerobic and anaerobic conditions were investigated in XYP broth. Seed inoculum was prepared by transferring a loopful of single colony of each strain in XYP broth and static incubated at 55 • C for 18 h or until the OD 600 reached 0.8-0.9 (10 8 CFU/mL). For aerobic conditions, 2% (v/v) of seed inoculum was transferred into a test tube containing 10 mL broth; in case of anaerobic conditions, 2 mL of sterilized paraffin oil was added to cover the broth surface. Furthermore, growths in the presence of various carbon sources (glucose, xylose, carboxymethyl cellulose (CMC), locust bean gum (LBG), and beechwood xylan) were also investigated in XYP broth. In these experiments, cultivation of bacteria was performed in liquid media based on XYP broth in which xylose was substituted with carbon sources as mentioned above. Moreover, the optimal temperatures (45 to 55 • C) and pH (4.0 to 9.0) for growth were investigated in XYP broth. The lactic acid production test was comparatively studied under controlled and non-controlled pH conditions. Shortly, the inoculum (2%, v/v) was transferred to a 125 mL Erlenmeyer flask containing 50 mL of XYP broth with 10 g/L of xylose as the sole carbon source. The fermentation was carried out at 55 • C for 24 h under static condition. In the pH-controlled fermentation, the culture was manually adjusted to pH 6.0 by addition of 1.0 M NaOH every 6 h. After the cultivation, the culture was harvested by centrifugation. The supernatant was determined for lactic acid content. The acetic acid, formic acid, and ethanol were also determined by high-performance liquid chromatography (HPLC) as described by Unban et al. [22]. Additionally, the selected strains were investigated for the ability to produce some lignocellulolytic enzymes via plate screening technique. Briefly, a loopful of each strain was spotted on NA plate that was supplemented with 0.01% (w/v) trypan blue as an indicator, and different carbon sources (CMC, LBG, and xylan) at a level of 0.5% (w/v) as the sole carbon source. After 24 h of incubation at 55 • C, the clear zones on a dark blue background were considered a positive result.

Potential Evaluation of Weizmannia coagulans Strains for Lignocellulolytic Enzyme Production
A total of 2% (v/v) inoculum of each strain of W. coagulans was transferred to a 125 mL Erlenmeyer flask containing 50 mL of nutrient broth containing 0.5% (w/v) of different carbon sources including CMC, LBG, and beechwood xylan as the sole carbon source. The fermentation was carried out at 50 • C for 24 h. After the incubation, the culture was centrifuged at 12,000 rpm and at 4 • C for 10 min. The clear supernatant was used for determination of cellulase, β-mannanase, and xylanase activities according to their corresponding substrates as described below (Section 2.7.1).

Effect of Different Agricultural Residues on Lignocellulolytic Enzyme Production
Five different agricultural residues (corn stover, sugarcane bagasse, sugarcane trash, rice straw, and water hyacinth) were used as source of lignocellulosic substrate for investigation of ability of W. coagulans MA42 in production of lignocellulolytic enzymes including cellulase, xylanase, β-mannanase, β-glucosidase, β-xylosidase, and β-mannosidase. The results would lead to selection of the most appropriate lignocellulosic feedstocks for lactic acid production by W. coagulans MA42. Primarily, the selected agricultural residues were washed with water and dried at 60 • C until constant weight. The dried residues were ground by using a hammer mill and passed through a 20-mesh sieve to achieve a fine particle. To prepare culture media, 2.5 g ground agricultural residue was used as the sole carbon source of XYP broth in which xylose was substituted with agricultural residue prior to sterilization at 121 • C for 20 min. The seed inoculum of 10% (v/v) of W. coagulans MA42 was transferred to 125 mL Erlenmeyer flasks containing 50 mL of the prepared XYP broth. The inoculated culture media were statically incubated at 50 • C for 48 h. Samples were collected every 12 h and centrifuged at 14,000 rpm for 5 min. The clear supernatant was used as crude enzyme for determination of lignocellulolytic enzyme activity.
2.6. CBP of Lignocellulosic Feedstock to Lactic Acid 2.6.1. Sources of Lignocellulose Five lignocellulosic agricultural waste feedstocks including corn stover, sugarcane bagasse, sugarcane trash, rice straw, and water hyacinth were selected as the available agricultural wastes. Prior to using these materials as a substrate for CBP for lactic acid production, their initial total available carbohydrate content was determined. A total of 0.5 g of each lignocellulose substrate was mixed with 8 mL of 72% (v/v) H 2 SO 4 and incubated at 30 • C on a 100-rpm rotary shaker for 1 h. After the incubation, the reaction was transferred to 300 mL of distilled water. This mixture was then autoclaved under 121 • C for 30 min and left to cool. The acid-heat hydrolysate was centrifuged at 14,000 rpm for 5 min. The supernatant was neutralized with CaCO 3 prior to centrifugation to collect a clear supernatant which was determined for total carbohydrate content by phenol-sulfuric acid method [23]. Total carbohydrate is referred to as an available fermentable carbohydrate for lactic acid production.

CBP of Lignocellulosic Feedstocks to Lactic Acid
For production of lactic acid, the media and culture conditions were the same as those that were described in Section 2.5. All lignocellulosic feedstocks were used as the sole carbon source of XYP broth in which xylose was substituted with lignocellulosic feedstocks prior to sterilization at 121 • C for 20 min. Fermentation was carried out at 50 • C for 48 h. During the fermentation, the pH was maintained between 6 and 7 by manually adding 1.0 M NaOH every 6 h of the fermentation. Samples were collected every 6 h for determination of lactic acid content. Lactic acid production was calculated in terms of total lactic acid (g) per total available carbohydrate (g). At the end of the fermentation, the residual substrate was harvested by centrifugation at 14,000 rpm for 10 min. The pellet was once washed with deionized water and filtered through a sheet of cloth. The residue was dried at 60 • C until constant weight prior to the determination of total available carbohydrate according to the method described in Section 2.6.1. Finally, the total available residual substrate was calculated.

CBP of Pretreated Lignocellulosic Feedstocks to Lactic Acid
All lignocellulosic feedstocks were prepared by acid and alkali pretreatments. Briefly, 2.5 g of the ground lignocellulose obtained from Section 2.5 were independently pretreated with 100 mL of 0.05 M H 2 SO 4 , 0.25 M H 2 SO 4 , 0.05 M NaOH, and 0.25 M NaOH. The suspension was autoclaved at 121 • C for 20 min. After cooling to room temperature (30 • C), the pretreated lignocellulose was separated by filtration and washed with excess tap water to remove the remaining H 2 SO 4 or NaOH. Finally, the pretreated lignocellulose was dried at 60 • C until constant weight. The pretreated lignocellulose with an equivalent amount of 2.5 g unpretreated lignocellulose was used as the sole carbon source for preparation in 50 mL XYP broth in which xylose was substituted with pretreated lignocellulose. Fermentation was carried out at 50 • C for 48 h. During the fermentation, the pH was maintained manually between 6 and 7 by 1.0 M NaOH.

Enzyme Activity Assay
Cellulase, β-mannanase, and xylanase activities were assayed by the method modified from Wongputtisin et al. [24] based on determining the release of reducing sugars from the corresponding substrates, CMC, LBG, and beechwood xylan that could be detected by 3,5-dinitrosalicylic acid (DNS) method [25]. Each substrate was prepared in 50 mM sodium phosphate buffer, pH 7.0. The reaction was carried out at 50 • C for 30 min and was performed by mixing an equal volume (125 µL) of substrate and culture broth supernatant (crude enzyme). The reaction was terminated by adding 250 µL of DNS solution and then placed on a boiling bath for 10 min. Finally, 2 mL of distilled water was added and mixed before measuring an absorbance at 540 nm. Enzyme activities were calculated and expressed as units per milliliter. One unit of enzyme is defined as the amount of enzyme that catalyzes the hydrolysis of the substrate to liberate 1 µmole reducing sugars per minute under the assay conditions.
For determination of β-glucosidase, β-xylosidase, and β-mannosidase activities, their corresponding substrates were p-nitrophenyl-β-D-glucopyranoside (pNPG), p-nitrophenylβ-D-xylopyranoside (pNPX), and p-nitrophenyl-β-D-mannopyranoside (pNPM), respectively. Each substrate was prepared in 50 mM sodium phosphate, pH 7.0. The reaction mixture consisted of an equal volume (125 µL) of crude enzyme and substrate. The reaction was carried out at 50 • C for 30 min. The reaction was terminated by adding 50 mM Na 2 CO 3 . After that, it was measured at an absorbance of 405 nm. The amount of pNP was calculated based on a standard plot. One unit of enzyme activity was defined as the amount of enzyme liberating 1 µmole of pNP per minute under the described conditions.

Lactic Acid Production Analysis
D-or L-lactic acid was measured by D-/L-lactic acid assay kit (Megazyme, Wicklow, Ireland) as described by manufacturer's instructions. Lactic acid production efficiency was calculated according to ratio of the total lactic acid produced (g) to total available carbohydrate (g) present in lignocellulose.

Isolation of Presumptive Weizmannia coagulans
The numbers of bacterial colonies were isolated from 15 soil samples. There were approximately 1000 colonies on agar plates; around half of them produced yellow clear zone on NA supplemented with 10 g/L of xylose as the sole carbon source, and bromocresol purple. Well growth single colonies of 100 isolates were randomly selected for conformation on thermoaciduran agar, which is a B. coagulans selective medium, and simultaneously tested for xylose-utilizing capability on XYP agar. All tested bacterial isolates showed growth on XYP agar, but only 15 bacterial isolates exhibited well growth by forming colonies with clear zone on thermoaciduran agar and XYP agar. Thermoaciduran agar supports the abundant growth of W. coagulans and gives more abundant sporulation than normal media [26]. In addition, thermoaciduran medium (pH 4.6) was specifically developed and validated as a selective medium for the selection of W. coagulans by the study of Nakajo et al. [19]. Furthermore, XYP agar was used as a selective media to confirm lactic acid production based on xylose as a carbon source. Therefore, these bacteria were presumptively considered W. coagulans and further evaluated for lactic acid production and lactic acid purity. The results revealed that all tested strains were able to produce lactic acid in the range of 0.6 to 2.3 g/L with high optical purity of L-lactic acid from XYP broth (Table 1). Three out of 15 bacterial isolates, namely MA42, P13, and S5 produced only L-lactic acid with lactic acid contents of 1.60 ± 0.21, 1.74 ± 0.07, and 1.43 ± 0.16 g/L, respectively. Although the lactic acid producing ability of these isolates was lower than those produced in the previous study [17,27,28], it is highly recommended that high optically pure D-or L-lactic acid producers are more promising than those that produce racemic mixture of lactic acid due to application of lactic acid for the biodegradable plastic polymerization process [29]. In addition, lactic acid production from isolates MA42, P13, and S5 would require further characterization to evaluate their potentials. According to Vijayakumar et al. [1], the temperature range of lactic acid bacteria growth is 20 to 45 • C. However, mesophilic has the disadvantage of contamination by mesophilic microbes during fermentation, which is a serious problem at the industrial scale. To overcome this problem, the thermophilic lactic acid microbes were extensively studied. The temperature range of thermophiles ranges between 45 to 80 • C. Additionally, thermophilic is also useful for industrial bioprocesses at high temperatures such as biorefinery which requires commercial enzymes that are stable at high temperatures to change the structure of the biomass [6]. Since they can function in a wider range of conditions and are more resilient, thermophilic organisms are normally preferable than mesophilic.

Biochemcial Characterization and Molecular Identification of the Presumptive Weizmannia coagulans
Summary of biochemical characteristics of isolates MA42, P13, and S5 are shown in Table 2. Isolates MA42, P13, and S5 are Gram positive, rod-shaped, and endospore forming bacteria. They exhibited growth in medium with D-glucose, D-xylose, CMC, LBG, and xylan. Growth was observed both under aerobic and anaerobic conditions, thus being assigned as facultative anaerobic microorganisms ( Figure S1). The optimal temperature for growth was 50 • C for isolates MA42 and P13, and 45 • C for isolate S5. All isolates had optimal pH values for growth in range of 5-7. Isolates MA42 and P13 demonstrated positive results for cellulase, β-mannanase, and xylanase, but isolate S5 showed positive results for cellulase and β-mannanase but negative for xylanase. Lactic acid production efficiency of all isolates was limited by pH conditions. L-lactic acid content produced under non-control pH was much lower than that produced under control pH condition (pH 6-7). Isolates MA42, P13, and S5 produced the highest lactic acid contents of 7.88 ± 0.20, 6.76 ± 0.13, and 7.13 ± 0.16 g/L from 10 g/L xylose, respectively, which are considered to be highly effective in producing lactic acid compared to previous studies (Table S1) [7,17]. It is noted that 95% of organic acids produced by these strains was lactic acid, and the other 5% was identified as acetic acid and/or formic acid and/or ethanol, after confirmed by high-performance liquid chromatography (HPLC). Therefore, all isolates were classified as homofermentative lactic acid producers. Table 1. L-lactic acid content and optical purity of L-lactic acid produced by various strains of the presumptive Wiezmannia coagulans.   A total of 1419, 1415, and 1421 nucleotides were sequenced of the 16S rRNA gene sequence of isolates MA42, P13, and S5, which are closely related to B. coagulans ATCC 7050 with 99.72%, 99.79%, and 99.86% identities, respectively. The phylogenetic tree of the 16S rRNA gene sequence clearly distinguished these isolates from other Bacillus spp. and placed isolates MA42, P13, and S5 in the group of B. coagulans (Figure 1). Based on the results, these isolates could be clearly identified as B. coagulans. The current review summarized that B. coagulans is a Gram-positive, facultative anaerobic, non-toxic, rodshaped, endospore-producing bacterium with its optimal temperature for growth ranging from 30 to 50 • C and pH ranging from 5.5 to 7.0. It is able to decompose and utilize a variety of pentose and hexose carbohydrates and can produce lactic acid [30]. These biochemical properties are in agreement with the results of this study. However, the current elucidation of the evolutionary relationship of Bacillus has reclassified members of genus Bacillus into 17 new genera and has renamed B. coagulans as Weizmannia coagulans [31]. Weizmannia coagulans could act as a potential probiotic with regard to its ability to be strongly tolerant to stomach acid, cholic acid, and other harsh environments. Its habitat has been recorded to occur in canned evaporated milk [32], industrial wastewater drainage [33], bean processing waste [6], lignocellulosic biomass [34], and soil (this study). Weizmannia coagulans strains MA45, P13, and S5 were evaluated for their potential in the production of various cellulolytic enzymes in the following experiments.

Selection of Weizmannia coagulans Strain for CBP of Lactic Acid Production Based on Lignocellulosic Enzyme Production
Lignocellulosic-enzyme-producing ability is one of the important criteria for selection of the most suitable strains of W. coagulans for CBP of lignocellulose to lactic acid. In preliminary studies, three strains of W. coagulans were grown in NB containing CMC, LBG, and xylan for quantitative analysis of their corresponding enzyme activity toward each substrate, namely cellulase, β-mannanase, and xylanase, although detection of these enzymes had been investigated by plate screening methods (Table 3).

Selection of Weizmannia coagulans Strain for CBP of Lactic Acid Production Based on Lignocellulosic Enzyme Production
Lignocellulosic-enzyme-producing ability is one of the important criteria for selection of the most suitable strains of W. coagulans for CBP of lignocellulose to lactic acid. In preliminary studies, three strains of W. coagulans were grown in NB containing CMC, LBG, and xylan for quantitative analysis of their corresponding enzyme activity toward each substrate, namely cellulase, β-mannanase, and xylanase, although detection of these enzymes had been investigated by plate screening methods (Table 3). Basically, an extracellular and endo-type enzyme plays a crucial role in microbial metabolism of carbohydrates as it randomly cleaves a polymeric structure of the carbohydrate to oligosaccharides which in part can be directly assimilated into cells and further specifically cleaved into monosaccharides by glycosidase prior to assimilation [35]. Cellulase is an endo-type enzyme that catalyzes the random hydrolysis of cellulose, the major structural component of lignocellulosic materials. Xylanase and β-mannanase are also endo-type enzymes that catalyze the random hydrolysis of xylan and mannan, two important components of hemicellulose. It is noteworthy that xylan is the core structure of most hemicellulose [36], thus xylanase must be considered a representative key enzyme that associates with degradation of hemicellulose. Isolate MA42 produced the highest β-mannanase and xylanase activities, and its cellulase activity was less than that produced by isolate P13, but this level was still high, up to 0.370 ± 0.01 U/mL. It is important to note that although isolate P13 produced the highest levels of cellulase and β-mannanase, its xylanase activity was much lower than that produced by isolate MA42. Isolate S5 produced cellulase, β-mannanase, and xylanase inefficiently due to much lower enzyme activities than those produced by isolates MA42 and P13. Based on the overall enzyme activities, isolate MA42 is considered the most promising strain for CBP of lignocellulose to lactic acid.

Effect of Different Agricultural Residues on Lignocellulolytic Enzyme Production
The ability to produce lignocellulolytic enzymes from lignocellulosic feedstocks is very important for the lactic acid production industry. As commercial enzymes are mainly used now for enzymatic saccharification and hydrolysis of cellulose and hemicellulose to release fermentable sugars needed for fermentative production of lactic acid [8], the capability and potential of this bacterial strain to produce its own lignocellulosic enzymes has stoked interest as it has potential to reduce production cost by reducing commercial purchase of these enzymes. Yet, the potential for lignocellulolytic enzyme production by W. coagulans MA42 was investigated using different lignocellulose materials as the sole carbon sources including corn stover, sugarcane bagasse, sugarcane trash, rice straw, and water hyacinth. As previously described, not only extracellular and endo-type enzymes but also glycosidase are the most important key factors for efficient hydrolysis of lignocellulose to monosaccharides which are then fermented to lactic acid. Additional enzymes that are required for efficient hydrolysis of lignocellulose are β-glucosidase, β-xylosidase, and β-mannosidase. Although W. coagulans has been extensively studied for lactic acid production from biomass, its profile regarding production of various lignocellulolytic enzymes is still unknown. The results showed that W. coagulans MA42 was able to produce all tested enzymes and secreted them to the culture media containing different lignocellulosic feedstocks (Figure 2), indicating its ability to produce lactic acid from these substrates. The enzyme quantity varied depending on the lignocellulosic substrate and fermentation time with cellulase and xylanase as the major enzymes. However, overall, it is likely that W. coagulans MA42 preferred corn stover, sugarcane bagasse, and rice straw rather than sugarcane trash and water hyacinth due to slightly low levels of all produced enzymes. The extracellular cellulase and xylanase occurred in an average range of 0.20 to 0.98 U/mL while the amount of β-mannanase was found at lower levels. It can be explained that the simultaneous formation of these enzymes could be induced by several polysaccharides structurally resembling the carbohydrate constituents of lignocellulose [37] such as cellu-lose, hemicellulose, and lignin, and this statement agrees with previous studies [38]. The endo-type enzymes always occurred between 24 and 36 h of fermentation, which was in the initial to the middle stage of fermentation, whereas glycosidases reached the highest activity after 36 h of fermentation. This is in accordance with a reason previously described in Section 3.3. Glycosidases determined in this study included β-glucosidase, β-xylosidase, and β-mannosidase which were found at levels of 0.002-0.008 U/mL. Thus far, the information about production of lignocellulolytic enzymes from W. coagulans is limited, with the exception of lipase [39], α-amylase [40], cellulase [6], and xylanase [41]. Therefore, it is expected that these enzymes may have sufficient stability to promote CBP of lignocellulose to lactic acid. In addition, this study has provided a profile of lignocellulose-degrading enzymes from different lignocellulosic feedstocks. From previous reports, B. coagulans has been shown to have maximum cellulase activity (0.812 U/mL) at 37 • C and at neutral pH with CMC as a carbon source [42], while Acharya and Chaudhary [43] recorded cellulase activity of 0.128 U/mL with CMC as a carbon source at 50 • C. In addition, maximum xylanase enzyme production was obtained at 17.8 U/mL with wheat straw as a carbon source at 45 • C, and decreased enzyme secretion was seen by B. coagulans at higher temperatures (less than 3 U/mL at 50 • C) [41]. Moreover, Bukola et al. [44] reported that 40 • C was the optimum temperature required to attain maximum activity for mannanase (0.20 U/mL) produced by B. coagulans U3 from LBG. Furthermore, 0.004 U/mL of β-Glucosidase activity was produced by B. coagulans PR03 from soy germ [45]. However, it cannot be denied that the enzyme yield from Bacillus spp. was low from a practical viewpoint [43]. 0.004 U/mL of β-Glucosidase activity was produced by B. coagulans PR03 from soy germ [45]. However, it cannot be denied that the enzyme yield from Bacillus spp. was low from a practical viewpoint [43].

CBP Lactic Acid Production from Lignocellulose Materials
In CBP, an organism or mixed culture of organisms produces enzymes for hydrolysis of lignocellulosic biomass and ferments the released sugars into lactic acid [46]. With regard to this context, attempts to screen microorganisms for simultaneous production of lactic acid from lignocellulose biomass have been extensively investigated. Weizmannia coagulans MA42 can grow rapidly in the early stages of fermentation and enter stationary phase after 18 h of fermentation. However, growth did not approach the death phase during 48 h incubation ( Figure S2). Direct lactic acid production from five agricultural residues was not successful due to low lactic acid content ( Figure 3A). The highest lactic acid

CBP Lactic Acid Production from Lignocellulose Materials
In CBP, an organism or mixed culture of organisms produces enzymes for hydrolysis of lignocellulosic biomass and ferments the released sugars into lactic acid [46]. With regard to this context, attempts to screen microorganisms for simultaneous production of lactic acid from lignocellulose biomass have been extensively investigated. Weizmannia coagulans MA42 can grow rapidly in the early stages of fermentation and enter stationary phase after 18 h of fermentation. However, growth did not approach the death phase during 48 h incubation ( Figure S2). Direct lactic acid production from five agricultural residues was not successful due to low lactic acid content ( Figure 3A). The highest lactic acid production efficiency of 148.4 ± 3.9, 101.8 ± 1.2, 160.7 ± 1.7, 136.8 ± 1.5, and 110.2 ± 1.3 mg/g total available carbohydrate was obtained when the substrates were the following: sugarcane bagasse, sugarcane trash, corn stover, rice straw, and water hyacinth, respectively. Total available carbohydrates present in the residual biomass ranging from 50-75% were retained ( Figure 3B), indicating an inefficient lactic acid production. It was proposed that lactic acid production efficiency and properties of lignocellulolytic enzymes and substrate complexity may be the main drawbacks that hinder simultaneous lactic acid production from lignocellulosic feedstocks by W. coagulans MA42. Based on the previous studies, the drawback of lignocellulose bioconversion to lactic acid is the limit in cellooligosaccharides and cellodextrin utilization. However, efficient conversion of cellooligosaccharide to L-lactic acid by a recombinant cellulolytic lactic acid bacterium, Lactococcus lactis, has been reported [47]. The current lactic acid production from lignocellulosic feedstocks by W. coagulans has added benefits of exogenous enzymes to help fermentative bioconversion into lactic acid based on enhanced hydrolysis of lignocellulose which consequently promotes an efficient substrate utilization to lactic acid [27]. Alternatively, metabolic engineering of a native lignocellulolytic enzyme-producing microorganism with an organic acid-producing pathway can be employed [48]. Pretreatment is a step to alter structural characteristics of biomass to increase cellulose and hemicellulose accessibility to lignocellulolytic enzymes. Acid pretreatments employ acids as catalysts which have a stronger effect on hemicellulose and lignin than on crystalline cellulose. Acid pretreatments can solubilize hemicellulose of the biomass and consequently, make cellulose more accessible to enzymes. Alkali pretreatments can increase cellulose digestibility and they are more effective for lignin solubilization, having less effect on cellulose and hemicellulose than acid pretreatments [49]. To investigate the potential of W. coagulans MA42 and its lignocellulolytic enzyme production, all substrates were pretreated with acid and alkali to promote substrate availability for lignocellulolytic enzymes.
After acid and alkali pretreatments, lactic acid content produced by W. coagulans MA42 and lactic acid production efficiency substantially increased, specifically when sugarcane bagasse, sugarcane trash, rice straw, and water hyacinth were used as the sole carbon source ( Table 4). The acid pretreatment of 0.25 M H2SO4 significantly improved lactic acid production efficiency than the alkali with regard to the highest lactic acid content of Based on the previous studies, the drawback of lignocellulose bioconversion to lactic acid is the limit in cellooligosaccharides and cellodextrin utilization. However, efficient conversion of cellooligosaccharide to L-lactic acid by a recombinant cellulolytic lactic acid bacterium, Lactococcus lactis, has been reported [47]. The current lactic acid production from lignocellulosic feedstocks by W. coagulans has added benefits of exogenous enzymes to help fermentative bioconversion into lactic acid based on enhanced hydrolysis of lignocellulose which consequently promotes an efficient substrate utilization to lactic acid [27]. Alternatively, metabolic engineering of a native lignocellulolytic enzyme-producing microorganism with an organic acid-producing pathway can be employed [48]. Pretreatment is a step to alter structural characteristics of biomass to increase cellulose and hemicellulose accessibility to lignocellulolytic enzymes. Acid pretreatments employ acids as catalysts which have a stronger effect on hemicellulose and lignin than on crystalline cellulose. Acid pretreatments can solubilize hemicellulose of the biomass and consequently, make cellulose more accessible to enzymes. Alkali pretreatments can increase cellulose digestibility and they are more effective for lignin solubilization, having less effect on cellulose and hemicellulose than acid pretreatments [49]. To investigate the potential of W. coagulans MA42 and its lignocellulolytic enzyme production, all substrates were pretreated with acid and alkali to promote substrate availability for lignocellulolytic enzymes.
After acid and alkali pretreatments, lactic acid content produced by W. coagulans MA42 and lactic acid production efficiency substantially increased, specifically when sugarcane bagasse, sugarcane trash, rice straw, and water hyacinth were used as the sole carbon source ( Table 4). The acid pretreatment of 0.25 M H 2 SO 4 significantly improved lactic acid production efficiency than the alkali with regard to the highest lactic acid content of 149.2 ± 1.2, 91.8 ± 1.1, 117.6 ± 4.2, 114.3 ± 4.1, and 129.9 ± 2.2 mg/g substrate, and lactic acid production efficiency of 553.4 ± 2.9, 325.4 ± 4.1, 326.6 ± 4.4, 528.0 ± 7.2, and 547.0 ± 2.2 mg/g total available carbohydrate when sugarcane bagasse, sugarcane trash, corn cob, rice straw, and water hyacinth were used as the substrate, respectively. These values led to respective % increased values of 92.4, 105.8, 64.0, 78.7, and 154.4%. At the maximum lactic acid production efficiency, total available carbohydrate in the agricultural residue retained after the fermentation was approximately 45% (w/w) which was reduced by 20-30% (w/w) compared to those that used unpretreated substrates (Table S2). This resulted in approximately 55% of the pretreated substrates being employed by W. coagulans MA42 to saccharify and further ferment to lactic acid. Moreover, it is implied that H 2 SO 4 might provide a stronger effect on cellulose structure than NaOH, and its property in solubilizing hemicellulose may promote saccharification of xylan, thus promoting available cellulose and hemicellulose for saccharification and fermentation. Direct lignocellulose fermentation to lactic acid through CBP is highly challenging [48], and this is in agreement with the results of this study. Considering microorganisms used for the CBP of lignocellulosic feedstocks to lactic acid, three groups of microorganisms have been utilized including natural microorganisms in a single culture, natural microorganisms in co-culture, and genetically engineered microorganisms in a single or co-culture [11].
To develop a simple process for lactic acid production, natural microorganisms are preferable. Notably, it may require attempts to isolate and screen for potential microorganisms. There have been previous reports that described the production of L-lactic acid from lignocellulose using B. coagulans, formerly Weizmannia coagulans. However, most of those reports applied an enzymatic saccharification process prior to the fermentation step [50,51]. Moreover, various methods adopted in the treatment of lignocellulose substrates come with energy and time costs [52]. Therefore, lactic acid production from lignocellulose substrates using CBP is an attractive alternative. Thermo-alkaline condition on banana peels has been used as the substrate for CBP of lactic acid production by Enterococcus durans BP130. The maximum lactic acid content of 15.9 g/L was achieved after direct production from 1:10 (w/v) unpretreated banana peels by E. faecium FW 26; this value is equivalent to 0.159 g/g banana peels [53], whereas the maximum lactic acid concentration of 19.2 g/L was achieved when using 6.7% (w/v) of the pretreated banana peels, and this value is equivalent to 0.28 g/g substrate [54]. Recently, washed steam-pretreated beechwood solid matrix containing only microcrystal cellulose has been used as a substrate for CBP of lactic acid production by a co-culture of Trichoderma reesei and Lactobacillus pentosus. The maximum lactic acid concentration of 15.1 g/L was obtained from 3.86% (w/w) of substrate with the calculated lactic acid production efficiency of 0.4 g/g substrate, and after process optimization, the lactic acid content was elevated up to 19.8 g/L [55]. It is noticeable that glycosidases produced by W. coagulans MA42 may not be efficient to promote saccharification of the released lignocellulosic substrate derived from endo-type enzyme reactions. Based on the results, the organism could only ferment 50% of the total available carbohydrates.
To the best of our knowledge, this study considered the possibility of using various lignocellulosic agricultural waste feedstocks including corn stover, sugarcane bagasse, sugarcane trash, rice straw, and water hyacinth directly for lactic acid production via CBP at high-temperature conditions. On a positive note, our study might offer a potential option that could reduce the cost associated with the upstream process. To further efficiently use W. coagulans MA42 for direct lactic acid production from lignocellulosic feedstocks, it is recommended to supplement some lignocellulolytic enzymes, specifically glycosidases, to enhance lactic acid production efficiency. In addition, either strain improvement or co-culture fermentation with an appropriate microorganism could be an alternative.

Conclusions
In this research, we selected a new thermophilic and lignocellulolytic strain of W. coagulans MA42 isolated from soil collected from Chiang Mai, Thailand, for its potential for lactic acid production directly from lignocellulose substrates. The MA42 strain showed an ability to produce optically pure L-lactic acid under facultative anaerobic conditions from xylose (C5 sugar) and optimal pH and temperature of 5.0-6.0 and 50-55 • C, respectively. For direct bioprocessing of lignocellulosic feedstocks to lactic acid, W. coagulans produced mainly cellulase, β-mannanase, and β-xylanase in addition to glycosidases, namely βglucosidase, β-mannosidase, and β-xylosidase, which were produced at low levels. In terms of lactic acid production, the results indicated that W. coagulans MA42 is a potent bacterial candidate for CBP of a variety of lignocellulosic feedstocks for L-lactic acid production; however, further bioprocess development and genetic engineering techniques would provide higher lactic acid production efficiency, and this would lead to sustainable lactic acid production from lignocellulosic feedstocks.