Biomass Valorization: Sustainable Methods for the Production of Hemicellulolytic Catalysts from Thermoanaerobacterium thermostercoris strain BUFF

: Processing and selection of fruits and vegetables generate high quantities of wastes that represent an economic and environmental issue for the agroindustry sector. According to the so-called “bioreﬁnery” approach, this biomass can be exploited for the recovery of value-added molecules. In this study, the residues of industrial processing of tomato ( Lycopersicon esculentum variety “Hybrid Rome”), fennel ( Foeniculum vulgare ), potato ( Solanum tuberosum ) and carrot ( Daucus carota ) were used as sole carbon sources to support cheap and sustainable microbial growth as well as the production of secondary metabolites (hydrogen and ethanol) by Thermoanaerobacterium thermostercoris strain BUFF, a thermophilic anaerobic microorganism isolated from buffalo-dung compost. Moreover, the use of hemicellulolytic enzymes of T. thermostercoris was assayed in the bioconversion reaction of the polymer fraction extracted from the rhizome of giant reed ( Arundo donax ) and of the leaves and stems of cardoon ( Cynara cardunculus ), dedicated non-food crops employed in energy supply. ( w / v ) Congo containing 5% ( v / v ) ethanol for 30 min, destained with 1 M NaCl (15 min, 2 times). The xylanase activity was visible as a clear band against a red background. The β -xylosidase and α -arabinofuranosidase activities in the gel were detected by developing a zymogram against 50 µ g/mL 4-methylumbelliferyl- β -D-xyloside or 50 µ g/mL 4-methylumbelliferyl-α -arabinofuranoside as substrate, respectively, prepared in 50 mM sodium citrate buffer (pH 5.6). After the electrophoretic run, the gel was immersed in substrate solution for 15 min at 70 ◦ C in the dark. The enzyme bands in the gel were visualized under UV light at λ 365 nm. The acetyl-xylan esterase activity in the gel was revealed by incubating the gel, after the electrophoretic run, in 20% ( v / v ) isopropanol then in 50 mM sodium acetate buffer pH 5.6 and ﬁnally in 2 mM p -nitro-phenyl acetate. Yellow bands after 10 min of incubation appeared in the presence of the enzyme.


Introduction
The circular economy is an alternative to the current linear, "make, use, dispose", economy model, which aims to keep resources in use for as long as possible to extract the maximum value from them whilst in use and to recover and regenerate products and materials at the end of their service life. It offers an opportunity to minimize the negative impacts of biomass's exploitation while maximizing the benefits from vegetable wastes and providing environmental, economic and societal benefits [1]. In addition, given the need to increase circularity of industrial systems to address limited resource availability, the development of the food waste biorefinery concept is also welcome, taking into account several critical aspects related to the technical feasibility of the processes at an industrial scale and their techno-economic potential, including available quantities of waste and a life-cyclebased environmental assessment [2]. However, the optimization of feedstock availability and transport should be systematically implemented to ensure that the profitability of biorefineries does not lead to trade-offs, such as increasing environmental impacts and externalities due to logistics [3]. Generally, when we refer to agro-industrial wastes, we are taking into consideration both lignocellulosic biomass, such as agricultural and forest residues; dedicated non-food crops, both herbaceous and woody as well as starchy biomass, from agricultural processing wastes. The main fraction of the vegetable wastes is represented by lignocellulosic residues, essentially consisting of cellulose, hemicellulose and lignin in variable proportions based on the plant source. The use of lignocellulosic thermophilic anaerobic microorganism isolated from buffalo-dung compost, and then as promoters of the production of the bacterium's enzymatic pool. Finally, the enzymes produced by this way were tested for the bioconversion of the hemicellulolytic fraction of two most used energy crops, i.e., Arundo donax (giant reed) and Cynara cardunculus (cardoon). Overall, the goal was to exploit agro-wastes as an alternative to the standard substrates available on the market at high cost (for example growth media comprising yeast extract, peptone, meat extract, and so on), currently used for bacterial biomass and related enzymes of biotechnological interest.

Waste Biomass
Waste biomass deriving from industrial processing of tomato (Lycopersicon esculentum variety "Hybrid Rome"), fennel (Foeniculum vulgare), potato (Solanum tuberosum) and carrot (Daucus carota) was kindly supplied from Fontanella Industry (Salerno, Italy), Fenny s.r.l. (Pagani, Salerno, Italy), Amato V. "Sammy Export" (Pagani, Salerno, Italy) and Tafuri Brothers for Polli Industry (Battipaglia, Salerno, Italy), respectively. Rhizomes of Arundo donax and stems and leaves of Cynara cardunculus were provided by the Agraria Department of "Federico II" University, Naples, and were obtained from an experimental plantation located at the Sele river plain of Campania Region (Southern Italy, 40 • 61 N, 14 • 92 E, 30 m a.s.l.). Vegetable wastes were dried under vacuum until reaching constant weight in order to avoid presence of residual water and minced to comparable particle size for further use.

Thermoanaerobacterium thermostercoris Culture Conditions
Pre-culture of Thermoanaerobacterium thermostercoris strain BUFF (ATCC BAA-1776 T = DSM 22141 T ) was performed at a temperature of 60 • C in 25 mL of broth "TN" dispensed into 120 mL serum bottles with agitation of 100 rpm in strict anaerobic conditions [22,23]. Culture medium contained (g/L): NaCl 10.0; KCl 0.1; MgCl 2 •H 2 O 0.2; NH 4 Cl 1.0; K 2 HPO 4 0.3; CaCl 2 •2H 2 O 0.1; cysteine-HCl 1.0; yeast extract 2.0; tryptone 2.0; glucose 5.0; resazurin 0.001. Before sterilization, the pH was adjusted at 6.5 with 1 M NaOH at room temperature. Routine growth was performed using a standard culture medium TN. After sterilization, the medium was also supplemented with 10 mL/L of both filter-sterilized vitamins and trace element solutions from DSM medium 141 [24]. The ability of the strain to utilize different carbon sources for growth and hemicellulolytic enzyme production was investigated in a 1 L glass reactor. Cultures were set up at the above reported temperature in 0.5 Lt of the anaerobic medium deprived of glucose and enriched at a final concentration of 5 g/L with one of the following compounds: xylose, arabinose or xylan or at a final concentration of 10 g/L with one of the following vegetable waste biomass: tomato, potato, carrot or fennel. The headspace was replaced with nitrogen gas. Growth was interrupted after 14 h (early stationary phase of growth) and evaluated by measuring the optical density at λ540 nm; in addition, the fermentation end products were compared. All experiments were carried out in duplicate.

Partial Enzyme Purification and Protein Content Determination
The 500 mL of culture broths were centrifuged at 10,000 rpm for 40 min, and the cell-free supernatants were precipitated with 80% ammonium sulphate and named "PP(NH 4 ) 2 SO 4 ". The crude enzymes were dissolved in 50 mM sodium acetate buffer pH 5.6 and dialyzed against the same buffer at a temperature of 4 • C by using Visking dialysis tubing with 14,000 Da cut-off. The protein concentration was determined by using a BioRad protein assay kit with bovine serum albumin as standard protein according to the Bradford method [25].

Enzymatic Assays
Xylanase activity was assayed by using 2% birchwood xylan as the substrate. The mixture, containing 0.5 mL of an appropriate dilution of the enzyme and 0.5 mL of the substrate, was incubated at 60 • C for 20 min. The quantity of reducing sugars liberated was determined using 3,5-dinitrosalicylic acid (DNS method) [26,27]. One unit of xylanase activity was expressed as µmol of reducing sugars (xylose equivalent) released in 1 min under the abovementioned conditions. β-xylosidase, α-arabinofuranosidase, β-glucosidase, α-galactosidase, β-galactosidase and β-fucosidase assays were based on the release of p-nitrophenol from 1 mM of p-nitrophenyl-β-D-xylopyranoside, p-nitrophenyl-α-Larabinofuranoside, p-nitrophenyl-β-D-glucopyranoside, p-nitrophenyl-α-D-galactopyranoside, p-nitrophenyl-β-D-galactopyranoside and p-nitrophenyl-β-D-fucopyranoside, respectively, using appropriately diluted enzyme solution at 60 • C for 20 min in 1 mL of 50 mM sodium acetate buffer pH 5.6 as final volume [28]. The acetyl-xylan esterase assay was based on the release of p-nitrophenol from 0.1 mM of p-nitrophenyl-acetate by the enzyme solution at 60 • C in 50 mM sodium acetate buffer at pH 5.6 for 20 min in 2 mL as final volume [29].

Biochemical Characterization of Enzyme Activities
The optimum pH for the enzyme activities was determined by assaying the enzymes in reaction mixtures under standard conditions by varying the pH from 3.0 to 10.0 using glycine-HCl buffer (pH 3.0), acetate buffer (pH 3.5-4.0-4.6-5.0-5.6), phosphate buffer (pH 6.0-6.5), Tris-HCl (pH 7.0) and glycine-NaOH buffer (pH 8.6-9.0-10.0) for each appropriate pH range. The optimum temperature was determined by assaying the enzyme activity in 50 mM acetate buffer, pH 5.6 at temperatures ranging from 40 to 90 • C. In both cases, the enzyme preparation from standard growth conditions of T. thermostercoris was employed. Thermal stability determination of xylanase, β-xylosidase, α-arabinofuranosidase and acetyl-xylan esterase was carried out by pre-incubating the samples in a temperature range 65-80 • C at various time intervals; after that the residual activities were determined under standard assay conditions and expressed as % with the respect to the same enzyme assayed without pre-incubation (100%).

Electrophoresis and Zymogram
Protein samples were analyzed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) using 10% acrylamide as resolving gel and 5% stacking gel according to the method of Laemmli [30]. The molecular mass was determined by using a molecular mass marker (Pharmacia). The proteins on the gel were visualized by staining with silver or Coomassie Brilliant Blue R-250 according to the manufacturer's instructions. The zymogram analysis was performed according to Martìnez [31]. In the case of xylanase, protein samples were denatured by heating at 70 • C for 10 min in sample buffer containing β-mercaptoethanol and then subjected to a normal 10% SDS-PAGE gel containing 0.5% (w/v) xylan. After electrophoresis, the gel was washed twice with 20% (v/v) isopropanol in 50 mM sodium citrate buffer (SCB) pH 5.4 for 30 min each at room temperature, followed by two washes of 30 min each in SCB (pH 5.4). The gel was incubated in SCB (pH 5.6) at 60 • C for 30 min for reaction and then stained with 0.1% (w/v) Congo red containing 5% (v/v) ethanol for 30 min, destained with 1 M NaCl (15 min, 2 times). The xylanase activity was visible as a clear band against a red background. The β-xylosidase and α-arabinofuranosidase activities in the gel were detected by developing a zymogram against 50 µg/mL 4-methylumbelliferyl-β-D-xyloside or 50 µg/mL 4-methylumbelliferylα-arabinofuranoside as substrate, respectively, prepared in 50 mM sodium citrate buffer (pH 5.6). After the electrophoretic run, the gel was immersed in substrate solution for 15 min at 70 • C in the dark. The enzyme bands in the gel were visualized under UV light at λ 365 nm. The acetyl-xylan esterase activity in the gel was revealed by incubating the gel, after the electrophoretic run, in 20% (v/v) isopropanol then in 50 mM sodium acetate buffer pH 5.6 and finally in 2 mM p-nitro-phenyl acetate. Yellow bands after 10 min of incubation appeared in the presence of the enzyme.

Ethanol and Hydrogen Production by T. thermostercoris
The cell-free supernatants were tested for ethanol content using the Megazyme K-ETOH kit according to the manufacturer's instructions. Hydrogen was detected and quantified by gas chromatography (Focus GC-Thermo Scientific Instrument, 3m Hayesep Q column) by injection of 1 mL of headspace. The temperatures of the thermal conductivity detector, injector and analytical column were 200, 50 and 50 • C, respectively. Nitrogen was used as a carrier gas at a flow rate of 20 mL/min. Hydrogen was quantified using an external calibration curve and expressed as mL of H 2 produced from 1 L of culture in 1 h.

Enzymatic Degradation of Hemicellulolytic Extracts from Cynara Cardunculus and Arundo Donax Biomass
Next, 1 g of dry stem and leaves from Cynara cardunculus (C) and 1 g of dry rhizomes from Arundo donax (A) were treated for hemicellulosic fraction extraction. For this purpose, a previously reported procedure [13,32] was applied, and almost 17.6% and 17% (w/w), respectively, of the dried biomass was recovered. The lyophilized material, from C. cardunculus (CK) and from A. donax (AK), respectively, was used as substrate for bioconversion reactions. Enzymatic digestion (2 mL) of 8 mg of hemicellulosic material extracted from C. cardunculus (CK) and A. donax (AK), solubilized in 50 mM sodium acetate buffer (pH 5.6), was carried out by incubating the partially purified samples with ammonium sulphate enzymatic solution obtained from fennel growth (4.86 mg of total proteins) for 24 h at 60 • C. All enzymatic reactions were performed in duplicate, in sealed glass vessels, under magnetic stirring and stopped by cooling in an ice bath. The results of hydrolysis reactions were estimated through the reducing sugar assay [26], evaluating the optical density at λ546 nm; the contribution of unhydrolyzed substrate was subtracted from the values obtained after 24 h of incubation. The reaction products were further analyzed by thin layer chromatography (TLC) in solvent, n-BuOH/AcOH/H 2 O, 6:2:2 (v/v), to determine the mono-and oligosaccharidic composition of reaction products.

T. thermostercoris Growth Test on Different Substrates for Hydrogen and Ethanol Production
The production costs for microbial biomass, enzymes and other metabolites represent the great obstacle to the success of a possible application in the industrial field, and therefore the optimization of the growth conditions are important and crucial in defining the optimal production while limiting costs. With this purpose T. thermostercoris strain BUFF, a nonspore-forming, thermophilic and anaerobic Gram-positive staining microorganism, was grown on different substrates including agriculture wastes as reported in the Materials and Methods section. In Table 1, the parameters related to absorbance, production of hydrogen and ethanol after 14 h of growth in a glass bioreactor are reported ( Figure S1). A higher yield in terms of growth (O.D. λ540 nm), compared to the standard culture medium containing glucose (TN+ glucose) or the other added sugars, occurred when the microorganism grew in the media enriched with carrot, fennel and tomato wastes. On the contrary, potato waste did not support growth well. Moreover, the greatest production of hydrogen took place when the bacterium was grown on fennel waste (35 mL/L/h H 2 ) followed by the production carried out on carrot waste and xylan. In the case of ethanol production, the maximum yield was found for cell growth conducted on carrot waste followed by growth on xylan. These results showed the ability of Thermoanaerobacterium thermostercoris strain BUFF to produce hydrogen and ethanol when grown by using agro-industrial vegetable wastes as main carbon sources. The production of hydrogen through biological processes represents a new area of biotechnology development for the production of bioenergy. The cost-effective production of biofuels by bacterial fermentation on economically competitive substrates such as agro-industrial waste for the improvement of hydrogen yields could become commercially competitive. In addition, biological conversion of biomass into ethanol and hydrogen, through the combination of hemicellulolytic enzyme production, hydrolysis of hemicellulose and fermentation in a single step, could provide new insights into a lower cost process [33,34]. Utilization of the waste biomass for production of value-added products through biological intervention not only helps to combat environmental pollution but also contributes significantly to the economy.

Enzyme Production of T. thermostercoris in Selected Growth Media
The presence of hemicellulolytic enzyme activities was investigated in all cell-free supernatants partially purified after the addition of (NH 4 ) 2 SO 4 at 80% saturation. The enzyme assays, conducted as reported in the Materials and Methods section, were expressed as specific activity (U/mg protein). In Figure 1 it is possible to observe that the production of xylanase enzyme was higher when the strain BUFF was grown on xylose with a specific activity of 8 U/mg. Moreover, the specific activity of xylanase, when compared to standard medium containing glucose (1.4 U/mg), was found to be higher on the waste of tomato (2.1 U/mg) and carrot (1.8 U/mg). Similarly for beta-xylosidase the greatest production was obtained from the growth conducted on xylose with 1158 U/mg, while from the analysis of the growth on vegetable waste compared to the growth on glucose (266 U/mg), the results showed good yields on tomato (405 U/mg), fennel (284 U/mg) and carrot (274 U/mg) wastes ( Figure 2). Similarly for beta-xylosidase the greatest production was obtained from the growth conducted on xylose with 1158 U/mg, while from the analysis of the growth on vegetable waste compared to the growth on glucose (266 U/mg), the results showed good yields on tomato (405 U/mg), fennel (284 U/mg) and carrot (274 U/mg) wastes ( Figure 2). Similarly for beta-xylosidase the greatest production was obtained from the growth conducted on xylose with 1158 U/mg, while from the analysis of the growth on vegetable waste compared to the growth on glucose (266 U/mg), the results showed good yields on tomato (405 U/mg), fennel (284 U/mg) and carrot (274 U/mg) wastes ( Figure 2).

Figure 2.
Beta-xylosidase specific activity produced by strain BUFF on TN medium containing the indicated carbon source or biomass.
In the case of the alpha-arabinofuranosidase enzyme, the higher enzymatic production was found in the standard culture medium with glucose (4092 U/mg) as carbon source, followed by growth on xylose (1931 U/mg) and arabinose (1263 U/mg); taking into account vegetable wastes, the production of the enzyme was found to be higher in culture media enriched with potato waste (882 U/mg) (Figure 3). In the case of the alpha-arabinofuranosidase enzyme, the higher enzymatic production was found in the standard culture medium with glucose (4092 U/mg) as carbon source, followed by growth on xylose (1931 U/mg) and arabinose (1263 U/mg); taking into account vegetable wastes, the production of the enzyme was found to be higher in culture media enriched with potato waste (882 U/mg) (Figure 3).  Interestingly, production of acetyl-xylan esterase was detected in the case of the T. thermostercoris strain BUFF grown on xylose as predominant carbon source reaching a specific activity of 11,961 U/mg; among the vegetable waste tested, fennel and potato wastes showed a higher enzyme production with the respect to glucose (4302 U/mg), with specific activity values of 6019 and 5411 U/mg, respectively. The waste of the tomato in the growth media, on the other hand, was found to inhibit this enzymatic activity (Figure 4). Interestingly, production of acetyl-xylan esterase was detected in the case of the T. thermostercoris strain BUFF grown on xylose as predominant carbon source reaching a specific activity of 11,961 U/mg; among the vegetable waste tested, fennel and potato wastes showed a higher enzyme production with the respect to glucose (4302 U/mg), with specific activity values of 6019 and 5411 U/mg, respectively. The waste of the tomato in the growth media, on the other hand, was found to inhibit this enzymatic activity ( Figure 4).
As regards alpha-galactosidase enzyme, it was found that the greatest production was observed in microbial growth enriched with tomato waste (1194 U/mg) followed by growth on fennel waste (587 U/mg) ( Figure 5). thermostercoris strain BUFF grown on xylose as predominant carbon source reaching a specific activity of 11,961 U/mg; among the vegetable waste tested, fennel and potato wastes showed a higher enzyme production with the respect to glucose (4302 U/mg), with specific activity values of 6019 and 5411 U/mg, respectively. The waste of the tomato in the growth media, on the other hand, was found to inhibit this enzymatic activity (Figure 4). As regards alpha-galactosidase enzyme, it was found that the greatest production was observed in microbial growth enriched with tomato waste (1194 U/mg) followed by growth on fennel waste (587 U/mg) ( Figure 5). For the beta-galactosidase activity, bacterial growth on the medium containing arabinose substrate produced the greatest amount of the enzyme with a specific activity of 3800 U/mg; all vegetable wastes tested, compared to the standard growth conducted on glucose (167 U/mg), produced beta-galactosidase to a greater extent. In the case of tomato, fennel, carrot and potato wastes, the specific activity recorded was 3773, 1139, 417 and 322 U/mg, respectively ( Figure 6). For the beta-galactosidase activity, bacterial growth on the medium containing arabinose substrate produced the greatest amount of the enzyme with a specific activity of 3800 U/mg; all vegetable wastes tested, compared to the standard growth conducted on glucose (167 U/mg), produced beta-galactosidase to a greater extent. In the case of tomato, fennel, carrot and potato wastes, the specific activity recorded was 3773, 1139, 417 and 322 U/mg, respectively ( Figure 6).
The greatest production of beta-fucosidase enzyme occurred in the culture medium enriched with vegetable waste from the tomato, showing a specific activity of 5725 U/mg, followed by the growth on arabinose as substrate (4886 U/mg). The other vegetable wastes tested, compared to the standard growth on glucose (216 U/mg), produced comparable levels of beta-fucosidase ranging from 476 U/mg in the case of potato up to 652 U/mg in the case of fennel waste (Figure 7).
Finally, the beta-glucosidase activity was mainly produced when the microorganism was grown in the presence of arabinose as main substrate (15,955 U/mg); among vegetable wastes analyzed, in the case of tomato and potato wastes, specific activity of 2576 and 2714 U/mg were registered, respectively, compared to growth on glucose (826 U/mg) (Figure 8).
For the beta-galactosidase activity, bacterial growth on the medium containing arabinose substrate produced the greatest amount of the enzyme with a specific activity of 3800 U/mg; all vegetable wastes tested, compared to the standard growth conducted on glucose (167 U/mg), produced beta-galactosidase to a greater extent. In the case of tomato, fennel, carrot and potato wastes, the specific activity recorded was 3773, 1139, 417 and 322 U/mg, respectively ( Figure 6).

Figure 6.
Beta-galactosidase specific activity produced by strain BUFF on TN medium containing the indicated carbon source or biomass.
The greatest production of beta-fucosidase enzyme occurred in the culture medium enriched with vegetable waste from the tomato, showing a specific activity of 5725 U/mg, followed by the growth on arabinose as substrate (4886 U/mg). The other vegetable wastes tested, compared to the standard growth on glucose (216 U/mg), produced comparable levels of beta-fucosidase ranging from 476 U/mg in the case of potato up to 652 U/mg in the case of fennel waste (Figure 7). Finally, the beta-glucosidase activity was mainly produced when the microorganism was grown in the presence of arabinose as main substrate (15,955 U/mg); among vegetable wastes analyzed, in the case of tomato and potato wastes, specific activity of 2576 and 2714 U/mg were registered, respectively, compared to growth on glucose (826 U/mg) (Figure 8).  Finally, the beta-glucosidase activity was mainly produced when the microorganism was grown in the presence of arabinose as main substrate (15,955 U/mg); among vegetable wastes analyzed, in the case of tomato and potato wastes, specific activity of 2576 and 2714 U/mg were registered, respectively, compared to growth on glucose (826 U/mg) (Figure 8). The results showed that the composition of the biomass present in the growth media (vegetable biomass and sugars) influenced the production of the enzymatic activities. Indeed, carrot residues are composed mainly of pectin/hemicellulose/cellulose in the The results showed that the composition of the biomass present in the growth media (vegetable biomass and sugars) influenced the production of the enzymatic activities. Indeed, carrot residues are composed mainly of pectin/hemicellulose/cellulose in the following ratio: 3.88:12.3:51.6 [35], while potato residues are composed mainly of starch/hemicellulose/cellulose according to the following ratio 67.5:9.2:1.2 [36]. Tomato waste's polysaccharide fraction is mainly composed of xyloglucan [37]; a similar composition has been found for fennel waste's polysaccharide fraction [10]. The higher production of xylanase, beta-xylosidase and acetyl-xylan esterase obtained with tomato and fennel wastes can be correlated with the chemical composition of the wastes (as reported above). In the case of the low acetyl-xylan esterase activity it could be argued that its expression depends upon the acetylation degree of the polysaccharide. As reported in the literature, "the degree of O-acetylation can vary depending on species, tissue type, and developmental state" of the plant source; therefore, it can be hypothesized that the obtained results are due to different degrees of acetylation in the polysaccharide fractions.
In Table 2, production of the enzymatic activities is summarized, and bold indicates the enzyme activities obtained from microbial growth carried out on agro-industrial wastes, with values greater than enzyme activity recorded on glucose medium (standard growth condition). These results indicated the importance of using different biomass sources to encourage the production of specific enzymes in a sustainable way by exploiting surplus biomass, disposal of which affects the costs faced by companies. Table 2. Enzymatic production of the microorganism T. thermostercoris strain BUFF grown on culture media enriched with agro-industrial wastes in comparison with standard growth medium (glucose). One unit of enzyme was defined as the activity releasing 1 µmol of p-nitrophenol in 20 min at 60 • C. In order to study the potential industrial applications of these enzymes, biochemical characterization is an indispensable step for biotechnological purposes.

Enzyme Activity (U/mg) Glucose
Therefore, to identify the optimal chemical-physical parameters of the enzymatic activities produced by the microorganism T. thermostercoris strain BUFF, activity tests were carried out at different pH and temperatures values under standard conditions; in addition, the thermal stability assay and the determination of molecular weights, as described in the Materials and Methods section, were also conducted.
In Table 3 the optimal temperature and pH values, in addition to estimated molecular weights from SDS-PAGE analysis, are also reported. In most cases, the optimal temperature for the studied enzymes ranged from 60 to 80 • C. Regarding the optimal pH value, in the case of acetyl-xylan esterase it was found in acidic medium (3.5); β-xylosidase and α-arabinofuranosidase were found to be more active at a pH value of 5.6, while xylanase and β-glucosidase preferred higher pH values of 6.0 and 6.5, respectively. Table 3. Main biochemical properties of enzymatic activities produced by T. thermostercoris strain BUFF grown on standard glucose growth media. Most Thermoanaerobacterium strains possess xylanase activity [38]. Xylanase activity of T. thermostercoris strain BUFF was comparable to those previously reported in T. thermosaccharolyticum strain NOI-1 and T. saccharolyticum strain JW/SL-YS 485, of which the crude enzyme exhibited 2.55 and 0.36 U/mg xylanase activity, respectively [39,40]. Furthermore, the crude enzymes of strain BUFF exhibited up to 100-fold greater xylanase activity than that reported in T. xylanolyticum strain B6A-RI (0.065 U/mg) and T. brockii strain HTD4 (0.003 U/ mg) [38]. In the genus Thermoanaerobacterium, strains equipped with multi-enzymatic systems capable of digesting lignocellulose are also described; a thermophilic anaerobic bacterium, Thermoanaerobacterium thermosaccharolyticum strain NOI-1 when grown on xylan showed a multienzyme complex consisting of xylanase, β-xylosidase, α-L-arabinofuranosidase, β-glucosidase and cellobiohydrolase [40,41].

Enzymes
In order to assess the thermal stability of the enzymes, the extracellular PP (NH4)2SO4 samples of xylanase, β-xylosidase, α-arabinofuranosidase and acetyl-xylan esterase, obtained from T. thermostercoris grown on standard glucose growth media and from fennel waste medium, were pre-incubated at temperature range 65-80 °C at various Most Thermoanaerobacterium strains possess xylanase activity [38]. Xylanase activity of T. thermostercoris strain BUFF was comparable to those previously reported in T. thermosaccharolyticum strain NOI-1 and T. saccharolyticum strain JW/SL-YS 485, of which the crude enzyme exhibited 2.55 and 0.36 U/mg xylanase activity, respectively [39,40]. Furthermore, the crude enzymes of strain BUFF exhibited up to 100-fold greater xylanase activity than that reported in T. xylanolyticum strain B6A-RI (0.065 U/mg) and T. brockii strain HTD4 (0.003 U/ mg) [38]. In the genus Thermoanaerobacterium, strains equipped with multi-enzymatic systems capable of digesting lignocellulose are also described; a thermophilic anaerobic bacterium, Thermoanaerobacterium thermosaccharolyticum strain NOI-1 when grown on xylan showed a multienzyme complex consisting of xylanase, β-xylosidase, α-L-arabinofuranosidase, β-glucosidase and cellobiohydrolase [40,41].
In order to assess the thermal stability of the enzymes, the extracellular PP (NH 4 ) 2 SO 4 samples of xylanase, β-xylosidase, α-arabinofuranosidase and acetyl-xylan esterase, obtained from T. thermostercoris grown on standard glucose growth media and from fennel waste medium, were pre-incubated at temperature range 65-80 • C at various time intervals. Subsequently, the samples were tested under standard conditions and the percentage of residual activity, compared to the non-pre-incubated ones, was calculated. As shown in Table 4, in the case of extracellular enzyme preparation from glucose medium, the half-life of xylanase was reached after 4 h of pre-incubation at a temperature of 75 • C, while for the β-xylosidase and α-arabinofuranosidase activities after 60 min of pre-incubation the half-life was reached at 70 and 80 • C, respectively. The acetyl-xylan esterase activity was less thermostable, as 50% of activity was reached only after 10 min of incubation at 80 • C ( Table 4). In the case of (PP(NH 4 ) 2 SO 4 ) extracellular enzyme preparation from fennel waste medium, for xylanase and α-arabinofuranosidase a 50% of activity was reached after 240 min at a temperature of 75 • C and 60 min at a temperature of 80 • C, respectively; for β-xylosidase and acetyl-xylanesterase, residual activity of 51% and 44% was reached after 60 and 10 min of incubation at a temperature of 75 and 80 • C, respectively.

Bioconversion of Hemicellulose Fractions from Arundo Donax and Cynara Cardunculus
The partially purified (PP(NH 4 ) 2 SO 4 ) extracellular enzyme of T. thermostercoris strain BUFF, produced by using fennel wastes for cell growth, was used to hydrolyze the hemicellulosic fraction deriving from the maceration in KOH 2N for 72 h of the rhizome of A. donax and of the leaves and stems of C. cardunculus. The reaction mixtures, after incubation, were tested for the production of reducing sugars, and the hydrolysis yields were expressed as a percentage with respect to the initial quantity present in the substrates. In the case of the hemicellulosic fraction of A. donax, 48% bioconversion occurred, while for C. cardunculus it was 65%. The enzymatic pool of the anaerobic bacterium T. thermostercoris strain BUFF was able to degrade the hemicellulosic component of the tested substrates (Table 5). Subsequently, the reaction mixtures were analyzed by thin layer chromatography (TLC). An increase in the hydrolysis products between time zero and final time (after 24 h of incubation) was evident ( Figure S2); this difference was marked above all in the reaction mixture in which the polymer fractions were used as substrate. The most abundant hydrolyses products were xylose, xylobiose and xylotriose. Most of the microbially investigated β-xylosidases showed a strong xylose inhibition [42]. Conversely, T. ther-mostercoris strain BUFF β-xylosidase produced xylose displaying low substrate inhibition according to some recently investigated β-xylosidase activities [43] suggesting a possible use of T. thermostercoris strain BUFF β-xylosidase for the industrial sector. Degradation of lignocellulosic agro-industrial residues by means of complex hemicellulolytic enzymes is a promising approach providing efficient biomass decomposition for subsequent conversion to value-added products [44].

Conclusions
The anaerobic thermophile Thermoanaerobacterium thermostercoris strain BUFF is an interesting microbial cell factory for its capability to produce industrially relevant bioproducts with great economic value, especially in the green chemistry field. One main issue in this field is represented by the sustainable production of biofuels, for example bioethanol from lignocellulosic biomass. This work demonstrated that the sustainable production of T. thermostercoris cell biomass and its cellulolytic enzymes could be achieved by using residual biomass from the agroindustry sector. Moreover, we also found that T. thermostercoris could directly utilize raw vegetable wastes (such as energy crops' residues) for biohydrogen and bioethanol production. Experimental data revealed a powerful enzymatic pool comprising thermophilic xylanase, β-xylosidase, α-arabinofuranosidase, β-glucosidase, α-galactosidase, β-galactosidase and β-fucosidase activities suitable to a wide pH range that could serve as an excellent green factory for bioprocesses, in which value-added by-products are produced from food wastes. These enzymes proved to be very effective for the bioconversion of the hemicellulose fraction from dedicated crop biomass into xylooligomers and fermentable sugars. Therefore, they give this microorganism great potential in biotechnological fields as a catalyst for bioprocessing purposes. Nevertheless, there are still many opportunities for increasing the scientific value of this microbial system as a cell factory by means of exploring its genome as well as its high expression of enzymatic activities through cloning.