Hemicellulosic biomass conversion by Moroccan hot spring Bacillus paralicheniformis CCMM B940 evidenced by glycoside hydrolase activities and whole genome sequencing

Thermophilic bacteria, especially from the genus Bacillus, constitute a huge potential source of novel enzymes that could be relevant for biotechnological applications. In this work, we described the cellulose and hemicellulose-related enzymatic activities of the hot spring Bacillus aerius CCMM B940 from the Moroccan Coordinated Collections of Microorganisms (CCMM), and revealed its potential for hemicellulosic biomass utilization. Indeed, B940 was able to degrade complex polysaccharides such as xylan and lichenan and exhibited activity towards carboxymethylcellulose. The strain was also able to grow on agriculture waste such as orange and apple peels as the sole carbon source. Whole-genome sequencing allowed the reclassification of CCMM B940 previously known as B. aerius into Bacillus paralicheniformis since the former species name has been rejected. The draft genome reported here is composed of 38 contigs resulting in a genome of 4,315,004 bp and an average G + C content of 45.87%, and is an important resource for illuminating the molecular mechanisms of carbohydrate metabolism. The annotated genomic sequences evidenced more than 52 genes encoding glycoside hydrolases and pectate lyases belonging to 27 different families of CAZymes that are involved in the degradation of plant cell wall carbohydrates. Genomic predictions in addition to in vitro experiments have revealed broad hydrolytic capabilities of the strain, thus reinforcing its relevance for biotechnology applications. Supplementary Information The online version contains supplementary material available at 10.1007/s13205-021-02919-0.


Genome research reports
Lignocellulosic biomass, mainly composed of cellulose, hemicelluloses and lignin, is the most abundant complex biopolymer in the nature. It is of major interest for biotechnology as this feedstock is used for the production of renewable biofuels and other high-value chemicals, through an enzymatic saccharification process (Zhang et al. 2018). However, depolymerization of lignocellulose is challenging and it requires high-performance enzymes that act in synergy (Houfani et al. 2020). Therefore, enzymes produced by microorganisms, such as bacteria, are used for the deconstruction of the complex polysaccharides of cellulose and hemicelluloses which are tightly bound together in plant cell walls (Zeng et al. 2017).
Xylan is the main component of hemicelluloses and consists of a β-1,4-d-xylan backbone with short side chains of O-acetyl, β-l-arabinofuranosyl, d-α-glucuronic acid and phenolic acid (Biely et al. 2016). Other polysaccharides such as xyloglucan, lichenan and mannans are part of the hemicellulolytic fraction of plant cell walls. Xylanases, xyloglucanases, lichenases, and endoglucanases are necessary to completely degrade these polysaccharides into simple sugars to allow glycolytic fermentation to produce energy. These enzymes belong to one of the glycoside hydrolase (GH) families, a subgroup of the carbohydrate-active enzymes (CAZymes) (Lombard et al. 2014) and very often associated with carbohydrate-binding modules (CBM) that facilitate adhesion of enzymes to cellulose or hemicelluloses.
Most microorganisms living in extreme environmental habitats such as hot springs have been considered attractive producers of lignocellulolytic enzymes for industrial bioconversion processes (Thapa et al. 2020). Morocco has more than 20 hot springs distributed in different regions (Cidu and Bahaj 2000;Bouchaou et al. 2017) which are suitable habitats for thermophilic microorganisms growing optimally between 55 and 80 °C. In addition, thermophilic bacteria are known for their ability to produce thermostable enzymes of biotechnological interest (Knapik et al. 2019). Previously, Aanniz et al. (2015) investigated the diversity of thermophilic bacteria in 4 different hot springs in Morocco, which enabled the isolation of 79 bacterial strains belonging to the genus Bacillus (www. ccmm. ma). Some of these strains were identified as Bacillus aerius. They grow optimally at 55 °C and exhibit amylolytic, proteolytic or cellulolytic activity (Aanniz et al. 2015). However, strains belonging to B. aerius species should be reclassified as this species name has been rejected since 2015 (Dunlap 2015). For this reason, we chose the CCMM B940 strain as the representative of the B. aerius group isolated by Aanniz et al. (2015) from a hot spring for biological characterization of its potential hemicellulolytic and cellulolytic activities and whole-genome sequencing analysis.
B940 strain was grown twice on Luria-Bertani (LB) agar medium (BD Difco, Sparks, Maryland, USA) for 18 h at 37 °C under aerobic conditions. Individual colonies were then picked from LB plates and cultivated at 37 °C for 18 h in LB agar medium supplemented with 0.5% Icelandic moss Lichenan (Megazyme, Wicklow, Irlande), beechwood Xylan (Megazyme, Wicklow, Irlande), or medium viscosity Carboxymethylcellulose (CMC, Sigma, MO, USA), respectively. A second bacterial strain (CCMM B945) from the same B. aerius group, known for inability to degrade lichenan and xylan (data not shown) was used as a negative control for degradation of hemicellulose and following the same experimental protocol as for CCMM B940. Both strains grew well on LB agar medium supplemented with purified fibres (Fig. 1A).
To reveal whether the strains were able to degrade the complex polysaccharides added into the medium, the plates were stained with 0.1% Congo red as previously described (Patrascu et al. 2017). Clear halos around the streaks indicated that B940 strain can degrade all three polysaccharides, and thus capable of producing the appropriate enzymes. As expected, lichenases and xylanases were only produced by the CCMM B940 strain (Fig. 1B). Strain CCMM B945 exhibited significant activity against CMC, indicating production of endoglucanases, similar to the CCMM B940 strain.
The Congo red test demonstrated that live cells of the B940 strain exhibited lichenase and xylanase activities and also produced endoglucanases for glucan-polymer degradation. However, to evaluate biomass utilization potential of this strain for biotechnology applications, we searched for the presence of those enzymes in the rather easily accessible B940 strain supernatants. For this purpose, the strain was cultivated in the basal salt solution (BSS), a chemically defined medium modified from Parab et al. (2017), with a lower salt concentration (NaCl 6 g L −1 ). The medium was supplemented with fruits peels (orange or apple) as the sole source of carbon and then sterilized 15 min at 121 °C. Fruit peels are a rich source of cellulose, hemicellulose, lignin and pectin (Rivas et al. 2008;Joglekar et al. 2019). Orange peel is rich in structural polysaccharides, including cellulose (18-20%), hemicellulose (14-16%), pectin (20-22%) and lignin (5-7%) (de la Torre et al. 2019). Apple residues typically contain 7.2-43.6% cellulose, 4.3-24.4% hemicellulose, 15.2-23.5% lignin and 3.5-14.32% pectin (Dhillon et al. 2013). Both substrates were chosen because of their abundance as industrial waste around the world (de la Torre et al. 2019).
A single colony of CCMM B940 and CCMM B945 from LB agar medium was picked to inoculate 5 mL of LB Broth (Difco) and incubated overnight at 37 °C (with continuous stirring at 170 rpm), followed by second culture in 5 mL of BSS medium supplemented with 0.5% apple or orange peels and inoculated at 1%. Before adding them in the medium, apple and orange peels were dried for 24 h at 60 °C and crumbled finely using an electric grinder. After 48 h of incubation at 37 °C with stirring (170 rpm), both strains grew well in the BSS supplemented with fruit peels representing agro-industrial wastes as shown in Figure S1. Additionally, this test confirmed the ability of CCMM B940 and CCMM B945 strains to break down fibres from the peels and use the degradation products.
Protein samples from B940 and B945 strains cultivated for 48 h at 37 °C in 30 mL BSS medium supplemented with 0.5% orange and apple peels were prepared to detect and quantify hemicellulolytic and cellulolytic activities in the extracellular fluids. Briefly, cultures were centrifuged for 15 min at 5000×g and 4 °C, supernatants were filtered through a 0.22 μm Millipore filter, and then concentrated 25-fold in a Spin-X UF10K concentrator following the supplier's protocol (Corning B.V., Amsterdam, The Netherlands).
Agar-well plate enzymatic assays were performed to detect xylanase, lichenase and endoglucanase activities using concentrated extracellular proteins. Concentrated extracellular proteins from CCMM B940 and control strain CCMM B945 were prepared in the same way. The samples were loaded (60 μL/well) in Petri dishes containing BSS medium supplemented with either 0.1% chromogenic AZCL-Xylan (Megazyme, Wicklow, Ireland), 0.5% non-chromogenic CMC (Sigma-Aldrich, MO, USA) or lichenan (Megazyme) polysaccharides. The plates were incubated overnight at 37 °C and those with non-chromogenic polysaccharide were stained with Congo red as described above. Clear and blue halos around the wells indicated that the CCMM B940 cultivated with apple peels was able to produce lichenase ( Fig. 2A), endoglucanase (Fig. 2C, although weakly in the tested conditions), and xylanase (Fig. 2B). The same results were obtained with orange peels (not shown). The CCMM B940 strain is able to grow at 55 °C in Tryptic Soy broth (Aanniz et al. 2015) and we also observed that the strain is capable of growing in BSS medium supplemented with lichenan and xylan at the same temperature (data not shown), thus reinforcing the interest of its characteristics for industrial applications requiring thermostable enzymes.
In contrast to CCMM B940 strain results, the control strain CCMM B945 was not able to produce xylanase or lichenase when cultivated in the presence of apple or orange peels (Fig. 2).
Moreover, extracellular hemicellulolytic activities of the strain CCMM B940 were quantified using the dinitrosalicylic acid (DNS) spectrophotometric method for reducing sugars as described by Miller (1959). Protein concentration was determined by the Bradford protocol (Bradford 1976) with bovine serum albumin (BSA) as the standard. The amount of protein samples and the incubation time were chosen to provide assay conditions in which the measured activity was proportional to these two parameters. The enzymatic assays were all carried out at 55 °C to quantify the activities under conditions close to those of industrial processes using thermostable enzymes. Reducing sugars that were released after enzyme-substrate incubations were quantified in triplicates using glucose or xylose as a standard. One international unit (IU) was defined as the amount of enzyme, which produced 1 µmol of reducing sugars in 1 min. The xylanase and lichenase activities of the extracellular proteins from Fig. 2 Detection of lichenase A, xylanase B and carboxymethylcellulase C activities of Bacillus CCMM B940 and B945 strains, using agar-well plate assays with concentrated extracellular proteins of BSS cultures with 0.5% apple. Agar plate BSS medium was supplemented with 0.5% (w/v) lichenan, carboxymethylcellulose (CMC), or 0.1% AZCL beechwood xylans. Plates were incubated overnight at 37 °C. Clear and blue halos around the wells indicate a positive sample for the corresponding glycoside hydrolase activity No reducing sugars was detectable for strain CCMM B945 proteins incubated with xylan or lichenan. The lichenase activity of CCMM B940 was approximately two times higher than the xylanase activity in the cultures with natural peels. Both activities were higher in apple peel cultures than in orange peels, suggesting that the produced enzymes were either more numerous or more active in metabolizing carbohydrates in apple peels. This could be due to the nature of fibres present in the peel (sugar composition and types of glycosidic bonds) and/or to differential induction of glycoside hydrolases encoding gene expression. Strain CCMM B940 exhibited xylanase activities approximately twofold and over tenfold higher in apple peel cultures than thermophilic strains B. licheniformis 2D55 (Kazeem et al. 2017) and Bacillus sp. 275 (Gong et al. 2017), respectively. To further investigate the putative functions of the CCMM B940 strain and to achieve its reclassification within the genus Bacillus, we sequenced and examined the whole genome of this strain. Thus, genomic DNA was extracted with Wizard ® Genomic DNA Purification Kit (Promega, Charbonnières-les-Bains, France). The concentration and quality were determined using Nanodrop and Qubit spectrophotometers. Genomic DNA was sequenced at Eurofins Genomics (Cologne, Germany) using Illumina HiSeq technology. The library was prepared with 2 × 150-bp paired-end read length, including DNA fragmentation, adapter ligation, amplification, and size selection. All the steps were performed according to Eurofins Genomics protocols, producing 7,612,190 reads.
After eliminating low coverage and small contigs (< 350 bp), we obtained a total of 38 contigs and 34 scaffolds resulting in a genome of 4,315,004 bp (NCBI accession number JADRJL000000000) and an average G + C content of 45.87% which is close to other genomes from the Bacillus subtilis group (Rey et al. 2004;Du et al. 2019).The N50 of the draft genome is 440,121 bp, with an average of 126,912 bp and a notable four gap region (Fig. 3). Species identity was confirmed by comparing the genome reference sequences of B. licheniformis and Bacillus paralicheniformis available in the NCBI database (https:// www. ncbi. nlm. nih. gov) using the BLAST service deployed on local servers with the genome obtained sequences and fastANI (Jain et al. 2018). The draft genome was most closely related to B. paralicheniformis Bac84 (accession number ASM299392v1) living in sea environment (Othoum et al. 2019) with an ANI value of 98.94%, and shared only 94.36% identity with B. licheniformis DSM 13 T (NC_006270.3). B. aerius CCMM B940 was, therefore, reclassified as B. paralicheniformis CCMM B940, placing it in the B. subtilis group; a group known for its wide array of uses in biotechnology mainly through species like B. subtilis and B. licheniformis (Fan et al. 2017). Taxonomic identity of the CCMM B940 strain was also confirmed using the GTDB-Tk tool for genome classification (Chaumeil et al. 2020) with 98.95% identity to the B. paralicheniformis KJ-16 genome (accession number ASM104248v2). The genome was then annotated using the default parameters of Prokka 1.14 (Seemann 2014) showing 4259 protein-coding genes, 51 tRNAs and 4 complete rRNAs.
We further used eggNOG-mapper v2.0.1 (Huerta-Cepas et al. 2017) to assign proteins sequences into functional categories. More than 76% of the predicted proteins were attributed to functional subsystems. Proportions of genes with known functions (3047 CDS) were assigned to each category as presented in Fig. 4. Interestingly, a high proportion of CCMM B940 genes was assigned to the carbohydrate transport and metabolism (11.3%) as well as aminoacid transport and metabolism (12.5%) proteins. In a study conducted on Bacillus velenzis, Bacillus safensis and Bacillus altitudinis genomes, only 6.1% of the predicted proteins were assigned to carbohydrate transport and metabolism (Datta et al. 2020), while in Bacillus amyloliquefaciens and Bacillus siamensis, it represents a maximum of 8% in the pan and the core genomes (Chun et al. 2019). B. paralicheniformis has not been thoroughly studied for carbohydrate metabolism yet, but is phylogenetically very close to the B.    licheniformis that has been widely used in the fermentation industry for production of enzymes, antibiotics and other chemicals (Madslien et al. 2012).
Hereafter, we focused on CCMM B940 glycoside hydrolases related to hemicellulose, cellulose and pectin degradation, comparing their protein sequences to the sequences of GH and pectate lyase (PL) of seven strains of B. paralicheniformis (A4-3, ATCC 9945a, BL-09, CBMAI 1303, FA6, MDJK30, ZAP17) available in the CAZY database (Lombard et al. 2014; database release 22 November 2020). Blast results was obtained using default parameters. Best hits with almost 90% amino-acid identity and 50% length alignment were selected and illustrated in Table 2 (the corresponding sequences are provided in Fig S2). Among 74 GH and PL sequences from CCMM B940 showed homology to other strains of B. paralicheniformis, 52 are dedicated to the degradation of carbohydrates in plant cell walls, representing 27 different CAZymes families according to the classification of B. Henrissat (Lombard et al. 2014). Thirty-two sequences of GH1, GH3, GH5, GH8, GH11, GH16, GH26, GH30, GH31, GH32, GH42, GH43, GH51, GH53, GH68, GH73 were linked to the degradation of hemicellulose, 4 sequences of GH5, GH9, GH12, GH48 for cellulose metabolism, 1 sequence (GH5) either for hemicellulose or cellulose degradation, and 11 sequences of GH (GH28, GH105) and PL (PL1, PL3, PL9, PL11, PL2) were attributed to pectin degradation (Table 2). Four sequences corresponded to enzymes (GH1, GH4) involved in glycolytic pathways. The remaining sequences were mainly related to the hydrolysis of starch and peptidoglycan (not shown). The accession numbers correspond to the CDS sequences annotated with Prokka as indicated in the manuscript. The corresponding sequences are provided in Figure S2 These predictions first confirmed the experimental observations in this study for the degradation of xylan, lichenan and CMC as the CCMM B940 genome carries glycoside hydrolases from GH5, GH9, GH11 and GH16 families of CAZYmes known to target these substrates. They also provided additional information on the carbohydrate metabolism of CCMM B940 such as pectinase activity as well as the presence of carbohydrate-binding modules (CBM3, CBM6, CBM66) associated with GHs helping them to bind to cellulose or hemicellulose polysaccharides to improve their efficiency. A few other studies (Wang et al. 2017;Chen et al. 2018) examined the genomic sequences of B. paralicheniformis and predicted their CAZymes, but without the experimental validation. To our knowledge, this is the first report that combines in silico genomic analysis and in vitro experiments to study the carbohydrate metabolism of B. paralicheniformis. These combined analyses reinforce the practical and potential use of CCMM B940 strain for industrial purposes. Indeed, Bacillus species are known to produce 60% of commercially available enzymes, mostly being homologous proteins that are naturally secreted in the growth medium (Westers et al. 2004). The strain CCMM B940 exhibits cellulolytic and hemicellulolytic activities at high temperature and produces extracellular thermostable enzymes.
Based on whole-genome sequencing, we demonstrated that the B. aerius strain CCMM B940 could be reclassified as B. paralicheniformis. Genomic analysis confirmed the fibrolytic potential of the strain CCMM B940 observed in vitro using purified and natural substrates and provided information on its ability to degrade pectins. This statement reinforces the strain's interest especially in agro-waste bioconversion. To our knowledge, this is the first report of lichenase and xylanase activities by a thermophilic strain of B. paralicheniformis isolated from a hot spring. The CCMM B940 strain is part of the large Coordinated Moroccan Collections of Microorganisms (CCMM), which comprise more than 1800 isolates of bacteria, yeasts and fungi. Many of these microorganisms have also been isolated from extremophile environments (Berrada et al. 2012;Aanniz et al. 2015) and they may harbor beneficial attributes such as production of heat-resistance enzymes as already observed in B. amyloliquefaciens (Beladjal et al. 2018) and B. subtilis (den Besten et al. 2017) strains. For this purpose, complete genome sequencing not only enables accurate identification and reclassification of CCMM isolates, i.e. the thermophilic strains of the B. aerius group, but also opens the door to the field of biotechnological applications such as production of second-and third-generation biofuels from raw or pretreated biomass, medical and nutraceutical fields and food processing (Benedetti et al. 2019), in the favor of the fibrolytic potential (CAZymes) of these strains. Enzymatic activities of these strains should be further studied to determine the most suitable markets in biotechnology (agro-waste bioconversion, depollution, health, pharmaceuticals, etc.…).

Accession numbers
The assembled genome and all relevant sequences were deposited in NCBI's GenBank on November 24, 2020 under the following accession numbers, BioProject PRJNA680612; BioSample SAMN16895636; Genome JADRJL000000000.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.