Genome sequence of the H₂-producing Clostridium beijerinckii strain Br21 isolated from a sugarcane vinasse treatment plant

Abstract

We report on the nearly complete genome sequence of Clostridium beijerinckii strain Br21, formerly isolated from a sugarcarne vinasse wastewater treatment plant. The resulting genome is ca. 5.9 Mbp in length and resembles the size of previously published C. beijerinckii genomes. We annotated the genome sequence and predicted a total of 5323 genes. Strain Br21 has a genetic toolkit that allows it to exploit diverse sugars that are often found after lignocellulosic biomass pretreatment to yield products of commercial interest. Besides the whole set of genes encoding for enzymes underlying hydrogen production, the genome of the new strain includes genes that enable carbon sources conversion into butanol, ethanol, acetic acid, butyric acid, and the chemical block 1,3-propanediol, which is used to obtain polymers. Moreover, the genome of strain Br21 has a higher number of ORFs with predicted beta-glucosidase activity as compared to other C. beijerinckii strains described in the KEGG database. These characteristics make C. beijerinckii strain Br21 a remarkable candidate for direct use in biotechnological processes and attest that it is a potential biocatalyst supplier.

Keywords:
Clostridium; biofuels; biohydrogen; beta-glucosidase

Hydrogen (H₂) has attracted attention because it is an energy carrier with higher energy per unit of weight (120 kJ/g) than fossil fuels, such as petroleum (42 kJ/g) and coal (24 kJ/g). In addition, H₂ combustion does not emit CO₂. Besides physical chemical approaches, H₂ can be obtained by fermentation of renewable materials, such as pure carbohydrates or carbohydrate-rich wastes and wastewater, at low pressure and temperature (Elsharnouby et al., 2013).

Bacteria in the genus Clostridium, mainly C. acetobutylicum and C. beijerinckii, can generate various products of industrial interest, including H₂ (Elsharnouby et al., 2013). During their exponential growth phase, these bacteria excrete acetate, butyrate, H₂, and CO₂ (Schiel-Bengelsdorf et al., 2013). At the end of the exponential growth phase, these bacteria take up acetate and butyrate and convert them into acetone, butanol, and ethanol in the so-called ABE fermentation (solventogenesis), and start endospore synthesis (Jones and Woods, 1986; Schiel-Bengelsdorf et al., 2013). Elucidating the acidogenesis, solventogenesis, and sporulation metabolic networks is crucial if we are to take advantage of this metabolism to obtain desired industrial products.

We present the genome sequence of a new isolate within the phylum Firmicutes, namely the bacterial strain Br21 belonging to the family Clostridiaceae. This bacteria was previously isolated from a sludge collected from an Upflow Anaerobic Sludge Blanked (UASB) bioreactor employed to treat wastewater from a sugar and ethanol production plant. To ensure the emergence of spore-forming bacteria, we acidified the sludge at pH 3.0 for 12 h before isolating the new Clostridium strain, as described previously (Fonseca et al., 2016).

The ability of the new isolate to produce H₂ from different monosaccharides was assayed in a preceding work. Strain Br21 affords the highest H₂ yield using glucose, galactose, mannose, and xylose, that are the main biomass substrates (Fonseca et al., 2016). In the same formerly work, Strain Br21 16S rRNA gene was sequenced (GenBank accession no. KT626859) revealing that this strain is affiliated to the family Clostridiaceae (order Clostridiales) and has 99.78% 16S rRNA gene sequence identity with C. beijerinckii NCIMB 8052 and C. diolis DSM 5431 as the two most closely related, validly described species (Fonseca et al., 2016). However, to confirm the new isolate identity, as well as to get deeper insight about its biotechnological potential its whole genome was sequenced as described below.

Bacterial cells were imaged by high-resolution scanning electron microscopy (SEM) (JEOL, Ltd.; Tokyo, Japan) (Figure S1). After 24 h, the cells consisted of elongated and round straight bacterial stems measuring ca. 3-8 μm × 0.7 μm (Figure S1 A,B). At the end of the exponential growth phase (at 60 h), stem-shaped cells began to form endospores (Figure S1 C). All the morphological characteristics described above agree with literature data for C. beijerinckii (Jones and Woods, 1986).

For genome sequencing, we obtained strain Br21 DNA from a cell pellet after cultivating the bacterium in liquid CH medium for 48 h, as described in Fonseca et al. (2016). We generated one short insert size paired-end library by using the Nextera DNA preparation kit and an additional long-insert library, 5-7 Kbp, with the Nextera Mate Pair Library preparation kit. We sequenced both libraries on HiSeq2500, which produced a total of ~1.4 x 10⁷ reads (2 x 100 bp). We preprocessed mate pair and paired-end reads with FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and Trimmomatic (Bolger et al., 2014). Mate-pair reads were further processed with NExtClip and types A, B, and C reads were kept for de novo assembly and scaffolding (Leggett et al., 2014). We estimated genome size by kmer statistics with Kmergenie (Chikhi and Medvedev, 2014). The high-quality reads were assembled in SPAdes v3.9.0 (Bankevich et al., 2012). The resulting genome assembly is ~5.9 Mbp in length (99.8% of the predicted genome size), similar in size to previously published C. beijerinckii genomes (https://goo.gl/3SgkeS), with a final coverage of 230x.

This Whole Genome Shotgun project was deposited at DDBJ/ENA/GenBank under the accession number MWMH00000000. The version described in this paper is version MWMH01000000.

The assembly has 28 scaffolds; the longest is 1.19 Mbp with mean, median, and N50 lengths of 214,033.75 bp, 81,771 bp, and 604,572 bp, respectively. The genome assembly was annotated with the NCBI Prokaryotic Genome Annotation Pipeline (Tatusova et al., 2016), which predicted a total of 5323 genes, of which 5099, 16, 54, 7, 147, and 1 encode for proteins, rRNA genes, tRNAs, ncRNAs, putative pseudogenes, and CRISPR array, respectively. The 16S rDNA phylogenetic tree shown in Figure 1 (Table S1) places strain Br21 in a clade with C. beijerincki and C. diolis, although with low bootstrap support, which prompted us to carry out genome-wide analyses to confirm strain assignment to the species level. The multilocus phylogenetic tree shown in Figure S2 (Table S2) was inferred based on a set o 168 single-copy gene markers, proposed to resolve phylogenetic relationships among Firmicutes (Wang and Wu, 2013). The multilocus tree clearly shows all C. beijerincki strains, as well as the single C. diolis strain and strain Br21 forming a clade with 100% bootstrap support. Particularly, Br21 forms a subclade (with 100% bootstrap support) with the C. beijerincki strains DSM 53, NRRL B-593 and NRRL B-528 (individual gene alignments and phylogenetic trees are available under DOI:10.6084/m9.figshare.5993164). Further genome-wide comparisons conducted with the Genome-to-Genome distance calculator (Meier-Kolthoff et al., 2013) revealed that strain Br21 has a predicted DNA-DNA hybridization value (DDH) of 66.3%, 65.3%, and 76.5% against C. diolis DSM 15410, C. beijerinckii NCIMB 8052, and C. beijerinckii NRRL B-528, respectively (Table S3). DDH values above 70% are required for species-level assignment. Computation of genome wide ANI values and alignment fraction are being increasingly used in bacterial taxonomy and have been posited as objective and precise criteria for bacterial species delimitation (Varghese et al., 2015). The OrthoANIu (Lee et al., 2016) values are 96.18%, 96.12%, and 97.61% respectively, with 95%-96% being usually used as cut-off for species demarcation (Figure 2, Table S3). The DDH of C. diolis DSM 15410 vs C. beijerinckii NCIMB 8052 is 79.2%, their OrthoANIu is 97.74%, suggesting that they are members of the same species, in agreement with recent findings (Figure 1 and Figure S1; Poehlein et al., 2017). An extended description of the procedures followed to assign strain Br21 to the species level is available in Supplementary Material Text S1.

Figure 1
Maximum likelihood phylogenetic tree based on the 16S rDNA sequences, representing the evolutionary relationships between strain Br21 (in bold) and closely related strains in the genus Clostridium. The scale shows 0.02 nucleotide changes per nucleotide position. Sequences with at least 94% identity to the 16S rDNA sequence extracted from the strain Br21 genome sequence were identified with the EzBiocloud identification tool and kept for further analysis. Sequences were aligned with MAFFT’s Q-INS-i option. Phylogeny was inferred with RAxML under the GTR+Γ+I evolutionary model, with automatic bootstrapping. (T) Type strain. (*) At least one strain in the species has had its genome sequenced. (**) The 16S rDNA sequence was extracted from the genome sequence. See Text S1 for further details and Table S1 for details of the sequences included.

Figure 2
Heatmap of OrthoANIu results among members of C. beijerincki, C. diolis and Br21. Values above 95%, usually considered as species boundary, appear in shades of red. Strain HUN142 is very distinct from any other C. beijerincki strain. Strain DSM 15419 formally assigned to C. diolis, cannot be distinguished from the main group of C. beijerincki strains. Strain Br21 forms a subcluster (lower left corner) together with C. beijerincki strains DSM 53, NRRL B-593 and NRRL B-528, a group that is also supported by the multilocus phylogenetic analysis (Figure S2). See Text S1 for further details and Table S1 for details of the sequences included.

According to Biebl and Spröer (2002), some C. diolis strains are very close to C. beijerinckii (as judged from molecular analyses) and not very distant in terms of the DNA-DNA hybridization data, but growth and nutrition differences suggested their classification as a separate species. For example, unlike C. beijerinckii, C. diolis does not ferment starch, raffinose, or inositol (Biebl and Spröer, 2002). Experiments conducted in our laboratory showed that strain Br21 can grow by consuming starch, raffinose, or inositol as the only carbon source, giving H₂ as product (data not shown). Thus, we named strain Br21 as C. beijerinckii. The phylogenetic analyses, the predicted DDH values and the OrthoANIu results support strain Br21 classification as C. beijerinckii, being strains NRRL B-593, NRRL B-528, and DSM 53 its closest relatives, within a cohesive and distinct clade.

We divided the enzymes identified in the C. beijerinckii Br21 genome into functional classes according to the EC nomenclature so that we could focus on the substrate specificity range of the enzyme catalogue, with especial emphasis on the glycosyl hydrolase group (EC 3.2.1.-). We detected 49 genes encoding for enzymes that hydrolyze or modify sugars, including α-galactose, cellobiose, starch, glycogen, maltose, chitin, pullulan, 6-phospho-D-glucosides (including 6-phospho-beta-D-glucosyl-(1,4)-D-glucose, trehalose-6-phosphate, and sucrose 6-phosphate), xylose, alpha-D-mannose, and α-L-arabinosides (Figure 3, Table S4). Accordingly, former experimental data showed strain Br21 can grow and produce H₂ from glucose, sucrose, xylose, cellobiose, and starch (Fonseca et al., 2016). Closely related bacteria like C. beijerinckii NCIMB 8052 and ATCC 35702 present a slightly higher number of total genes encoding for glycosyl hydrolases (58 and 61, respectively) as compared to the 49 genes identified in strain Br21. Unexpectedly, strain NCIMB 8052 does not contain genes encoding for alpha-mannosidases (EC 3.2.1.24), alpha-xylosidases (EC 3.2.1.-), or neopullulanases (EC 3.2.1.135), and strain ATCC 35702 does not have genes for alpha-mannosidases (EC 3.2.1.24) and neopullulanases (EC 3.2.1.135) in their genomes, as revealed by searches in the KEGG database (Kanehisa and Goto, 2000).

Figure 3
Distribution of glycosyl hydrolases identified in the genome of C. beijerinckii Br21 and related strains (NCIMB 8052, ATCC 35702 and NCIMB 14988) based on their function as defined by the fourth level of EC nomenclature. Only enzymes for which the relative percentage is higher than 5% relative to the total are specifically shown.

Interestingly, strain Br21 has a larger number of ORFs with predicted beta-glucosidase activity (EC 3.2.1.21; 9 genes) as compared to C. beijerinckii NCIMB 8052 and ATCC 35702 (5 and 6 ORFs, respectively) (Figure 3, Table S4). Beta-glucosidases (beta-D-glucoside glucohydrolases, EC 3.2.1.21) have a key role in cellulose hydrolysis, as they complete the final degradation step (Singhania et al., 2013). These enzymes have recently attracted attention due to their functions in the production of bioethanol and other biofuels from agricultural residues (Singhania et al., 2013).

As mentioned above, the C. beijerinckii Br21 genome comprises 49 genes encoding glycosyl hydrolases; some of these genes are absent in the genomes of closely related strains (Figure 3, Table S4). This corresponds to the hydrolysis of (1,6)-α-, (1,2)-α-, (1,4)-β-, (1,4)-α-branch linkages, which is known to play essential roles in the sugar industry, pulp and paper industry, as well as in medicine (Ferrer et al., 2016).

Concerning C. beijerinckii Br21 application in biofuel production, the genome analysis showed the presence of genes encoding for the electron carriers and enzymes involved in H₂ evolution and butanol fermentation. C. beijerinckii Br21 has 19, 3, and 5 genes, respectively, encoding for the electron carrier ferredoxin, pyruvate-flavodoxin oxidoreductase-PFOR (EC 1.2.7.-), and Fe-Fe hydrogenase (EC 1.12.7.2), which directly account for H₂ generation (Table S5). Moreover, the Br21 strain genome presents the most important genes encoding for the enzymes underlying butanol production, including genes related to acetyl-CoA acetyltransferase or -thiolase (EC 2.3.1.9), butyrate-acetoacetate CoA-transferase and acetyl-CoA acetyltransferase (CoA transferase, EC: 2.8.3.9), 3-hydroxybutyryl-CoA dehydrogenase (EC 1.1.1.157), butyryl-CoA dehydrogenase (EC: 1.3.8.1), NADH-dependent butanol dehydrogenase A (EC: 1.1.1.-), and an alcohol dehydrogenase that can transform butyraldehyde into butanol (see Table S5). However, strain Br21 does not present the gene adc encoding for acetoacetate decarboxylase, which catalyses the conversion of acetoacetate to acetone and carbon dioxide (Table S5).

Remarkably, C. beijerinckii Br21 presents genes encoding for enzymes that convert glycerol into 1,3-propanediol, a high-value chemical block used to produce a thermoplastic for the textile and automobile industries (Papanikolaou et al., 2000). The gene encoding for enzyme 1,3-propanediol dehydrogenase (EC 1.1.1.202) appears in its genome (Table S5), but not in the C. beijerinckii NCIMB 8052 and ATCC 35702 genomes, as revealed by KEGG database analysis. 1,3-Propanediol generation by C. beijerinckii DSM 791 has been described only recently (Wischral et al., 2016).

Strain Br21 genome analysis highlights its biotechnological potential, particularly regarding its use in the production of biofuels and chemicals from a wide spectrum of substrates.

Acknowledgments

We gratefully acknowledge the time allotted to us by the NGS facility at Laboratório Nacional de Ciência e Tecnologia do Bioetanol (CTBE), which integrates the Centro Nacional de Pesquisa em Energia e Materiais (CNPEM). We thank Mr. Douglas Antonio Alvaredo Paixão for his help during the construction of the DNA libraries. We greatly acknowledge financial support from the Sao Paulo State Foundation (FAPESP, grant numbers 2015/04309-1 and 2015/06074-1).This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.

Conflict of Interest

The authors declare that there is no conflict of interest that could be perceived as prejudicial to the impartiality of the reported research.

Author contributions

BCF isolated the microorganism and prepared the material for sequencing. DMRP carried out genome assembly, genome annotation, phylogenetic analyses and final taxonomic assignment. MEG and VR planned and supervised the study. All authors wrote the manuscript, read, and approved the final version.

References

• Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, Lesin VM, Nikolenko SI, Pham S, Prjibelski AD et al. (2012) SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol 19:455-/457.
• Biebl H and Spröer C (2002) Taxonomy of the glycerol fermenting Clostridia and description of Clostridium diolis sp nov. Syst Appl Microbiol 25:491–/497.
• Bolger AM, Lohse M and Usadel B (2014) Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30:2114-/2120.
• Chikhi R and Medvedev P (2014) Informed and automated k-mer size selection for genome assembly. Bioinformatics 30:31-/37.
• Elsharnouby Y, Hafez G, Hesahm NM and Naggar E (2013) A critical literature review on biohydrogen production by pure cultures. Int. J. Hydrog. Energy 38:4945–/4966.
• Ferrer M, Martinez-Martinez M, Bargiela R, Streit WR, Golyshina OV and Golyshin PN (2016) Estimating the success of enzyme bioprospecting through metagenomics: current status and future trends. Microb Biotechnol 9:22-/34.
• Fonseca BC, Guazzaroni ME and Reginatto V (2016) Fermentative production of H₂ from different concentrations of galactose by the new isolate Clostridium beijerinckii Br21. Int J Hydrogen Energ 41:21109-/21120.
• Kanehisa M and Goto S (2000) KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res 28:27-/30.
• Jones DT and Woods DR (1986) Acetone-butanol fermentation revisited. Microbiol Rev 50:486–/524.
• Lee I, Kim YO, Park SC and Chun J (2016) OrthoANI: An improved algorithm and software for calculating average nucleotide identity. Int J Syst Evol Microbiol 66:1100-/1103.
• Leggett RM, Clavijo BJ, Clissold L, Clark MD and Caccamo M (2014) NextClip: An analysis and read preparation tool for Nextera Long Mate Pair libraries. Bioinformatics 30:566-/568.
• Meier-Kolthoff JP, Auch AF, Klenk HP and Goker M (2013) Genome sequence-based species delimitation with confidence intervals and improved distance functions. BMC Bioinformatics 14:/60.
• Papanikolaou S, Ruiz-Sanchez P, Pariset B, Blanchard F and Fick M (2000) High production of 1,3-propanediol from industrial glycerol by a newly isolated Clostridium butyricum strain. J Biotechnol 77:191–/208.
• Poehlein A, Solano JDM, Flitsch SK, Krabben P, Winzer K, Reid SJ, Jones DT, Green E, Minton NP, Daniel R and Dürre P (2017) Microbial solvent formation revisited by comparative genome analysis. Biotechnol Biofuels 10:/58
• Schiel-Bengelsdorf B, Montoya J, Linder S and Durre P (2013) Butanol fermentation. Environ Technol 34:1691-/1710.
• Singhania RR, Patel AK, Sukumarana RK, Larroche C and Pandey A (2013) Role and significance of beta-glucosidases in the hydrolysis of cellulose for bioethanol. Biores Technol 127:500–/507.
• Tatusova T, DiCuccio M, Badretdin A, Chetvernin V, Nawrocki EP, Zaslavsky L, Lomsadze A, Pruitt KD, Borodovsky M and Ostell J (2016) NCBI prokaryotic genome annotation pipeline. Nucleic Acids Res 44:6614-/6624.
• Varghese NJ, Mukherjee S, Ivanova N, Konstantinidis KT, Mavrommatis K, Kyrpides NC and Pati A (2015) Microbial species delineation using whole genome sequences. Nucleic Acids Res 43:6761-/6771.
• Wang Z and Wu M (2013) A phylum-level bacterial phylogenetic marker database. Mol Biol Evol 30:1258-/1262.
• Wischral D, Zhang J, Cheng C, Lin M, De Souza LMG, Pessoa FLP, Pereira Jr N and Yang S-T (2016) Production of 1,3-propanediol by Clostridium beijerinckii DSM 791 from crude glycerol and corn steep liquor: Process optimization and metabolic engineering. Bioresour Technol 212:100-/110.

Supplementary material

The following online material is available for this article:

Text S1 Extended description of the procedures followed for species level assignment of strain Br21.

Table S1 Details about the strains shown in Figure 1.

Table S2 Details about the genomes used for strain identification using genome-wide information.

Table S3 Complete results from the Genome-to-Genome Distance Calculator and OrthoANIu.

Table S4 Genes encoding for glycosyl hydrolases identified in the C. beijerinckii strain Br21 genome.

Table S5 Genes encoding for enzymes and electron carriers related to biofuel and chemical production identified in the C. beijerinckii strain Br21 genome.

Figure S1 Scanning electron micrograph of C. beijerinckii strain Br21 during the logarithmic growth phase.

Figure S2 Maximum-likelihood phylogeny with 168 markers for the genome sequences of all clostridia deposited in the NCBI RefSeq database

• Associate Editor: Ana Tereza R. Vasconcelos

Publication Dates

• Publication in this collection
  31 Jan 2019
• Date of issue
  Jan-Mar 2019

History

• Received
  31 Oct 2017
• Accepted
  12 Apr 2018