General genomic features of Serratia AORB19

The final Serratia AORB19 genome assembly contained 42 contigs sequences longer than 500 bp with a total consensus genome size of 5587018 bp (∼5.5 Mb) and GC content of 54.99%, which was comparable to the genome sizes of previously sequenced Serratia genomes. The largest contig size was 1236132 bp, and the N50 size was 370852 bp. The BUSCO evaluation showed 99.7% completeness of this genome, indicating a high-quality genome assembly. All annotation statistics are listed in Table 1.

Table 1 General features of strain AORB19 genome

The genome had 5655 protein-coding genes, of which 4002 genes were functionally assigned, while the remaining genes were annotated as hypothetical proteins (29.23%). In addition, 108 non-coding RNAs, including tRNAs, rRNAs, and snRNA were identified in the Serratia genome. The circular plot of the genome is represented in Fig. 1.

Fig. 1

Overview of S. quinivorans AORB19 genome. A: Circular plot generated by DNA plotter. Circles indicate, from inside outwards: GC skew (yellow/green); GC content (moss green/purple); putative lignocelluloses enzymes (cyan); putative laccase –like enzyme genes (purple); tRNA coding genes (green) and rRNA (red); protein-coding genes on reverse strand (blue); and protein-coding genes on forward strand (red)

The functional annotation reveals that out of the total protein-coding genes, 4955 genes that have been assigned were annotated with Clusters of Orthologous Genes (COG) terms and grouped into 21 classes, generally covering the essential functions of cellular processes, metabolism, information storage and general function. The top five functional COG terms were transcription, K (10%), amino acid transport and metabolism, E (10%), carbohydrate transport and metabolism, G (8%), inorganic ion transport and metabolism, P (6%), and cell wall/membrane/envelope biogenesis, M (6%) (Fig. 2). In the Kyoto Encyclopedia of Genes and Genomes (KEGG) annotation of the genome, 3320 genes were successfully annotated with 2333 KEGG Orthologous (KO) terms. Detailed analysis of the KEGG pathways (Fig. 3) revealed that the genes were enriched in 23 pathways and four major functions of metabolism, including genetic information processing, environmental information processing and cell process. In the category of metabolic processes, 11 pathways were annotated among which 400 and 303 genes were enriched in the carbohydrate metabolism and amino acid metabolism pathways, respectively. Five pathways in the genetic information processing category were annotated, with the translation pathway having the most annotations of 82 genes. Under the environmental information processing category, 256 and 159 genes were enriched in the membrane transport and signal transduction pathways, respectively. A total of 2366 genes from AORB19 were annotated in the Gene ontology (GO) database under the three broad categories of biological process (BP), cellular component (CC), and molecular function (MF). As shown in Fig. 4, the top three annotated BPs were the metabolic processes, cellular metabolic process and organic substance metabolic process, in which 1370, 1216, and 1202 genes were enriched, respectively. There were 1030, 988, and 824 genes enriched in the top three CCs, including intracellular anatomical structure, cytoplasm and cytosol respectively. The top annotated MFs include catalytic activity, binding and ion-binding with 1244, 963 and 496 genes.

Fig. 2

Distribution of Clusters of Orthologous Genes (COG) functional categories in the complete genome sequence of S quinivorans AORB19 genome

Fig. 3

Kyoto Encyclopedia of Genes and Genome (KEGG) pathway annotation of the assembled genome of strain AORB19. Percentage of gene sequences assigned to each sub-category of the four top KO categories, namely metabolism (red), genetic information processing (green), environmental information processing (blue), cellular processes (yellow) were calculated and displayed

Fig. 4

GO classification of bacterial gene function annotation. MF stands for molecular function; BP for biological process and CC for cellular components

Whole genome comparison

The GenBank now contains numerous whole-genome sequences due to advancements in next-generation sequencing, enabling the utilization of whole-genome sequencing as a novel approach for species differentiation. Average Nucleotide Identity (ANI) and in-silico DNA-DNA hybridization (isDDH) analysis are the most commonly used parameters for taxonomic assignments. In the context of genome-based species delineation, in-silico DDH is widely regarded as a reliable alternative to traditional DDH methods.

ANI is a measure of nucleotide-level genomic similarity between the coding regions of two genomes. The ANI matrix (Fig. 5) reveals that the genome of AORB19 showed an ANI < 95% with most of the other type strains in the genus Serratia including the S. proteamaculans CCUG 14510. A 95% ANI cutoff is the most frequently used standard for species demarcation [28]. However, AORB19 shows a high ANI value (> 95%) with S. quinivorans NCTC11544 and therefore this genome is very close to the S. quinivorans genome. Genome-to-Genome Distance Calculator (GGDC) analyses among the 15 Serratia strains (Fig. 5) yielded in-silico DDH estimates greater than 81% between strain AORB19 and S. quinivorans NCTC11544 indicating that they potentially belong to the same species. All other 13 pairwise comparisons with AORB19 indicated that they were different species based on 70% species delimitation threshold [29].

Fig. 5

Matrix showing the values of ANI and in-silico DDH comparison of S. quinivorans AORB19 with 14 other type strains of Serratia. The upper half of the matrix (highlighted in yellow) shows the results of ANI analysis whereas the lower half of the matrix (highlighted in blue) shows the results of in-silico DDH analysis

To further infer the phylogenomic relationship of this genome, 16S phylogenetic neighbour joining type tree was constructed. The 16S rRNA tree (Supplementary file 1: Fig. S1) reveals that S. quinivora AORB19 is placed in a clade along with S. proteamaculans, S. liquefaciens, S. grimesii and S. quinivorans CP6a but quite distinctly placed from S. quinivorans NCTC 11544 strain. Classical 16S rRNA gene-based trees might not possess sufficient resolution to differentiate between closely related species like S. liquefaciens, S. grimesii, S. proteamaculans and S. quinivorans [30]. Core genome-based phylogeny provides a better resolution and is more robust than conventional marker gene-based phylogeny. Thus, we investigated the genetic diversity within the Serratia genus by inferring the phylogenomic relationship based on the core genome. For this purpose, we used the up-to-date bacterial core gene set (UBCG2) consisting of 81 core genes from 3,508 bacterial species covering 43 phyla [31]. Phylogenetic tree analysis inferred by the maximum-likelihood method based on core genes defined by UBCG2 (Fig. 6) clearly indicated that AORB19 belongs to the same clade as S. proteamaculans, S. quinivorans, S. liquefaciens and S. grimesii strains. But AORB19 is closer to S. quinivorans strains than to S. proteamaculans strains in this analysis. The maximum likelihood-based whole genome-based phylogenetic tree (Fig. 7) constructed on 74 publicly available Serratia genomes was consistent with the ANI and in-silico DDH results. It also corroborated the fact that this genome is closest to the Serratia quinivorans species although it had close relationships with S. proteamaculans and S. liquefaciens strains. Despite this strain having been previously reported as S. proteamaculans AORB19 [16] based on its 16S rRNA sequence, comparative genomic analyses of the entire genome suggest that the strain should be designated as Serratia quinivorans AORB19.

Fig. 6

Core genome-based phylogenetic tree constructed by UBCG pipeline

Fig. 7

Whole genome phylogeny constructed for S. quinivorans AORB19 and 74 publicly available Serratia genomes. Tree was constructed using the BV-BRC system (https://www.bv-brc.org)

Lignocellulose degrading enzymes of strain AORB19

CAZyme (carbohydrate-active enzyme) annotation for the AORB19 genome was completed utilizing the dbCAN 3 meta server (https://bcb.unl.edu/dbCAN2/). The different classes of CAZymes, including glycosyltransferase (GT), glycoside hydrolase (GH), polysaccharide lyase (PL), carbohydrate esterase (CE), auxiliary activity (AA), and carbohydrate-binding modules (CBMs), play crucial roles in sugar metabolism, specifically in the synthesis, binding, and breakdown of carbohydrates. The total annotated CAZymes gene number in AORB19 genome was 123 including 47 GTs, 58 GHs, 5 CEs, 8 AAs and 13 CBMs (Supplementary file 2: Fig. S2). No gene was assigned to PLs. A few genes were assigned to more than one class of CAZymes. For example, SQAORB19_5091 was assigned to both CBM32 and GH144 classes.

The glycoside hydrolases in the GH23 family were widely represented in this genome, including 8 predicted encoding genes involved in the deconstruction of peptidoglycan along with chitinase activity. The chitin degrading ability of Serratia marcescens has been previously described [32]. The genome of strain AORB19 contains 5 genes encoding GH1, and 5 for GH4s. GH1 members are a widespread group of enzymes that hydrolyze the glycosidic bond between two or more carbohydrate units. The GH1 family comprises the majority of bacterial β-glucosidases used in cellulose hydrolysis [33]. Members of GH4 are also involved in cleavage of glycosidic bonds, but exhibit unusual cofactor requirements for activity involving NAD + [34]. A total of 26 encoded proteins in the AORB19 genome were found to be involved in lignocellulose degradation including chitinase activity (Table 2). These included 10 cellulases, which are key enzymes involved in the decomposition of cellulose into glucose; they are members of the GH1, GH3, GH8, GH43, GH45, GH144 families. A variety of annotated GHs and CEs in AROB19 are predicted to participate in hemicellulose degradation. A total of 4 hemicellulases, including xylanases and mannanases were found in the genome belonging to GH43, CE4 and GH2 families.

Table 2 List of potential lignocellulose degrading enzymes found in the genome of strain AORB19

The second most frequent enzyme family contained in this genome was the glycosyltransferases GT family (47 encoding genes). The transfer of sugar residues from activated donor molecules to saccharide or non-saccharide acceptor molecules to form glycosidic linkages is facilitated by GTs. The most abundant GTs belong to the GT2 family including cellulose synthase, chitin synthase, mannosyltransferase, glucosyltransferase, galactosyltransferase. The output from dbCAN2 also included multiple hits corresponding to carbohydrate esterases (CEs) represented with 5 predicted genes attributed to CE11, CE14, CE4 and CE9 families. CEs accelerate the degradation of polysaccharides by acting on ester bonds in carbohydrates, thereby facilitating the access of glycoside hydrolases. The most abundant CEs in the genome belonged to the CE4 family acting on acetylated xylan and chitin. The genome also had 13 genes predicted to encode carbohydrate-binding modules (CBMs). The majority of predicted CBMs belonged to CBM50 and CBM48. CBM50 proteins play a crucial role in the binding of enzymes with cleavage activity of chitin or peptidoglycan, whereas CBM48 encodes specific modules with glycogen-binding function and is appended to GH13 modules.

The bacterial strain AORB19 also possessed an array of CAZymes and lignin-degrading enzymes (including aromatic compound-degrading and detoxifying enzymes) for the degradation of lignocellulose. CAZymes database (www.cazy.org) includes a class AA (Auxiliary Activities) which hosts a wide range of catalytic modules related to lignocellulose conversion and involved in plant cell wall degradation. The AA class works together with PL, CE and GH enzymes to get access to the carbohydrates within the cell wall of plants and facilitate lignin degradation. Within the AA classes, lignin oxidizing enzymes (LO) are classified into three subclasses including AA1, AA2, and AA3; and lignin degrading enzymes are classified into four: AA4, AA5, AA6, and AA8 [35]. Currently, the AA class encompasses 9 families of ligninolytic enzymes and 7 families of lytic polysaccharide mono-oxygenases. Our study revealed the presence of eight genes encoding ligninolytic enzymes under the AA class in the AORB19 genome. The genes among the AA class were: one in AA1 class which encodes the laccase enzymes, one from AA2 class, two from AA3_2 class, which comprise enzymes like aryl alcohol oxidase and glucose-1-oxidase, and one gene in the AA6 class and three in the AA10 class, known as copper-dependent lytic polysaccharide monooxygenases (LPMOs, Table 3). LPMOs are copper-dependent enzymes with multiple functionalities in plant biomass degradation and play a crucial role in lignin breakdown [36]. The presence of specific genes for different classes of lignin degrading enzymes indicates the strain’s ability for efficient break down of lignin and provides the evidence of the potential of AORB19 to be further genetically modified on specific metabolic pathways and regulatory mechanisms based on practical applications.

Table 3 Category and number of annotated open reading frames (ORFs) of auxiliary activities (AA) families in AORB19Analyses of lignin degradation pathways of Serratia quinivorans AORB19

Microbial lignin degradation is a complex process involving multiple enzymes and intermediate products. The specific pathway and products produced may vary depending on the specific microorganisms, conditions involved in the degradation process, as well as source or type of lignin. Degradation process can be broadly categorised into two stages based on conditions and nature of compounds: i) the peripheral, or upper pathway that includes lignin depolymerization to central intermediate products such as catechol and protocatechuate and ii) the central, or lower pathway that includes aromatic ring opening to precursor molecules and compounds such as muconic acid derivatives, acetyl-CoA, succinyl-CoA and pyruvate [35, 37,38,39].

Liquid chromatography with UV detection (LC-UV) was employed to identify (using UV spectra) and monitor potential intermediates produced during the degradation of the alkali lignin by Serratia quinivorans AORB19 over an 8-d period (Fig. 8; Supplementary file 3: Table S1). Alkali lignin medium without bacterial inoculation was used as a negative control. In the negative control, a number of lignin-derived phenolic compounds were detected including p-hydroxybenzaldehyde, vanillic acid, and vanillin. The detection of compounds in the negative control can be attributed to the residual by-products from the alkali treatment during the manufacturing process [40].

Fig. 8

Zoomed LC-UV (λ = 280 nm) chromatograms of Alkali Lignin (Sigma Aldrich) incubated for 168 h without strain (control) or for increasing times (48–168 h) with Serratia quinivorans AORB19

In the treated samples, three phenolic compounds were observed with concentrations varying as follows: p-hydroxybenzaldehyde, a typical phenolic compound derived from the degradation of lignin, was found to gradually increase over time; vanillin, which experienced a significant decrease after 48 h compared to the control, then gradually rose over the remaining time period of the study; and vanillic acid, which was completely consumed after 48 h, remained undetectable until the end of the experiment (Fig. 8; Supplementary file 3: Table S1). The preponderance of vanillin and vanillic acid in the supernatants is in agreement with the origin of alkali lignin used in this study, i.e., softwood. The increasing concentration of p-hydroxybenzaldehyde and vanillin in the samples is indicative of lignin degradation by the Serratia quinivorans AORB19 bacterium. Both chemicals result from the cleaving oxidation of the benzyl carbon adjacent to the p-hydroxyphenyl (H) and guaiacyl (G) units, respectively [41,42,43]. The Serratia quinivorans AORB19 genome harbored several oxidative lignin degrading enzymes including a few potential laccase-like genes—multi multicopper oxidases (SQAORB19_2332; SQAORB19_1606; SQAORB19_2924), DyP-peroxidases (SQAORB19_3208; SQAORB19_3467; SQAORB19_3468) and dehydrogenases (SQAORB19_610; SQAORB19_635) contributing to the formation of the two difunctionalized aldehydes. In similar lines, a thermoalkaliphilic laccase from Caldalkalibacillus thermarum TA2.A1 has also been reported to depolymerize kraft lignin to p-hydroxybenzaldehyde as a metabolic intermediate [44]. Once formed, the aldehydes can be oxidized into p-hydroxybenzoic acid or vanillic acid by the action of dehydrogenase enzymes, as previously reported [45, 46]. In agreement, two aldehyde dehydrogenases genes (SQAORB19_5141; SQAORB19_610) were identified during the genome analyses of strain AORB19. Similarly, an aldehyde dehydrogenase gene responsible for the conversion of syringaldehyde to syringic acid was identified in a previous study [39].

The fact that vanillic acid, present in the controls appeared completely depleted in the presence of S. quinivorans AORB19 suggests that the latter has the capability to transform the acid. The genome of strain AORB19 possessed the candidate genes of vanillate O-demethylase oxygenase (SQAORB19_2890; SQAORB19_5632) that is known to convert vanillic acid into protocatechuic acid [47]. The o-catechol and protocatechuate obtained from the depolymerization of lignin monomers can then be converted to cis,cis-muconate and 3-carboxy-cis,cis-muconate, respectively, by the ring opening dioxygenases (SQAORB19_3419; SQAORB19_3420) identified in the genome of strain AORB19. The β-ketoadipate pathway plays a vital role in the degradation of aromatic compounds, including lignin, by enabling microorganisms to utilize carbon from complex polymers, and derives its name from the characteristic intermediate, β-ketoadipate, which serves as a branching point for further metabolism [48,49,50,51]. The genome of strain AORB19 also possessed numerous key genes for the β-ketoadipate pathway that could be involved in the degradation of aromatic compounds. SQAORB19_989 and SQAORB19_3417 encode CMD γ -carboxymuconolactone decarboxylase; SQAORB19_1365 encodes ELH, enol lactone hydrolase; SQAORB19_616, SQAORB19_1321, SQAORB19_1366 encode TH, β-ketoadipyl-CoA thiolase; SQAORB19_1367 (subunit B), SQAORB19_1368 (subunit A) encode TR, β-ketoadipate:succinyl-CoA transferase; SQAORB19_3416 encodes CMH, β-carboxymuconolactone hydrolase and CMLE, β-carboxy-cis, cis-muconate lactonizing enzyme; SQAORB19_3419 (alpha chain) and SQAORB19_3420 (beta chain) encode P3,4O, protocatechuate 3,4-dioxygenase.

Notably, genes for additional pathways linked with lignin degradation such as the gentisate, anthranilate, homogentisic and phenylacetate–CoA pathways were also detected in the genome analyses (see Supplementary file 4: Table S2; Supplementary file 5: Table S3). These degradation pathways allow prokaryotes to break down and utilize aromatic compounds as a carbon and energy sources. Taken together, the results of Serratia quinivorans AORB19 genome analysis and the formation of detectable traces of transient aromatic compounds in LC-UV analysis validated each other and confirmed the lignin-degradative traits of the strain Serratia quinivorans AORB19. More so, with the genetic pathways characterized, the strain Serratia quinivorans AORB19 has the potential to undergo genetic modification and optimization in order to enhance the cost-effectiveness and sustainability of lignocellulosic biorefineries, addressing the urgent need for efficient techniques to convert lignin into bioproducts.

Determination of laccase production using agro-industrial biomasses

Considering the strain's inherent capacity for natural extracellular laccase production and its lignin-degradative traits identified through genomic analysis, the strain was tested for its ability to enhance laccase production in various agro-industrial biomasses sourced from Canadian industries. Various industrial waste biomasses including pea hull, flax seed meal, canola meal, okara and barley malt sprouts were used separately as a carbon source to assess their impacts on the strain’s laccase production. It was observed that for all substrates, the enzyme activity maximized at 48 h. At this time, flax seed meal exhibited maximum laccase activity of 257.71 U/L, which was three-fold higher than the laccase activity observed when alkali lignin (Sigma Aldrich) was utilized as the carbon source (83.65 U/L). It was followed by pea hull activity of 230.11 U/L, canola meal activity of 209.56 U/L, okara activity of 187.67 U/L and barley malt sprouts activity of 169.27 U/L (Fig. 9). These results indicated the excellent adaptability of the strain to grow well on a broad range of substrates, particularly on flax seed meal leading to enhanced laccase production. Of note, the strain Serratia quinivorans AORB19 efficiently produced laccase enzymes using these heterogeneous biomasses without requiring further substrate optimization.

Fig. 9

Laccase production by Serratia quinivorans AORB19 using different agro-industrial biomasses as carbon source

Microorganisms that encompass broad substrate utilization including lignocellulose-rich plant or agricultural residues are deemed essential for robust enzymatic degradation of lignocellulose [52, 53]. Agro-industrial residues such as pea hull, flax seed meal, canola meal, okara, and barley malt sprouts are the most common feed ingredients for poultry and livestock in Canada [54]. Multiple cohesive enzymes, including laccases are required to facilitate the degradation of intact cell walls of these agro-industrial residues. Apart from the impressive growth characteristics and laccase production observed in this study using flax seed meal, other research has also reported the induction of laccase production in different organisms using various substrates. For instance, pea peels induced laccase production (0.85 U/mL) in Bacillus aquimaris AKRC02 through submerged fermentation [55]. Similarly, the fungus Pleurotus ostreatus DAOM 197961 was able to produce laccase when grown in canola meal and was found to decrease the meal’s phenolic content upon solid-state fermentation [56]. Moreover, a 2.11-fold increase in laccase production was observed when 0.1% okara was used in submerged fermentation with gram-negative Rheinheimera sp. [57].

Bacterial laccases are exceptional environmentally friendly catalysts known for their wide substrate specificity, offering numerous potential applications in areas such as bioremediation, lignocellulose processing, waste valorization and beyond. In the study of bioprocessing using mixed cellulosic feedstocks for ethanol production [58], the increased release of reducing sugars was attributed to the synergistic action of cellulases and ligninolytic laccase, which potentially enhances the accessibility of holocellulolytic enzymes to holocelluloses, consequently leading to improved production of fermentable sugars. Similarly, strain AORB19’s ability to thrive and produce laccase using low-cost raw substrates as carbon sources suggested it may, at the same time, produce holocellulolytic enzymes that synergistically work together with laccase to efficiently hydrolyze lignocellulose. The existence of relevant gene encoding sequences in the genome sequence supports such a hypothesis that can be confirmed by enzymatic activity analyses in future research. Our results suggested the great potential for strain AORB19’s applications in low-cost enzyme production, biomass pretreatment and valorization.