In silico genome analysis reveals the metabolic versatility and biotechnology potential of a halotorelant phthalic acid esters degrading Gordonia alkanivorans strain YC-RL2

Environmental pollution has become a global concern that requires concerted effort for monitoring, control and remediation plans across all sectors of society. Industrialization and urban development, advanced and commercial agriculture have all accelerated pollution. Recalcitrant organic xenobiotic compounds such as polychlorinated biphenyls (PCBs), plasticizers, synthetic polymers have attracted attention not only due to their persistence in the environment but all also their toxicity (El-Shahawi et al. 2010; Schäfer et al. 2011). Metal pollution is another concern for terrestrial and aquatic ecosystems (Fu and Wang 2011). Metals are naturally occurring in earth’s crust but anthropogenic activities have resulted in elevated levels that are detrimental to life. Metals at the right concentrations play crucial roles in biochemical and physiological activities but elevated levels can lead to an equilibrium shift leading to abnormalities and disease (Tchounwou et al. 2012; Wang and Shi 2001). Therefore, there is an urgent need for bioremediation and microbes have a rich potential for remediation of contaminated sites and synthesis of secondary metabolites that would be eco-friendly alternatives to their synthetic congeners (Das and Chandran 2011; Megharaj et al. 2011).

Members of phylum Actinobacteria possess a great potential for remediation of contaminated environments through degradation of xenobiotics, sequestration of heavy metals and production of novel secondary metabolites. In the phylum Actinobacteria, the genus of Gordonia has recently gained attention due to the biotechnological potential of its members (Drzyzga 2012). Members of genus Gordonia are high G+C gram positive bacteria of the order Actinomycetales (Arenskötter et al. 2004). Their ability includes but is not limited to degradation of various phthalic acid esters (PAEs), poly aromatic hydrocarbons (PAHs), alkylpyridines, 1,3,5-triazines and synthetic isoprene rubber (Drzyzga 2012). Production of extracellular polysaccharides (Ta-Chen et al. 2008), carotenoids (Arenskötter et al. 2004), steroids (Schneider et al. 2008), and surfactants (Drzyzga 2012) has been documented. With major technological advances in genome sequencing, prediction and elucidation of the biosynthetic pathways of these novel compounds through genome mining is now possible. Varied microbial natural products such as polyketide synthases (PKSs, NRPSs, and the ribosomally synthesized and post-translationally modified peptides (RiPPs) (Letzel et al. 2014; Liu et al. 2016) have recently gained attention in the pharmaceutical industry. Most Gordonia species contain genes for polyketide and carotenoid biosynthesis and alkane degradation (Sowani et al. 2018).

A halotorelant PAEs degrading Gordonia alkanivorans strain YC-RL2 was previously isolated from contaminated soil. It had an outstanding ability to utilize PAEs and other aromatic compounds as the sole carbon and energy source in varied environmental conditions (Nahurira et al. 2017). Strain YC-RL2 could transform dialkyl phthalates (DAPs) into phthalate (PA) via monoalky phthalates (MAPs) and then to benzoic acid (BA). Typically, the biodegradation process of PAEs is initiated by the hydrolysis of ester linkage forming monoester and subsequently PA and then protocatechuate. Though PA degrading genes are well characterized, not so much is known about genes and enzymes involved in the hydrolysis of PAEs to PA and its transformation to benzoate under aerobic conditions. Furthermore, not much is known about the genetic basis of YC-RL2 outstanding ability to utilize diversified substrates and adapt to a wide range of environmental conditions. This study focused on analyzing genome information to unveil the underlying genetic information of strain YC-RL2’s ability to degrade various aromatic compounds, PAEs, synthesize secondary metabolites and survive in contaminated environments.

From the results of 16S rRNA alignment and ANI calculation, strain YC-RL2 could be classified as G. alkanivorans. With a plethora of prokaryotic genomes in public databases, genomic classification was proposed by Chun et al. (2018) as the most reproducible and reliable method to infer phylogenetic relationships among prokaryotes (Chun et al. 2018).

The aerobic degradation pathway of PAEs in bacteria is initiated by the hydrolysis of ester bond to form corresponding MAPs. MAPs are further hydrolyzed to form PA (Liang et al. 2008) and esterases are the key enzymes (Luo et al. 2012). PA is then completely mineralized via protocatechuate (Benjamin et al. 2015) which is a central intermediate of aromatic ring degradation.

A previous study involving strain YC-RL2 showed that the strain could degrade di-2-ethylhexyl phthalate (DEHP) to PA and then BA instead of protocatechuate (Nahurira et al. 2017). More so, a phthalate degradation operon could not be identified in the genome of YC-RL2. We therefore hypothesized that PA could be decarboxylated to BA. Similar pathways were reported in Microbacterium sp. CQ0110Y (Chen et al. 2007), Mycobacterium sp. YC-RL4 (Ren et al. 2016). Thus the genome of strain YC-RL2 would have benzoate and catechol degradation clusters. We therefore analyzed the diversity of esterases, dioxygenases and decarboxylases. The putative esterases (Fig. 4) clustered majorly into two groups though the group containing CarEW from Bacillus sp K91was distinct. This is the only reported PAE esterase capable of hydrolyzing two bonds. This group contained WP_006358366.1 and WP_006358508.1 both annotated as carboxylesterases.

In typical phthalate degradation, further steps are accomplished by dioxygenases, dehydrogenases and later decarboxylases. A total of 22 dioxygenases were annotated in the genome and the phylogenetic tree between related dioxygenases shows that they are clustered into two major groups; Group I and II. The benzene ring is the most widely spread and thermodynamically stable chemical moiety which leads to its persistence in nature. Therefore, many aromatic compounds are major environmental pollutants (Díaz 2004). Ring cleavage is a crucial step in the aerobic degradation of aromatic pollutants and this is effected by oxygenases especially dioxygenases. Aromatic ring hydroxylating dioxygenases incorporate two atoms of molecular oxygen into their substrates to form cis-1, 2-dihydroxycyclohexadiene intermediates that are further acted upon by cis-diol dehydrogenases. This group of dioxygenases includes toluene, benzene, phthalate, naphthalene and biphenyl dioxygenases which are very important in xenobiotics degradation. Ring-cleaving dioxygenases on the other hand, catalyze fission of catecholic compounds. No dioxygenases showed similarity to reported phthalate dioxygenases and a complete phthalate operon could not be annotated in the genome. This shows that strain YC-RL2 possesses a unique pathway of phthalate degradation that needs to be explored further.

There were complete catechol and benzoate gene clusters annotated in the genome. The gene organization of YC-RL2 benABCD operon differs from that of most reported benABCD operons (Ezezika et al. 2006; Kitagawa et al. 2001; Zhan et al. 2008). The genes benC and benD are fused together with a C-terminal ligand binding domain (FCD) (Additional file 1: Figure S3) characteristic of GntR regulatory proteins. This domain is found in regulators of sugar biosynthesis operons that bind DNA through a helix-turn-helix (HTH) motif. There are two regulator coding genes; one upstream (R₁) and other downstream (R₂). The regulator R₁ belongs to LsyR regulators while R₂ belongs to the LuxR unlike most LsyR regulators common to catechol and benzoate catabolism operons (Collier et al. 1998; Ezezika et al. 2006). LuxR regulators have a characteristic alpha-helical protein domains fold with sequence-specific DNA binding domains. R₁ showed 35% identity to ben and cat operon transcriptional regulator from Acinetobacter baumannii (SCZ15558.1) (Zhan et al. 2008), 37% to uncharacterized HTH-type transcriptional regulator YnfL (P77559.1) (Aiba et al. 1996) and 33% to HTH-type transcriptional regulator BenM (O68014.2) (Collier et al. 1998). R₂ also showed 59% to HTH type of regulators Rv0890c (P9WMG1.1) and Mb0914c (P59969.1) which are mycobacterial proteins.

In addition to degradation of xenobiotics, Gordonia members are known to be a source of natural products. A total of 53 putative BGCs were predicted in the genome. Predicted Nrps clusters shared little similarity to known gene clusters such as chartreusin (15%), cahuitamycins (12%), kanamycin (2%), echosides (11%). Predicted terpene clusters showed 6% and 62% similarity with SF2575 and sioxanthin biosynthetic gene clusters respectively. Sioxanthin is a class of glycosylated carotenoids reported in the genus of Salinispora; marine actinomycetes (Richter et al. 2015). Although the genus of Gordonia is well known for carotenoid biosynthesis, no clusters were annotated as carotenoid BGCs. However, 1 arylpolyene and 1Type I PKS cluster were predicted. Arylpolyene are class of pigments that are functionally similar to carotenoids and protect the bacterium from reactive oxidation (Schöner et al. 2016). Arylpolyenes are polyketide derivatives each contain a terminal aryl moiety connected to a polyene carboxylic acid. To our knowledge, arylpolyene production has not been experimentally proven but was also predicted in G. lacunae BS2^T (Durrell et al. 2017).

Annotated ectoine cluster shared 75% similarity to BGC0000853_c1 ectoine cluster. The predicted ectoine cluster comprised of acetyl glutamate-5-kinase, TetR family transcriptional regulator, haloalkane dehalogenase, ectoine synthase, aminotransferase class III and GCN5 related-acetyltransferase genes. Ectoine is one many compatible solutes that protect and stabilizes cell membranes, proteins and DNA against drying and high temperatures, high osmolarity, and even pollutants (Bursy et al. 2008; Graf et al. 2008; Nakayama et al. 2000; Schnoor et al. 2004). In addition to ectoine biosynthesis, more genes coding for betaine, glycine and trehalose synthesis and transport were predicted in the genome. This explains the halotolerant nature of strain YC-RL2.

Some of the BGCs annotated as putative shared similarity with known BGCs for instance bacillomycin (20%), coelichelin (27%), nataxazole (7%), rifamycin (3%) and paromomycin (5%). The ability to produce diversified secondary metabolites would not only enable the strain to survive in different ecological niches but also be harnessed for therapeutical, industrial and pharmaceutical uses.

Metals have a central role in the life processes of microorganisms. Some metals such as copper, iron, and zinc are essential in biochemical, structural and osmotic processes (Sterritt and Lester 1980). However, high metal concentration can be detrimental to microorganisms by displacement of essential metals from their native binding sites or through ligand interactions. Bacteria adapt to metals through a variety of genome-mediated resistance systems (Hassen et al. 1998; Tchounwou et al. 2012). Some Gordonia strains such as Gordonia sp WQ-01A, G. amicalis HS-11 and G. amarae MTCC 4818 are known to reduce some metals by formation of nanoparticles (Sowani et al. 2018).

The present study reveals that strain is metabolically versatile and has a high potential for secondary metabolites can be harnessed for various clinical and industrial applications. Presence of diverse dioxygenases, esterase/hydrolase and metal resistant genes provide insight into the ability of this strain to survive in varied ecological niches. Cloning and expression of esterases involved in PAEs hydrolysis is currently underway. Based on these results, strain YC-RL2 genome will provide a rich genetic resource for further biotechnological and remediation studies.