Biosurfactants, which are mainly derived from the secondary metabolites of microorganisms (Nie et al., 2012), contain hydrophilic and hydrophobic groups in their structure. They harbor surface activities and emulsification activities that increase the bioavailability and degradation rates of hydrophobic substrates, including hydrocarbons (Desai and Banat, 1997). They are a structurally diverse class of compounds, with glycolipids and lipopeptides reported as two primary isolated families. So far, researches regarding the effects of biosurfactants on the bioremediation properties of Achromobacter strains are scarce. It has been demonstrated that surfactants SDS, Tween 60 and Tween 80 (derived from chemical processes) can enhance the detoxification of chromium-containing slag by Achromobacter sp. CH-1 (Chai et al., 2009).

Strain HZ01 is capable of producing biosurfactants that efficiently emulsify diverse hydrophobic compounds, exhibiting the potential for bioremediation (Deng et al., 2016). It has been demonstrated that citric acid is related to biosurfactant production (Pirog et al., 2008). In this work, the emulsification activities of the culture broth using citric acid as the sole carbon source were investigated. Although the fermentation broth was not concentrated, stable emulsions were formed using soybean oil, coconut oil, diesel oil, kerosene, and hexane as emulsion substrates, respectively (Figure 3), indicating that biosurfactants might exist in the fermentation broth. Further studies are needed to confirm the biosurfactant production in strain HZ01 using citric acid as the sole carbon source. Diesel oil and kerosene, both derived from petroleum, contain diverse hydrocarbons. The emulsification of diesel oil, kerosene, and hexane suggests that the fermentation products produced by strain HZ01 may promote its utilization of hydrocarbons.

Genome annotation revealed that a total of 197 genes were assigned to “secondary metabolites biosynthesis, transport, and catabolism” (Supplementary Table S6). The genome harbors candidate genes associated with biosurfactant production, such as 3-oxoacyl-ACP reductase, acyltransferase, phosphomannomutase, and glycosyltransferase (Supplementary Table S7), which are essential for glycolipid synthesis (Maier and Soberon-Chavez, 2000). Eleven genes were annotated as LuxR family transcriptional regulators that play an important role in the regulation of glycolipid synthesis (Satpute et al., 2010). Genes coding for cell envelope biogenesis protein OmpA (gene_2469) and outer membrane protein OmpAb (gene_3528) were identified. OmpA is an active constituent of alasan (a bioemulsifier) and has been demonstrated to be involved in the degradation of hydrophobic compounds, including hydrocarbons (Walzer et al., 2006). OmpH (gene_1336), which is possibly related to bioemulsifier production (Schneiker et al., 2006), is also present in strain HZ01. Generally, lipopeptides are synthesized through the ribosome-independent pathway using non-ribosomal peptide synthetases (NRPSs) (Roongsawang et al., 2010). One gene (gene_1007) was assigned to “non-ribosomal peptide structures.” Additionally, thioesterases and peptide synthetases, both related to lipopeptide biosynthesis (Yao et al., 2003; Moyne et al., 2004), were also identified (Supplementary Table S7). Strain HZ01 was previously reported to produce a cyclic lipopeptide containing the peptide chain of Gly-Gly-Leu-Met-Leu-Leu linked to a C₂₁ fatty acid chain (Deng et al., 2016). We infer that the biosynthetic pathway of this biosurfactant includes three major processes (Supplementary Figure S7), i.e., (a) a 3-hydroxy-heneicosanoic acid is derived from specific carbon sources via fatty acid biosynthesis; (b) the hexapeptide is obtained by a series of enzymatic condensations from the N-terminal of Leu to the C-terminal of Gly; and (c) the 3-hydroxy-heneicosanoic acid may undergo an enzymatic condensation process, being incorporated at the C- and N-terminals of the hexapeptide to produce a cyclic lipopeptide. Further studies are needed to clarify this biosynthetic pathway.

In summary, the identified genes provide a genetic basis for biosurfactant production in strain HZ01, which may facilitate its uptake and degradation of petroleum hydrocarbons.

Petroleum is a complex mixture primarily containing saturated hydrocarbons, PAHs and asphaltenes, of which alkanes are its major constituents (Rojo, 2009). Strain HZ01 is capable of degrading C₁₂–C₂₇ n-alkanes and PAHs (Deng et al., 2014). Its genome contains numerous genes for hydrocarbon degradation. Methyl-accepting chemotaxis proteins (MCP; gene_441, gene_1384, gene_1386, gene_1387, and gene_2249), related to alkane chemotaxis (Smits et al., 2003), may be beneficial for strain HZ01 to seek hydrocarbons as substrates. Cytochrome o ubiquinol oxidase (Cyo; gene_2136, gene_2137, gene_2138, gene_2181, gene_2182, gene_2183, and gene_2184) exhibits a negative regulation over the expression of alkane degradation genes, and there is a marked correlation between the Cyo levels and the repression extents of alkane degradation pathway (Dinamarca et al., 2003). Transcriptome sequencing showed that the transcriptional level of Cyo in strain HZ01 was not affected during petroleum degradation (Hong et al., 2016). The Cyo is not reported in hydrocarbon-degrading Alcanivorax borkumensis SK2 (Schneiker et al., 2006), Oleispira antarctica RB-8 (Kube et al., 2013), Geobacillus thermodenitrificans NG80-2 (Feng et al., 2007), Polymorphum gilvum SL003B-26A1^T (Nie et al., 2012), or Desulfatibacillum alkenivorans AK-01 (Callaghan et al., 2012).

n-Alkanes are usually subjected to degradation via the terminal or subterminal oxidation pathway (Rojo, 2009). In the terminal oxidation pathway, alkane degradation is initiated by oxygenation of a terminal methyl group to generate a primary alcohol, which is converted into its corresponding aldehyde and is further oxidized to a fatty acid. Subsequently, the fatty acid combines with CoA and is subjected to β-oxidation to produce acetyl-CoA. The acetyl-CoA is further processed by the TCA cycle to finally generate H₂O, CO₂, and energy. Diverse enzymes involved in this process were identified in strain HZ01, such as oxygenases, hydroxylases, alcohol dehydrogenases, aldehyde dehydrogenases, and acyl-CoA synthetases (Supplementary Table S8). Representative alkane hydroxylase AlkB and its homologs that initiated the oxidation of medium-chain alkanes were not identified in the genome, which was consistent with the genome features of oil-degrading Geobacillus thermodenitrificans NG80-2 (Feng et al., 2007) and Polymorphum gilvum SL003B-26A1^T (Nie et al., 2012). The lack of AlkB and its homologs in hydrocarbon-degrading marine bacteria is not uncommon (Venter et al., 2004). Some bacteria use the enzymes of cytochrome P450 family to perform similar functions as that of AlkB and its homologs (Rojo, 2009). One gene coding for cytochrome P450 (gene_1029) in strain HZ01 was identified. Its sequence showed an identity of 60% to the sequence of cytochrome P450 monooxygenase in Pseudomonas sp. Lz4W (GenBank accession no. EMI05095.1), suggesting that the cytochrome P450 of strain HZ01 may have similar functions as that of strain Lz4W. RNA-seq demonstrated that two transcripts of cytochrome P450 were up-regulated after petroleum treatment (Hong et al., 2016). Additionally, ferredoxin and ferredoxin reductase, required for the electron transfer from NAD(P)H to cytochrome P450, were also present in the genome (Supplementary Table S8), contributing to the alkane oxidation. Thus, the cytochrome P450 may play an important role in the initial oxidation of n-alkanes in strain HZ01. Comparative analysis revealed that, although strains NG80-2 and SL003B-26A1^T lacked AlkB and its homologs, long-chain alkane monooxygenase LadA was included in their genomes for alkane degradation (Feng et al., 2007; Nie et al., 2012). LadA belongs to the bacterial luciferase family and plays an important role in the initial oxidation of C₁₅–C₃₆ n-alkanes (Feng et al., 2007). On the contrary, Alcanivorax borkumensis SK2 and Oleispira antarctica RB-8 contained AlkB and its homologs, as well as cytochrome P450, whereas LadA was not identified (Schneiker et al., 2006; Kube et al., 2013). LadA was also not identified in strain HZ01 by genome sequencing. Nevertheless, RNA-seq showed that strain HZ01 contained one transcript coding for luciferase-like monooxygenase family protein 1 (Hong et al., 2016), which might have the functions similar to those of LadA.

Fatty acids are the major intermediate products of alkane degradation and are processed by β-oxidation to generate acetyl-CoA (Rojo, 2009). Besides acyl-CoA synthetase, which catalyzes the conjugation of fatty acid to CoA, fatty acid hydroxylase (gene_696, gene_1843, and gene_2856) is also contained in strain HZ01. The fatty acid hydroxylase catalyzes the hydroxylation of fatty acids, generating hydroxy fatty acids. Subsequently, the hydroxy fatty acids are converted into dicarboxylic acids, followed by β-oxidation (Coon, 2005). A total of 40 genes in strain HZ01 were assigned to the “fatty acid degradation” pathway, including the genes related to β-oxidation (Supplementary Table S9 and Figure S8). RNA-seq demonstrated that the genes involved in β-oxidation were activated after petroleum treatment (Hong et al., 2016). In addition, an intact TCA cycle pathway for processing the resulting acetyl-CoA is present in strain HZ01 (Supplementary Figure S4).

A total of 35 genes were assigned to the “methane metabolism” pathway (Supplementary Table S10). For instance, strain HZ01 contained fdoG, fdfH, fdoH, and fdoI coding for formate dehydrogenase subunits. However, some enzymes essential for methane oxidation were not identified, including methane monooxygenase, methanol dehydrogenase, and formaldehyde dehydrogenase. Strain HZ01 contains esterases, but the Baeyer-Villiger monooxygenase, an essential enzyme involved in the cycloalkane degradation and the subterminal oxidation of n-alkanes, was not included in the genome. The identified genes suggest that (i) subterminal oxidation pathway is not present in strain HZ01; (ii) this bacterium cannot perform methane oxidation or cycloalkane degradation; and that (iii) the degradation of n-alkanes in strain HZ01 is performed via the terminal oxidation pathway (Figure 4). The results of genome sequencing further verify the pathway for n-alkane degradation predicted by RNA-seq (Hong et al., 2016). The metabolic reaction initiated by fatty acid hydroxylase (Figure 4) was not included in the previous predicted pathway. Besides, RNA-seq showed that the following steps of terminal oxidation pathway were activated after strain HZ01 was treated with petroleum for 16 h: (i) the initial oxidation of n-alkanes; (ii) reactions catalyzed by dehydrogenases and acyl-CoA synthases; and (iii) fatty acid β-oxidation (Hong et al., 2016).

In the previous RNA-seq, some transcripts related to the degradation of aromatic compounds were identified in strain HZ01 (Hong et al., 2016). Nevertheless, the potential pathways for degradation of aromatic compounds in this bacterium have not been revealed so far.

Genome sequencing of this study showed that abundant genes related to the degradation of aromatic compounds were contained in strain HZ01. For instance, a total of 28 genes were assigned to “degradation of aromatic compounds” (Supplementary Table S11). During the degradation of aromatic compounds, reactive dihydroxylated intermediates, such as catechol and protocatechuate, will be produced. These intermediates are further degraded via the intradiol (ortho) or extradiol (meta) ring cleavage (Strachan et al., 1998). The resulting products of the ortho-pathway (acetyl-CoA and succinyl-CoA) and the meta-pathway (pyruvate and acetaldehyde) are further processed by the TCA cycle (Pessione et al., 1999).

The homogentisate pathway, the protocatechuate and catechol branches of the β-ketoadipate pathway, and the phenylacetate pathway are four common pathways for the degradation of aromatic compounds (Jimenez et al., 2002). The genome sequencing of this work showed that only PhaA and PhaB, both belonging to the enoyl-CoA hydratase/isomerase family proteins, were identified in the phenylacetate pathway (Figure 5).

Protocatechuate and catechol are representative intermediates of the protocatechuate and catechol branches of the β-ketoadipate pathway, respectively, and these two intermediates are generally produced during the degradation of aromatic compounds. For instance, the catechol intermediate may be generated from benzoate, benzamide, (chloro)-biphenyl, benzonitrile, benzaldehyde, salicylate, and mandelate (Chain et al., 2006). Both the protocatechuate and catechol branches will produce β-ketoadipate enol-lactone, which is sequentially processed by PcaD (3-oxoadipate enol-lactonase; gene_4113), PcaIJ (not identified) and PcaF (beta-ketoadipyl CoA thiolase; gene_443). Then, acetyl-CoA and succinyl-CoA are generated and further metabolized by TCA cycle (Jimenez et al., 2002). PcaC (4-carboxymuconolactone decarboxylase; gene_4173) catalyzed the production of β-ketoadipate enol-lactone from γ-carboxymuconolactone and was identified in the protocatechuate branch (Figure 5). Nevertheless, PcaGH and PcaB were not identified. In the upper-stream of the protocatechuate branch, Acd (acyl-CoA dehydrogenase) and VanB (vanillate monooxygenase ferredoxin subunit) are present in strain HZ01. Acd and VanB play important roles in the production of vanillate from feruloyl-CoA and protocatechuate from vanillate, respectively.

So far, some Achromobacter strains capable of degrading aromatic compounds have been reported. Catechol dioxygenases play an important role in the degradation of aromatic compounds (Moon et al., 1997). Similar to strain HZ01, Achromobacter sp. DMS1 and Achromobacter xylosoxidans DN002 contain catechol 1,2-dioxygenase (Supplementary Table S12). Achromobacter xylosoxidans KF701, strain DN002 and Achromobacter xylosoxidans T7 contain catechol 2,3-dioxygenase, which is not found in strain HZ01. The catechol branch of the β-ketoadipate pathway is comparatively intact than the protocatechuate branch in strain HZ01. CatA (catechol 1,2-dioxygenase; gene_376) performs the intradiol cleavage of catechol and some of its derivatives. This enzyme also catalyzes the extradiol cleavage of meta-substituted substrates, such as 3-methoxycatechol and 3-methylcatechol (Strachan et al., 1998). Within the catechol branch, CatA transforms catechol into cis,cis-muconate, which is processed by CatB (muconate cycloisomerase; gene_772) to generate muconolactone (Figure 5). Then, CatC (muconolactone D-isomerase; gene_4111) catalyzes muconolactone to produce β-ketoadipate enol-lactone. In the upper-stream of catechol, BenA (benzoate/toluate 1,2-dioxygenase subunit alpha; gene_3355 and gene_3376) and BenC (benzoate/toluate 1,2-dioxygenase reductase component; gene_2862) were identified, but BenB and BenD were not present. BenABC can transform benzoate into benzoate diol, which is further metabolized by BenD to generate catechol (Figure 5).

Regarding the homogentisate pathway, only Mai (maleylacetoacetate isomerase; gene_2657 and gene_5130) was identified in the downstream of homogentisate. This enzyme catalyzes maleylacetoacetate to generate fumarylacetoacetate (Figure 5). In the upper-stream of homogentisate, PhhB (pterin-4-alpha-carbinolamine dehydratase; gene_3287), TyrB (aromatic amino acid aminotransferase; gene_533) and Hpd (4-hydroxyphenylpyruvate dioxygenase; gene_3903) were identified, but PhhA was not found. PhhA and PhhB play an important role in the transformation of phenylalanine into tyrosine (Jimenez et al., 2002). Then, the tyrosine is catalyzed by TyrB to generate 4-hydroxyphenylpyruvate, which is transformed into homogentisate by Hpd (Figure 5).

Additionally, the genes coding for benzoate/toluate 1,2-dioxygenase subunit alpha (gene_3355 and gene_3376) and benzoate/toluate 1,2-dioxygenase reductase component (gene_2862) were identified in strain HZ01. RNA-seq showed that the expression levels of benzoate 1,2-dioxygenase subunit alpha and benzoate 1,2-dioxygenase large subunit were not affected by petroleum treatment (Hong et al., 2016). The benzoate 1,2-dioxygenase system performs its functions via catalyzing the double hydroxylation of benzoates with the incorporation of two oxygen atoms (Yamaguchi and Fujisawa, 1980). Salicylate hydroxylase (gene_2661), a key enzyme in naphthalene catabolism, catalyzes the decarboxylation and hydroxylation of salicylate, generating catechol, H₂O, and CO₂ (You et al., 1991). Glutathione S-transferase (GST), related to PAHs degradation (Xia and Min, 2003), were also identified in strain HZ01 (Supplementary Table S8). Laccases (gene_3832), a family of blue multicopper oxidases, play an important role in the oxidation of numerous aromatic compounds (Fang et al., 2011). In addition, 150 genes in strain HZ01 were assigned to “xenobiotics biodegradation and metabolism” using KAAS annotation (Supplementary Table S13), such as genes related to the degradation of fluorobenzoate, toluene, xylene, nitrotoluene, and dioxin. Several genes encoding GntR family transcriptional regulators and MarR family transcriptional regulators related to aromatic compound degradation were also identified.

Strain HZ01 has been reported to degrade anthracene, phenanthrene, and pyrene (Deng et al., 2014). The genome sequencing in this study further suggests that strain HZ01 has the potential for bioremediation of aromatic-compounds-contaminated environments. Since the enzymes involved in the catechol pathway are comparatively complete, this pathway may play a major role in the degradation of aromatic compounds in strain HZ01. Further studies are needed to elucidate the aromatic compounds degradation in this bacterium.

Unsaturated fatty acids are related to the fluidity of cell membrane and contribute to the adaptation of bacteria to low-temperature environments and the low solubility of substrates (including hydrocarbons). A total of 12 genes in the genome of strain HZ01 were assigned to “biosynthesis of unsaturated fatty acids.” For instance, gene_4447 encodes fatty acid desaturase, which is existent in almost all organisms and is a key enzyme in the biosynthesis of fatty acids (Alonso et al., 2003). These genes play important roles in the fatty acid synthesis in strain HZ01 and may enhance its capability for hydrocarbon degradation. Other candidate genes associated with the degradation of petroleum components were shown in Supplementary Table S8, mainly including the genes coding for dehydrogenases, rubredoxins, oxidoreductases, monooxygenases, and dioxygenases. In summary, the identified genes and pathways provide a genetic basis for petroleum degradation in strain HZ01.

Besides, the analysis contents of this study also include (i) genomic comparison, (ii) amino acid metabolism, (iii) nitrogen metabolism, (iv) biosynthesis of other secondary metabolites, (v) antibiotic resistance, (vi) two-component systems, (vii) cell motility, and (viii) membrane transport (Supplementary Results and Discussion). According to the genes identified by genome sequencing, the material transport and metabolic pathways in Achromobacter sp. HZ01 were depicted in Figure 6.