Introduction

N-alkanes, one of the main components of petroleum, enter into marine environment through natural seep and human activites1. It has a strong ecological impact on the marine ecosystem because of its persistence2, potential for bioaccumulation and toxicity3, and ability to travel long distances with marine snow and particles4,5, even down to the bottom of the Mariana Trench at the water depth of 10,500 m6,7.

When the ocean ecosystem is exposed to emergency oil spills, microbial degradation is the dominant and ultimate natural removal process for n-alkanes. At the surface ocean, the canonical microbial n-alkane degradation normally begins with an activation reaction that introduces oxygen-containing functional groups. This step is typically mediated by hydroxylases and requires energy to overcome the chemical inertness, which is the initial and rate-limiting step in the n-alkanes degradation8,9. There are several hydroxylase systems responsible for the activation of different length of chain n-alkane. For example, soluble cytochrome P450s and integral membrane non-heme iron monooxygenases AlkBs predominately participate in the activation of n-alkane with chain length ranging from 5 to 17 carbons10. The putative flavin-binding monooxygenase AlmA of the genus Alcanivorax functions as the hydroxylase for n-alkane with chain length ranging from 22 to 36 carbons10,11. The oxidation step is related to electron transport and intracellular oxidative status. In the AlkB-type system, reductase (AlkT) and hemoglobin reductase (AlkG) are responsible for transferring two electrons from NADH to AlkB12. The intermediate fatty alcohols are subsequently oxidized into fatty aldehydes and fatty acids by dehydrogenase and then enter the β-oxidation and tricarboxylic acid (TCA) cycle for energy metabolism13.

In the deep ocean (> 1000 m water depth), the microbial degradation of n-alkanes is regulated by high hydrostatic pressure (HHP), and therefore reveals distinct patterns compared with that at surface ocean. At the cellular level, HHP can induce intracellular oxygen stress14, leading to a shift in anabolism (biomass synthesis) and catabolism (CO2 production, etc.), and altering the n-alkanes degradation efficiency and pathway in deep ocean as what have been reported from both in situ investigations and laboratory cultivations15,16,17,18. For example, in the cultivation of a deep sea sediment sample, the TCA cycle rate under 30 MPa can rarely meet high energy demand, resulting in a slow degradation of long-chain n-alkane17. In Marinobacter hydrocarbonoclasticus strain #5, HHP (35 MPa) had no impact on n-hexadecane(n-C16) degradation but inhibited the synthesis of n-hexadecane-derived wax esters that accumulated in the cells as individual lipid bodies19. Moreover, HHP has a selective impact on the dominant taxa of alkane degraders. Among all the identified alkane degraders, the genus Alcanivorax is the most widely distributed in the deep oceans, indicating its adaptation to HHP6,20. Laboratory enrichment under HHP also showed Alcanivorax was selectively enriched with n-alkane addition, indicating its competitive edge in the deep ocean7,21.

However, studies on the effect of hydrostatic pressure (HP) on the n-alkane biodegradation is affected by HP, in terms of its conversion fate, remain scarce primarily because of two limitations: (1) lack of ability to simulate extreme HHP in the deep ocean in the laboratory, resulting in the inability to HHP cultivation and detection; (2) lack of n-alkane-degrading bacteria that can grow under extreme HHP, resulting in no suitable biomaterial for research. In this study, we used Alcanivorax xenomutans A28, a piezotolerant strain with n-alkane degradation capacity isolated from 7663.5 m of deep-ocean sediments, for culture experiments under different HPs (0.1, 40, and 80 MPa) and at 4 °C in specially designed high-pressure incubation vessels. Chemical parameters, transcriptomics, and metabolomics were used to evaluate the microbial utilization of n-C16 and the degradation process of this strain under different HPs. This study enhances our understanding of the microbial degradation efficiency and mechanisms of n-alkanes and provides certain basic knowledge to assist the deep-ocean petroleum remediation.

Results

General characteristics of Alcanivorax xenomutans A28

The commonly known n-alkane-degrading genus Alcanivorax was abundant (from approximately 1.00% to 21.02% in the first 16 cm and 1.42% in the 0–2 cm layer) in the in situ sediment (Supplementary Fig. 1). A piezotolerant bacterium strain Alcanivorax xenomutans A28 was isolated from 0 to 2 cm sediment layer. Strain A28 was taxonomically affiliated with the family Alcanivoracaceae, genus Alcanivorax. 16S rRNA gene sequence analysis revealed 99.93% and 99.86% sequence similarity with Alcanivorax xenomutans JC109 and Alcanivorax dieselolei B5, respectively (Fig. 1A). Based on genomic average nucleotide identity (ANI), it exhibited 98.52% and 93.62% similarity with A. xenomutans JC109 and A. dieselolei, respectively. The complete genome consists of one chromosome with a total length of 4,710,204 base pairs (bp) with a G + C content of 61.44% (Fig. 1B; Supplementary Table 1). Strain A28 possess two 16S rRNA gene copies in its genome.

Fig. 1: Genetic tree based on 16S rRNA gene and circular genome of Alcanivorax xenomutans strain A28.
figure 1

A Phylogenetic tree based on 16S rRNA gene sequences showing the relationship of strain A28 within the genus Alcanivorax. The tree was reconstructed by the Neighbor-Joining method and rooted by using Ketobacter alkanivorans GI5 and Halomonas stenophila N12 that belong to the family Alcanivoracaceae as the outgroups. Numbers at nodes are the percentage of bootstrap support (based on 1000 replications). GenBank accession numbers for 16S rRNA gene sequences are shown in parentheses. Bar indicates 1 nucleotide substitution per 100 nt. B Circular representation of Alcanivorax xenomutans A28 genome. From the inside to the outside, the first circle represents the scale; The second circle represents GC Skew; The third circle represents the GC content; The fourth and seventh circles represent the COG (Cluster of Orthologous Groups of proteins) to which each CDS (Coding Sequence) belongs; The fifth and sixth circles represent the position of CDS, tRNA and rRNA on the genome.

Full size image

Genomic analysis suggested the n-alkanes degradation capacity of A. xenomutans A28. The genome harbored three types of n-alkane hydroxylase systems, cytochrome P450s (chr_2087 and chr_2372), the non-heme diiron integral membrane n-alkane monooxygenase genes alkB (chr_102, chr_700, and chr_4245), and the flavin-binding monooxygenase gene almA (chr_2065). The genes of two enzymes participating in electron transfer to AlkB, rubredoxin reductase gene alkT (chr_4201) and the rubredoxin alkG (chr_4202) were also annotated in genome (Table S2). The almA sequence of strain A28 showed 100% similarity with the amino acid sequence of almA from A. dieselolei B5 (Supplementary Fig. 2).

Degradation performance of Alcanivorax xenomutans A28 for n-hexadecane under different HPs

We incubated Alcanivorax xenomutans A28 in artificial seawater with and without n-C16 as the sole carbon source under different HPs (0.1, 40, and 80 MPa) at 4 °C. The growth of strain A28 was supported by the addition of n-C16 but minimally impacted by HP (Fig. 2; Supplementary Fig. 3). The cell numbers significantly increased with n-C16 compared to those without n-C16 (one-tailed T test, p < 0.05). Additionally, no significant differences in cell numbers were detected among the three HPs incubations (ANOVA analysis, p < 0.05). The cells under 0.1 MPa had a higher protein content than those under 40 and 80 MPa (Supplementary Fig. 4).

Fig. 2: Cell number on day 7 and 14 under different HPs (0.1 MPa, 40 MPa, and 80 MPa).
figure 2

The statistical significance analysis among three HPs with n-C16 addition was conducted by one-way ANOVA. Error bars represent standard deviation of three biological replicates under the same treatment.

Full size image

To test the biodegradation performance of strain A28 for n-C16 under different HPs, we quantified the n-C16 consumption rate, amount of CO2 production, and the complete mineralization ratio (amount of carbon of CO2 production to n-C16 consumption). HHP (40 and 80 MPa) inhibited n-C16 degradation (Fig. 3). After two weeks of incubation, the total removal of n-C16 reached 20%, 15%, and 11% under 0.1, 40, and 80 MPa, respectively. The calculated consumption rates based on the slopes of the trend lines were 0.46, 0.35, and 0.30 mg n-C16 per day under 0.1, 40, and 80 MPa, respectively. n-C16 consumption decreased with HP elevation, while CO2 production increased, resulting in a higher complete mineralization ratio under 40 and 80 MPa than that under 0.1 MPa (Fig. 3). The incubation under 80 MPa showed lower n-C16 consumption rate but higher CO2 production proportion compared with those under 0.1 and 40 MPa. The averaged mineralization ratios (calculated from four sampling points, see the details in Methods) ranged from 0.0039 ± 0.0013‰ (mean ± s.d.), 0.0059 ± 0.0012‰, and 0.0090 ± 0.0012‰ under 0.1, 40, and 80 MPa, respectively. This pattern was consistent with the pH changes caused by CO2 production (Supplementary Fig. 5).

Fig. 3: Amount of n-C16 consumption, CO2 production and mineralization ratio by strain A28 under different HPs (0.1, 40, and 80 MPa).
figure 3

A Amount of n-C16 consumption under different HPs (0.1, 40, and 80 MPa). The dotted lines are the fitting lines by simple liner regression and the grey box shows the fitting slope and goodness of fit. B The CO2 production and mineralization ratio based on the amount of carbon of n-C16 consumption to CO2 production under different HPs. Error bars represent standard deviations of three biological replicates under the same HP.

Full size image

Cellular metabolism of n-hexadecane oxygenation under different HPs

Using transcriptome and metabolome analyses, we examined throughout the n-alkane degradation pathway of strain A28 and focused on n-alkane oxygenation steps. The n-alkane hydroxylases were downregulated under 40 and 80 MPa compared with those under 0.1 MPa (Fig. 4; Supplementary Table 2). Among the three types of n-alkane-degrading enzyme genes, alkB was the most severely affected at the transcriptional level under HHP. The expression of alkB in chr_700 was significantly downregulated under 80 MPa (log2 FC: −2.07 on day 7) and 40 MPa (log2 FC: −4.57 on day 7 and −2.96 on day 14). In addition, two other genes of AlkB-depending enzymes, alkG and alkT, were also significantly downregulated under HHP. In particular, alkG was significantly downregulated both on day 7 and 14 under 80 MPa than that under 0.1 MPa. The putative almA was significantly upregulated under 40 MPa, with 1.48 log2 FC on day 7 and 1.17 on day 14 (Fig. 4; Supplementary Table 2).

Fig. 4: Transcriptional profiles of key genes related to n-alkane oxidation pathway, β-oxidation of fatty acid and the TCA cycle in strain A28 under different HPs (0.1, 40, and 80 MPa).
figure 4

Arrows represent direction of biochemical reactions. The number in the block represents the Log2FC of gene expression under HHP (40 or 80 MPa) compared with that under 0.1 MPa. Color of number indicates different sampling points that green represents the day 7 and purple represents the day 14. Block with pink or blue color indicates the transcriptional level of differentially expressed genes compared with that under 0.1 MPa. Pink block indicates significantly upregulation (Log2FC ≥ 1, adjusted p value < 0.05) while blue block indicates significantly downregulation (Log2FC ≤ -1, adjusted p value < 0.05). Each transcriptional analysis involved three biological replicates under the same HP.

Full size image

Conversion of the aldehyde into fatty acid during n-alkane oxygenation is carried out by alcohol dehydrogenases (ADHs) and aldehyde dehydrogenases (ALDHs). We identified five ADH homologues (chr_245, chr_405, chr_1636, chr_2506, and chr_2884) and two ALDH homologues (chr_1622 and chr_4234) in strain A28 genome. Among the five ADH homologues, the ADH homologues chr_405 and chr_245 were significantly upregulated under 40 MPa, and chr_245 was only upregulated under 80 MPa, compared with those under 0.1 MPa (Fig. 4). Both two ALDH homologues, chr_1622 and chr_4234, were only significantly upregulated under 80 MPa, whereas they were unaffected under 40 MPa, compared with those under 0.1 MPa.

Using metabolome analysis, we identified the potential hexadecane intermediates produced during n-alkane oxygenation. The final oxidation product hexadecenoic acid showed no significant differences among the three HPs, indicating that the net production of hexadecenoic acid was unaffected by HP (Fig. 5A). Lipid-related products, such as PG (16:1(9Z)-16:0), PE (34:2), PE (20:1(11Z)-14:1(9Z)) showed higher concentrations under 0.1 MPa than HHP (ANOVA, p < 0.01; Supplementary Fig. 6).

Fig. 5: Relative content of potential metabolites in n-C16 degradation.
figure 5

Relative content of potential hexadecane-derived metabolites (A) and TCA cycle related intermediate (B, C, D). The relative content was the peak area normalized to total peak area. The significant analysis was conducted by unpaired and two-tailed T test. Box plots indicate upper and lower quartiles (top and bottom of the box), median (line within box), and the dot represents the value of each biological replicate. Error bars represent standard deviation of six replicates under the same HP.

Full size image

Energy metabolism under different HPs

In the initial step of β-oxidation, seven long-chain acyl-CoA synthetase homologues were identified in strain A28 (chr_1199, chr_1604, chr_2040, chr_2190, chr_1909, chr_3554, and chr_3666). The gene expressions of long-chain acyl-CoA synthetases were generally upregulated under HHP. Gene chr_1199 was significantly upregulated under both 40 MPa (day 14) and 80 MPa (day 7 and 14), and chr_3666 was significantly upregulated under 80 MPa (day 7 and 14). Gene chr_1604 was significantly upregulated under 40 MPa (day 14). No significant differences were observed in the expression of chr_2040 and chr_2909. Among the seven long-chain acyl-CoA synthetase homologues, chr_2190 was the only one that was significantly downregulated under 80 MPa (log2FC = -1.21 on day 7 and -1.42 on day 14, adjusted p < 0.05) compared with that under 0.1 MPa.

For other steps of β-oxidation and the entire TCA cycle, 25 of 39 genes were significantly upregulated and no gene was significantly downregulated under 80 MPa; whereas, 9 of 39 genes were significantly upregulated and 15 of 39 genes were significantly downregulated under 40 MPa (Fig. 4). Sixteen genes were significantly differentially expressed both under 40 MPa and 80 MPa, compared with those under 0.1 MPa. These 16 genes were all significantly upregulated under 80 MPa and 11 genes were downregulated under 40 MPa, compared with those under 0.1 MPa.

In the TCA cycle, HHP stimulated all the genes related to two CO2-producing steps. In the reaction from isocitrate to α-ketoglutarate, two isocitrate dehydrogenase homologues (chr_687 and chr_1990) were significantly upregulated under 40 MPa (log2FC > 1.67, adjusted p < 0.05) and 80 MPa (log2FC > 1.08, adjusted p < 0.05), compared with that under 0.1 MPa. For the step from α-ketoglutarate to succinyl-CoA, α-ketoglutarate dehydrogenase homologues (chr_2681, chr_2682, and chr_2683) were upregulated under 80 MPa (log2FC > 1.18, adjusted p < 0.05) while downregulated under 40 MPa (log2FC < -1.46, adjusted p < 0.05), compared with those under 0.1 MPa. On the contrary, the expression of glyoxylate shunt (GS) related genes (chr_3263 and chr_1968) was significantly downregulated under 40 MPa (chr_3263: log2FC < -1.82, adjusted p < 0.05; chr_1968: log2FC = −1.14, adjusted p < 0.05 on day 7) and only malate synthase homologues chr_1969 was upregulated under 80 MPa. Three identified TCA-related intermediates, citrate, cis-Aconitase, and succinyl-CoA, all had minimum concentrations under 80 MPa among the three HPs except for succinyl-CoA on day 7 (Fig. 5B–D). Besides, the expression levels of homologous involved in oxidative phosphorylation machinery, such as ATP synthase, succinate dehydrogenase, cytochrome bc complex, and cytochrome bd complex, were significantly upregulated under 80 MPa (Supplementary Fig. 7). This further indicated the intensified energy metabolism during n-alkane biodegradation under elevated HPs.

Oxidative stress level of Alcanivorax xenomutans A28 under different HPs

HHP triggered a higher reactive oxygen species (ROS) production and upregulated antioxidant related genes (Fig. 6). Intracellular H2O2 accumulated with incubation time. The concentration of intracellular H2O2 between the incubation under 0.1 MPa and 40 MPa was closely similar in the first three sampling points (p = 0.1163, paired T test) and the largest difference reached 2.98 × 10−8 μM per cell at the last sampling point day 14. Among the three HPs, the concentration of intracellular H2O2 under 80 MPa was higher than those under 0.1 and 40 MPa (except for day 4). The total antioxidant capacity quantified by the ferric reducing ability of the plasma (FRAP) method showed the cells under HHP possessed similar levels of total antioxidant capability (one-way ANOVA analysis, F(1.182, 3.545) = 0.5880, p = 0.5189; Fig. 6B).

Fig. 6: Oxidative stress response of Alcanivorax xenomutans A28 under different HPs (0.1 MPa, 40 MPa, and 80 MPa).
figure 6

A The concentration of intracellular H2O2. The concentration was normalized to cell number. B Total antioxidant capacity by the FRAP method. The total antioxidant capacity is expressed as the concentration of the FeSO4 standard solution. The concentration was normalized to cell number. The RM (repeated measured) One-Way ANOVA was conducted among three HPs. Error bars (A, B) represent standard deviation of three biological replicates under the same HP. C Transcriptional heat map of key genes related to antioxidant system. The number inside box indicates the value of log2 FC in the transcriptional comparison. The asterisk (*) indicates the transcriptional level gene is significantly differentially expressed with |log2 FC|≥ 1 and adjusted p-value < 0.05.

Full size image

For the genes related to antioxidant system, an overall up-regulation under HHP was observed (Fig. 6C). The sod1 gene (chr_3251), encoding for enzyme that participates in O2− removal to H2O2, was significantly upregulated under 40 MPa (log2FC = 2.24 on day 7 and 2.49 on day 14, adjusted p < 0.05) and 80 MPa (log2FC = 1.03 on day 7 and 1.99 on day 14, adjusted p < 0.05), comparted with that under 0.1 MPa. The ahpC gene (chr_3156), encoding for enzyme that directly participates in H2O2 removal, was also significantly upregulated under 40 MPa (log2FC = 1.44 on day 7 and 2.46 on day 14, adjusted p < 0.05) and 80 MPa (log2FC = 1.39 on day 7, adjusted p < 0.05). The trxB gene(chr_1998), encoding for thioredoxin reductase that provides electrons to thiol-dependent peroxidases, was significantly upregulated under 40 MPa (log2FC = 3.09 on day 7 and 1.23 on day 14, adjusted p < 0.05). Additionally, higher gene expression levels of trxB, ahpC and sod1 were observed under 40 MPa than those under 80 MPa.

Discussion

HHP slows down the microbial degradation of n-hexadecane by inhibiting the oxygenation of n-alkane

In this study, the effect of HP on n-C16 degradation was observed up to 80 MPa and HHP negatively affected the degradation of n-C16. This is consistent with the results of previous studies conducted under lower HP levels. Both laboratory cultures and field investigations have indicated that HHP (10 and 20 MPa) or water depth (down to 4000 m water depth where the HP is approximately 40 MPa) can inhibit the degradation of organic carbon such as n-alkanes by microorganisms (especially piezotolerant microorganisms)17,22. Furthermore, our findings indicate that HHP inhibits the rate of n-alkane degradation. This led to the discovery that oil pollutants, such as n-alkanes, undergo slower biodegradation when sinking into the deep ocean than those in the surface ocean. Additionally, a higher proportion of long-chain n-alkane (> C26) preserved during the four years following the Mexico oil spill accident15. Our study further confirms the important role of HP in deep-ocean n-alkane degradation.

Previous studies indicated the inhibited heterotrophic activity by HHP may be related to a decreased growth rate17,22. In this study, no significant impact from HHP was observed on the growth rate of strain A28 (Fig. 2). This indicates that HHP inhibits the n-alkane degradation rate by regulating intracellular processes rather than by controlling the population size. Using transcriptome and metabolome analyses, we found that the oxygenation of n-alkane was the potential bottleneck for n-C16 consumption under HHPs. The n-alkane oxygenation is generally the initial and rate-limiting step in aerobic n-alkane degradation, as n-alkane-activating enzymes must overcome the low chemical reactivity of n-alkanes13. It is catalyzed by n-alkane hydroxylases. Strain A28 has three n-alkane hydroxylase systems (AlkB, putative AlmA and cytochrome P450). These n-alkane hydroxylases have different substrate utilization ranges or induction patterns4.The enzymes AlkB and cytochrome P450 primarily degrade short- and medium-chain n-alkane (C5-C17)10,11, while AlmA is more involved in the degradation of longer chain n-alkane (>22 C)11. AlkB-type n-alkane hydroxylases exhibit greater affinities for C8-C20 substrates than cytochrome P45023, and were the most transcriptionally sensitive n-alkane hydroxylases under HHP in this study (Fig. 4; Supplementary Table 2). The enzyme AlkB is a class of non-heme diiron enzyme whose activation requires the transfer of two electrons from NADH via reductase AlkT and hemoglobin reductase AlkG12. The downregulation of alkB and alkG further suggests that the AlkB system was generally inhibited by HHP and may limit the oxygenation rate for n-C16 (Fig. 4; Supplementary Table 2). The inhibitory effect of HHP on n-alkane hydroxylase gene alkB has been reported in a few studies, the underlying mechanism remains underexplored24. The accumulation of ROS, such as H2O2, inhibits the activation of n-alkane oxygenases25. Based on our results, we propose that the function of AlkB was inhibited by the high ROS level, induced by HHP incubation (Fig. 6).

Notably, the n-alkane degradation rate was higher under 40 MPa than that under 80 MPa despite the observed downregulation of alkB gene expression. This discrepancy suggests that potential isozymes of AlkB may be functioning, e.g. AlmA26. The AlmA in strain A28 shares 100% sequence similarity with AlmA in A. dieselolei B5 (Supplementary Fig. 2); the latter has been proven to degrade n-C16 based on enzymatic activity assay27. The almA was significantly upregulated under 40 MPa compared with that under 0.1 and 80 MPa (Fig. 4; Supplementary Table 2). The unusually high expression of almA under 40 MPa when degrading n-C16 may help to offset the inhibition of alkB by HHP, resulting in a slightly higher degradation rate of n-C16 than that under 80 MPa. The high expression level of almA suggests its functioning advantage under HHP. The enzyme AlmA is a class of flavin-dependent enzymes. Due to the low electrophilicity, flavin-dependent enzymes often have a low sensitivity to H2O2, and can even directly eliminate H2O2 in some cases25. As shown in our results, the expression level of almA in deed has the similar trend as the H2O2 level (Fig. 6A). However, since the function of the AlmA-type hydroxylase involved is far from well understood, the relationship between its expression with HPs needs to be further investigated.

HHP accelerates the complete mineralization process of microbial n-hexadecane degradation by stimulating energy metabolism

The ratio of CO2 production to n-C16 consumption increased with HP elevation, despite the decrease in n-C16 degradation rate (Fig. 3). This suggests that HHP stimulates the complete mineralization of n-C16. Previous studies generally assumed that high CO2 production may be indicative of a high degradation rate of organic carbon28. However, whether this relationship remains stable under different HPs has yet to be investigated. Herein, we demonstrated the lack of a positive correlation between the amount of CO2 production and n-C16 consumption under different HPs (Fig. 3). It should be noted that since a sacrificial sampling method was used (see the details in Material and Method), biological batch effects between samples were inevitably introduced, which prevented us from calculating the carbon balance through cultivation periods. Therefore, we could only compare the CO2 accumulation at the same time point among different HPs. Nevertheless, the observed phenomenon also indicates that the removal of same amount of n-alkanes will generate more CO2 in deeper ocean environments.

The n-alkanes are used by microorganisms for anabolism (composition of biomass) and catabolism (production of CO2, etc.)8. The varying ratios of CO2 production to n-C16 consumption indicates that HHP changes the allocation of n-alkanes for anabolism and catabolism. That is, the proportion of n-C16 for catabolism increased under HHP, whereas that for anabolism decreased. Through the metabolomic analysis, we identified several metabolites involving in the construction of cell structures and membranes that were significantly depleted under 80 MPa (Supplementary Fig. 6). Additionally, the culture under 0.1 MPa exhibited a higher cellular protein content than that under HHP (Supplementary Fig. 4). These results further suggests that biomass synthesis is weakened under HHP. For example, hexadecenoic acid serves as substrates for biosynthesis, and an intensified biosynthesis process would cause less accumulation of hexadecenoic acid under 0.1 MPa. This explains why the cells under HHP consumed less n-C16 and transferred a higher proportion to CO2 but still accumulated same levels of hexadecenoic acid as those under 0.1 MPa (Fig. 5A).

For catabolism, n-C16 is completely degraded through β-oxidation and the TCA cycle, where CO2 is the final product. The complete oxidation of 1 mole n-C16 to CO2 can contribute a net reducing power of 106 mole ATP, 31 mole NADH and 14 mole FADH2, contributing to high energy and reducing power29. The expression of genes associated with β-oxidation and the TCA cycle were upregulated under 80 MPa, indicating that HHP stimulated catabolism (Fig. 4). The low intracellular concentrations of the relevant intermediate metabolites under HHP also further supported the high cycling rate of the TCA cycle (Fig. 5B–D). In particular, CO2 production through the TCA cycle under HHP was significantly upregulated compared with that under 0.1 MPa (Fig. 5). The glyoxylate shunt, which is generally activated during alkane degradation, was downregulated overall under 40 MPa and unaffected under 80 MPa (Fig. 4). The glyoxylate shunt can bypass the TCA cycle, contributing to less CO2 and NADH/FADH30. Upregulation of CO2-producing genes in the TCA cycle and avoidance of glyoxylate shunt contribute to a high rate of n-C16 mineralization to CO2. This may be related to the increased energy demand under HHP, which is evidenced by the upregulation of genes involved in oxidative phosphorylation for ATP synthesis under 80 MPa (Supplementary Fig. 7). This is consistent with the theory that the upregulation of energy metabolism is a common strategy to cope with oxidative stress induced by HHP31,32. Previous studies found that Halomonas titanicae ANRCS81 suffered an increased oxidative stress when cultivated under 40 MPa compared to that under 0.1 MPa, accompanied by the upregulation of TCA cycle-related genes32. This is consistent with the elevated H2O2 and the upregulation of antioxidant enzyme genes including sod1, ahpC (Fig. 6). Thus, enhanced HP promotes the catabolism of n-C16 under HHP.

Summary

In this study, we reported the regulatory mechanism of a piezotolerant strain Alcanivorax xenomutans A28 for n-C16 biodegradation up to 80 MPa. We found that HHP significantly inhibited the oxygenation of n-alkane by inducing intracellular oxidative stress, leading to a reduced n-alkane degradation rate. The high oxidative stress under HHP also accelerated the complete oxidation of n-alkane, resulting in a high mineralization ratio. These findings greatly enhanced our understanding for the natural capacity of n-alkanes bioremediation and the fate of n-alkanes at different water depths. This study provides a new perspective on ecological risk evaluation for deep-ocean n-alkane and oil pollution.

Methods

Isolation for Alcanivorax xenomutans A28

The surface sediment was collected from the West Philippine Basin (16°57'22” N, 129°44'38“E) at a depth of 7663.5 m by the lander “Fendouzhe” aboard the R/V TAN SUO YI HAO during cruise TS-21 in August 2021. The sediment was sliced into 2 cm intervals and preserved in sample bags (B00992, Whirl-Pak, USA) at 4 °C. The surface sediment (0–2 cm) was first incubated in the ONR7a medium33 supplemented with n-C16 and n-C18 addition (Sigma-Aldrich, USA) under ambient pressure, 4 °C. After two weeks of cultivation, the supernatant was spread on the ONR7a solid plate33 with n-C16 and n-C18 addition. A single colony was picked out and streaked onto a solid ONR7a plate. After repeated streaking, the pure strain was isolated and identified as Alcanivorax xenomutans based genomic analysis. Details on DNA extraction for genome analysis is described in the Supplementary Methods.

Microbial analyses

Culture under different HPs

To harvest sufficient cells for incubation under different HPs, strain A28 was first cultured in ONR7a medium with n-C16 addition under 25 °C, 0.1 MPa. About 5 L cultures with an OD600 of approximately 0.53 (about 5×108 cells per ml) were centrifuged at 3000 rpm for 10 min at 4 °C to harvest cell pellet. The pellet was washed twice by ONR7a medium to remove organic carbon. Cells were then resuspended in 7 L ONR7a medium. The cell suspensions were divided into 50-ml syringes (Kindly, China) without any headspace, in which 50 μl n-C16 (8.20633, Sigma-Aldrich, USA) was supplied as the sole carbon source. The syringes were cultured separately under atmospheric pressure (0.1 MPa) and in high-pressure vessels (HPK03, Shanghai Jiao Tong University, China; Supplementary Fig. 8) with 40 MPa and 80 MPa at 4 °C.

Before pressurization, all cultures in syringes were precultured under 0.1 MPa, 4 °C for 12 h to adapt low temperature conditions. The time at which culture syringes were transferred into the high-pressure vessels was considered T0. Culture syringes were sampled on days 4, 7, 11, and 14. Two blank control groups (cell suspension without n-C16 and sterile ONR7a medium with n-C16) were also set up and tested under 0.1, 40, and 80 MPa. The detailed experimental procedure is illustrated in Fig. 7.

Fig. 7: The schematic diagram of the experimental procedure in this study.
figure 7

The strain A28 was incubated under three HPs, 4 °C. Only the sample for omics analysis were sampled on day 7 and 14. The schematic diagram was created by the authors using Adobe Illustrator.

Full size image

Bacterial growth monitoring

To evaluate the bacterial growth, quantitative real-time PCR for 16S rRNA gene was used to monitor growth. DNA was extracted with a Bacterial Genomic DNA Extraction Kit (DP302, Tiangen, China). The primer pair 341f-519r (341F: 5′-CCTACGGGWGGCWGCA-3′ and 519R: 5′-TTACCGCGGCKGCTG-3′), which is specific to the 16S rRNA gene, was used for quantitative real-time PCR. According to the manufacturer’s instructions for PowerUp™ SYBRTM Green Master Mix (A25742, Thermo Fisher, USA), quantitative real-time PCR was performed on the Real-Time PCR System (QuantStudio 5, Applied Biosystems, USA) with a standard two-step amplification program. The blank control group without n-C16 was sampled for 16S rRNA gene copy number detection only on day 7 and 14. Cells from a 25 mL culture on day 7 and 14 were centrifuged at 10,000 × g for 5 min at 4 °C for protein extraction and quantification (Supplementary Methods).

Intracellular H2O2

Intracellular H2O2 was measured using a Hydrogen Peroxide Assay Kit (S0038, Beyotime, China) to assess the oxidative response of the cells. Approximately 2 ml of culture was centrifuged at 10,000 × g for 3 min at 4 °C to collect the cell pellet. The cell pellet was resuspended a the cell lysis buffer and subjected to ultrasonication and centrifugation. The resulting supernatant was used for intracellular H2O2 measurement with lysis buffer as the diluent in standard sample preparation. Intracellular H2O2 level was normalized to the cell number.

Total antioxidant capacity

Total antioxidant capacity was quantified using a commercial Total Antioxidant Capacity Assay Kit (S0116, Beyotime, China) based on the ferric reducing ability of plasma (FRAP) method. Approximately 10 ml of cultures were centrifuged to harvest cell pellet. The cell pellet was resuspended by PBS and ultrasonic decomposed. After centrifugation at 12,000 × g for 5 min at 4 °C, the supernatant was used for antioxidant capacity measurement. For the FRAP method, the total antioxidant capacity was expressed as the concentration of the FeSO4 standard solution.

Chemical analyses

Amount of CO2 and pH

To evaluate the decomposition of n-C16, the CO2 and pH levels in the medium were measured. As CO2 dissolved in seawater contributes H+ and decreases the pH, thus the pH is also an indicator of n-alkane decomposition17. Approximately 2 ml of culture medium was injected into a 10 ml sealed glass bottle under N2 atmosphere for CO2 measurement. Since there was no headspace in the culture syringes, CO2 was almost dissolved in the liquid phase. The CO2 was quantified using gas chromatography with a BID detector (Nexis GC-2030, Shimadzu, Japan) after adding 10% H3PO4 (695017-100 ml, Sigma-Aldrich, USA). For pH measurement, approximately 2 ml of culture medium was detected by pH meter (FiveEasy PlusTM, Mettler Toledo, Switzerland).

Extraction of n-C16 and analysis by GC-MS

To evaluate the consumption of n-C16, the amount of n-C16 was determined. Due to the low solubility of n-C16 in seawater, three individual culture syringes were used for n-C16 extraction to avoid heterogeneous sampling. Before extraction, 10 μg of n-C20 (219274, Sigma-Aldrich, USA) dissolved in n-hexane was added to culture syringes as an internal standard to assess the recovery rate. Then, 10 ml n-hexane was added into culture syringes and syringes were shaken to mix the organic and liquid phase well. After the standing and layering process, the organic phase was sequentially transferred to a smaller tube and dried under nitrogen. Four milliliters of n-hexane was used to dissolve the residual organic matter and was then analyzed using GC-MS/MS (TSQ 9000, Thermo Fisher, USA). The n-hexadecane solution (A117437, Aladdin, China) was used as the standard to identify the peak and quantify the amount of n-hexadecane.

The mineralization ratio at each sampling point was calculated by dividing the average carbon equivalent of the net CO2 production by the net n-hexadecane consumption, based on the three biological replicates. The average mineralization ratio was calculated by averaging the mineralization ratios from the four sampling points.

Omics analysis

Transcriptome analysis

Three of six biological replicates were used for transcriptome analysis. For each replicate, 25 ml of culture was centrifuged at 10,000 rpm for 5 min at 4 °C. The cell pellet was harvested for RNA analysis. RNA was extracted by the trizol method34 and sequenced by Shanghai Personalbio Biotechnology Co., Ltd. (Shanghai, China). Raw reads were trimmed to remove adapters and then filtered using fastp with parameters (-w 16 -q 20 -u 20 -g -c -W 5 -3 -l 90). The rRNA reads were removed from the clean reads by ribodetecter35. The obtained mRNA reads were mapped to the genome of strain A28 by Burrows-Wheeler-Alignment Tool36. Gene reads count were calculated by featureCounts37. DESeq2 package was used to analyze the differentially expressed genes, with adjusted p-value < 0.05 and the absolute value of log2(fold change) ≥1.

To validate the RNA sequencing, we performed real-time quantitative reverse transcription PCR (qRT-PCR) analysis on six differentially expressed genes, using the 16S rRNA gene as an internal reference (Supplementary Methods). All qRT-PCR primers are listed in Supplementary Table 3 and the results are described in Supplementary Fig. 9.

Intracellular metabolome analysis

Six biological replicates were used for the intracellular metabolite analysis. For each replicate, 25 ml of culture was centrifuged at 10,000 rpm for 5 min at 4 °C. The cell pellet was harvested for intracellular metabolite analysis. The detail method of extraction was described in the Supplementary Methods. LC-MS/MS analyses were performed using an UHPLC system (Vanquish, Thermo Fisher, USA) with a UPLC BEH Amide column (2.1 mm × 100 mm, 1.7 μm) coupled to Orbitrap Exploris 120 mass spectrometer (Orbitrap MS, Thermo Fisher, USA). Raw data were converted to the mzXML format using ProteoWizard and processed with an in-house program, developed using R and based on XCMS, for peak detection, extraction, alignment, and integration. An in-house MS2 database (Biotree Database, Biotree Biotech CO., Ltd., China) was applied in metabolite annotation. The cutoff for annotation was set at 0.3.

Statistics and reproducibility

The statistical analyses of data and sample size are detailed in the relevant subsections of the Methods and Materials and Results and are also included in the corresponding figure legends. GraphPad Prism 10.0.0 and Adobe Illustrator 2023 were used for data analysis and visualization. Statistical significance of differences was determined using two-tailed T tests. One-way analysis of variance (ANOVA) was used for three groups. All statistical analyses were performed using non-parametric tests due to the small sample size of our experimental groups, which may not satisfy the assumptions of normality and homoscedasticity required for parametric methods. All p-values are two-sided, and a p-value  <  0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.