Bioluminescence Contributes to the Adaptation of Deep-Sea Bacterium Photobacterium phosphoreum ANT-2200 to High Hydrostatic Pressure

Bioluminescence is a common phenomenon in nature, especially in the deep ocean. The physiological role of bacterial bioluminescence involves protection against oxidative and UV stresses. Yet, it remains unclear if bioluminescence contributes to deep-sea bacterial adaptation to high hydrostatic pressure (HHP). In this study, we constructed a non-luminescent mutant of ΔluxA and its complementary strain c-ΔluxA of Photobacterium phosphoreum ANT-2200, a deep-sea piezophilic bioluminescent bacterium. The wild-type strain, mutant and complementary strain were compared from aspects of pressure tolerance, intracellular reactive oxygen species (ROS) level and expression of ROS-scavenging enzymes. The results showed that, despite similar growth profiles, HHP induced the accumulation of intracellular ROS and up-regulated the expression of ROS-scavenging enzymes such as dyp, katE and katG, specifically in the non-luminescent mutant. Collectively, our results suggested that bioluminescence functions as the primary antioxidant system in strain ANT-2200, in addition to the well-known ROS-scavenging enzymes. Bioluminescence contributes to bacterial adaptation to the deep-sea environment by coping with oxidative stress generated from HHP. These results further expanded our understanding of the physiological significance of bioluminescence as well as a novel strategy for microbial adaptation to a deep-sea environment.


Introduction
Bioluminescent bacteria are commonly observed in marine habitats, especially deep ocean, where they have been found under either free-living, sessile or symbiotic lifestyles [1]. The majority of luminescent bacteria have been identified in five genera of Vibrio, Photobacterium, Aliivibrio, Photorhabdus and Shewanella. The molecular mechanism of bacterial bioluminescence has been extensively studied in the model strains of Aliivibrio fischeri and Vibrio campbellii [2][3][4][5]. As a highly conserved mechanism, luciferase reduces molecular oxygen in water with concomitant oxidation of long-chain aldehydes to long-chain acids. It is an exergonic reaction and the energy is released through light emission with a wavelength of 450-490 nm

Construction of luxA Mutant and Complementary Strain
To construct a ∆luxA mutant, two sequences of 500 bp in length flanking luxA were synthesized at GeneWiz Inc. (Suzhou, Jiangsu, China) and introduced into pUX19 at restriction sites of XmaI and SacII. The resulting plasmid was introduced into the P. phosphoreum ANT-2200 strain by bi-parental conjugation using E. coli WM3064 as a donor strain. The luxA deletion mutant was screened with a two-steps screening strategy. The first step is to screen for colonies carrying kanamycin resistance cassettes with cultivation on YPG plates containing kanamycin. The second step is to select for double crossover deletion mutant that cannot grow with the presence of 5% sucrose. The potential ∆luxA mutant strain was confirmed by PCR amplification and DNA sequencing.
To construct a complementary strain, the DNA fragment containing luxCDAB was amplified with forward primer: 5 -TAAAACGACGGCCAGTGAGTTACGAGCTTGGTAAATTCTTTTG-3 , and reverse primer: 5 -AGGAAACAGCTATGACATGATAAAGAAAATCCCAATG-3 and introduced into board-host vector pBBR1MCS-2 using pEASY-Uni Seamless Cloning and Assembly Kit (TransGen Biotech, Beijing, China, catalogue number CU101-01). The resulting plasmid was transformed into ∆luxA through bi-parental conjugation, and colonies with kanamycin resistance were then confirmed by sequencing.

Growth Experiment and Luminescence Assays
Cell samples were collected at different stages of growth. To be specific, 1.0 mL culture was collected for absorption measurement at 600 nm (OD 600 nm ) with Cary 60 UV-Vis spectrophotometer (Agilent Technologies, Santa Clara, CA, USA), and 200 µL culture was collected for luminescence measurement which was carried out using a black, flat-bottom 96-well plate with microplate reader Varioskan LUX (Thermo Fisher Scientific, Waltham, MA, USA). Three replicates were taken for each sample. The average luminescent intensity was divided by the average cell density resulting in the specific luminescent density.

Reactive Oxygen Species (ROS) Quantification
The 2 ,7 -dichlorodihydrofluorescein diacetate (DCFH-DA, Sigma-Aldrich, catalogue number D6883) was used to determine the intracellular ROS level. A stock solution (10 mM) was prepared by dissolving 50 mg of DCFH-DA in 10 mL DMSO (Sigma-Aldrich, China, catalogue number D1435). To quantify intracellular ROS, 200 µL culture was taken, and the fluorescent intensity was measured with an excitation of 488 nm and an emission of 526 nm. An average reading (relative fluorescence unit, RFU) was calculated from triplicate measurements. The average reading of the blank medium was subtracted from the average reading of a sample, and then the value was divided by cellular density (OD 600 nm ) and generated the specific fluorescence intensity (RFU/OD 600 nm ).
For the analysis of ROS after short-term incubation, a 15 µM DCFH-DA probe was added into a culture at the early-exponential phase with a cell density of 0.1 OD 600 nm . The mixture was incubated for one hour under atmospheric pressure, at 25 • C, to allow the diffusion of the probe into bacterial cells. The cells were washed and suspended in fresh YPG medium without the probe before being transferred into syringes and incubated for 2.5 h at different pressure conditions.
To monitor the intracellular ROS during the growth experiment, the DCFH-DA probe was added into the culture to a final concentration of 150 µM. The mixture was transferred into syringes and incubated at different pressure conditions. Samples were taken at different growth stages.

RNA Extraction and Real-Time PCR
Total RNA was extracted using Trizol reagent (Ambion, China, catalogue number AM9906) following the manufacturer's instruction. Briefly, 2.0 mL of mid-exponential phase cultures with a cell density of OD 600 nm 0.3 (approximately 10 8 cells per mL) were centrifuged at 4500× g for 30 min. The pellet was homogenized in 1.0 mL of TRIzol, and 200 µL of chloroform was added. After thoroughly shaking and incubation, the samples were centrifuged at 12,000× g for 15 min at 4 • C. The upper aqueous phase was transferred to a tube containing an equal volume of isopropanol. The mixtures were thoroughly mixed and incubated at −20 • C for 30 min before centrifuging at 12,000× g for 20 min at 4 • C. The pellet was washed in 75% ethanol (prepared by mixing anhydrous ethanol and DEPC-treated water at 3:1) and resuspended in 50 µL DEPC-treated water (Ambion, Shanghai, China, catalogue number AM9906). Extracted RNA samples were treated with RNase-free recombinant DNaseI (TaKaRa, Beijing, China, catalogue number 2270A) at 37 • C for 3 h to remove any residual genomic DNA. The cDNA was synthesized using PrimeScript TM II 1st Strand cDNA Synthesis Kit (TaKaRa, China, catalogue number 6210A). Reaction mixtures with total volumes of 20 µL contained FastStart Universal SYBR Green Master (ROX) (Roche, Mannheim Germany, catalogue number 04913914001), 0.6 µL of each primer (10 mM) and 1 µL of 10-times diluted cDNA as template. The thermal cycling protocol was as follows: initial denaturation for 30 s at 95 • C, followed by 40 PCR cycles of denaturation at 95 • C for 15 s, annealing at 55 • C for 10 s, and extension at 72 • C for 30 s, the final extension lasted for 5 min at 72 • C. Primers used for amplification are shown in Table S1. Reactions were carried out using Applied Biosystems StepOnePlus™ Real-Time PCR (Applied Biosystems, Waltham, MA, USA). All the analyses were performed in triplicates. Relative expression levels were calculated using the rpoD gene as the internal reference gene. The relative expression level was calculated as follows: ∆Ct = Ct (gene of interest) − Ct (housekeeping gene), ∆∆Ct = ∆Ct (treated sample) − ∆Ct (untreated sample), Relative expression levels = 2 −(∆∆Ct) .

HHP Treatment and Growth Experiment
Cells cultured to the early-exponential phase with a cell density of approximately 0.1 OD 600 nm were subdivided into two groups and incubated at atmospheric pressure and high hydrostatic pressure (100 MPa) conditions following the method described in Section 2.1. After an incubation of 4 h, cells were collected and diluted 100 times using a fresh medium. The cells were transferred into new syringes and incubated for 30 h in an incubator with light or in the dark by covering the syringe with aluminum foil. Cells were collected at different growth stages and OD 600 nm was measured, as described in Section 2.3.

UV Light Irradiation Assay
Cells grown to the mid-exponential phase were collected and diluted to a cell density of 0.05 OD 600 nm (approximately 10 7 cells per mL). Then 20 mL of bacterial cultures were transferred into a clean petri dish and exposed to a UV lamp (8 W) for 5 min. The cells were incubated under light or dark (covered by aluminum foil) conditions for 2 h. For each sample, 100 µL culture was taken for series dilution and spread on YPG plates. The plates were incubated under light or dark conditions at 25 • C for 48 h before the number of colonies was counted. The survival rate was calculated as follows: Survival rate (%) = CFU of UV-treated cell/CFU of non-treated cells × 100%.

Bioluminescence Favors Bacterial Recovery from High Hydrostatic Pressure
To elucidate the function of bioluminescence in deep-sea bacterium, the luxA gene encoding the catalytic subunit of luciferase in the deep-sea bioluminescent strain P. phosphoreum ANT-2200 was knocked out, generating a non-luminescent mutant ∆luxA. We then introduced a plasmid carrying the luxCDAB fragment into the dark mutant and obtained a complementary strain c-∆luxA which is luminescent ( Figure S1). The expression of lux genes in the three strains was examined by qPCR. As expected, the luxA was undetectable in the mutant, but its expression was doubled in the complementary strain compared to the wild-type strain, while expression of other lux genes was not affected ( Figure S2).
The growth and luminescence of the three strains were examined under pressures of 0.1 MPa, 22 MPa, 30 MPa, 40 MPa and 50 MPa, respectively. The mutant strain was non-bioluminescent by observation. Spectrophotometry analysis confirmed that its bioluminescence intensity was six orders of magnitude lower than the wild-type strain (10 2 RLU in ∆luxA versus 10 8 in the wild-type strain) ( Figure 1). The complementary strain restored luminescence, but its intensity was weaker by approximately an order of magnitude compared to the wild-type strain (10 7 RLU in c-∆luxA versus 10 8 in wild-type strain) ( Figure 1). Changing pressure from 0.1 MPa to 40 MPa had little influence on the specific luminescence intensity of all three strains. Resembling the wild-type strain, both the ∆luxA mutant and the complementary strain c-∆luxA had a higher biomass at 22 MPa than at 0.1 MPa, 30 MPa and 40 MPa. Further increasing the pressure to 50 MPa completely abolished the growth as well as bioluminescence of all three strains.
The growth experiment suggested that disruption of bioluminescence did not affect the piezophilic growth phenotype of the deep-sea strain ANT-2200 under pressures below 50 MPa. We then examined the role of bioluminescence under even higher pressures. Considering the fact that pressure over 50 MPa would be lethal for cells of ANT-2200, we cultivated the cells at 0.1 MPa to the early-exponential phase before transferring them to 100 MPa. After incubation for 4 h at 100 MPa, cells were cultivated again at 0.1 MPa. The growth experiment demonstrated that all three strains were able to grow after a treatment of 100 MPa. It was noted that despite a comparable biomass being reached in all three cultures by the late-exponential phase, the ∆luxA mutant exhibited a prolonged lag phase compared to the luminescent strains ( Figure 2). It seemed that bioluminescence participated in bacterial coping with HHP stress, although the effect appeared only at extremely high pressure under the condition used in this study. Microorganisms 2023, 11, x FOR PEER REVIEW 6 of 15 The growth experiment suggested that disruption of bioluminescence did not affect the piezophilic growth phenotype of the deep-sea strain ANT-2200 under pressures below 50 MPa. We then examined the role of bioluminescence under even higher pressures. Considering the fact that pressure over 50 MPa would be lethal for cells of ANT-2200, we cultivated the cells at 0.1 MPa to the early-exponential phase before transferring them to 100 MPa. After incubation for 4 h at 100 MPa, cells were cultivated again at 0.1 MPa. The growth experiment demonstrated that all three strains were able to grow after a treatment of 100 MPa. It was noted that despite a comparable biomass being reached in all three cultures by the late-exponential phase, the ΔluxA mutant exhibited a prolonged lag phase compared to the luminescent strains ( Figure 2). It seemed that bioluminescence participated in bacterial coping with HHP stress, although the effect appeared only at extremely high pressure under the condition used in this study.

Higher Level of ROS Is Generated by HHP in the Non-Luminescent Mutant
Previous studies in pressure sensitive bacteria and yeast suggested that HHP may lead to oxidative stress [21,27,28]. Meanwhile, molecular oxygen is consumed during bioluminescence, and one potential biological function of microbial bioluminescence is to de-

Higher Level of ROS Is Generated by HHP in the Non-Luminescent Mutant
Previous studies in pressure sensitive bacteria and yeast suggested that HHP may lead to oxidative stress [21,27,28]. Meanwhile, molecular oxygen is consumed during bioluminescence, and one potential biological function of microbial bioluminescence is to detoxify molecular oxygen [7]. It was thus presumed that bioluminescence facilitates deep-sea bacterium coping with oxidative stress generated under the HHP condition. To test this hypothesis, we first analyzed the intracellular ROS levels in cells treated with pressures ( Figure 3). Cells grown to exponential phase at atmospheric pressure were collected and incubated at different pressures for a short period of 2.5 h before the ROS level was measured. The results showed that compared to cells incubated at 0.1 MPa, raising incubation pressure increased intracellular ROS content in all three strains. When incubated at 50 MPa, the ROS level increased by two times in the wild-type strain and by around three times in the ∆luxA mutant strains. Meanwhile, the complementary strain c-∆luxA had the lowest ROS level under all three pressures tested; even incubation at 50 MPa hardly affected its intracellular ROS level.

Higher Level of ROS Is Generated by HHP in the Non-Luminescent Mutant
Previous studies in pressure sensitive bacteria and yeast suggested that HHP may lead to oxidative stress [21,27,28]. Meanwhile, molecular oxygen is consumed during bioluminescence, and one potential biological function of microbial bioluminescence is to detoxify molecular oxygen [7]. It was thus presumed that bioluminescence facilitates deepsea bacterium coping with oxidative stress generated under the HHP condition. To test this hypothesis, we first analyzed the intracellular ROS levels in cells treated with pressures ( Figure 3). Cells grown to exponential phase at atmospheric pressure were collected and incubated at different pressures for a short period of 2.5 h before the ROS level was measured. The results showed that compared to cells incubated at 0.1 MPa, raising incubation pressure increased intracellular ROS content in all three strains. When incubated at 50 MPa, the ROS level increased by two times in the wild-type strain and by around three times in the ΔluxA mutant strains. Meanwhile, the complementary strain c-ΔluxA had the lowest ROS level under all three pressures tested; even incubation at 50 MPa hardly affected its intracellular ROS level. Asterisks indicate significant differences: *** p < 0.001, * p < 0.05 (data were analyzed by one-way ANOVA and the differences among the means were tested using Tukey's multiple comparison test). Asterisks indicate significant differences: *** p < 0.001, * p < 0.05 (data were analyzed by one-way ANOVA and the differences among the means were tested using Tukey's multiple comparison test).

Alternative ROS-Scavenging Enzymes Are Induced in the Non-Luminescent Mutant in Response to HHP
Elevated pressure resulted in the increase of intracellular ROS content, especially in the non-luminescent cells, indicating that HHP triggers oxidative stress and the level of oxidative stress is dependent on bioluminescence. To corroborate this hypothesis, we examined the expression of ROS-scavenging enzymes in the three strains. A total of seven genes were analyzed, including three genes coding for putative superoxide dismutase (sodB, sod1 and sod2), katE and katG coding for putative catalases, dyp coding for Dyp-type peroxidase and prx coding for peroxiredoxin C ( Table 2).
Compared to cultures at 0.1 MPa, short incubation at 22 MPa up-regulated the expression of sodB and dyp by over 6-fold, prx gene by 4-fold and doubled the expression of the other two superoxide dismutases (sod1 and sod2) in the ∆luxA. In contrast, none of these genes was affected in the wild-type strain and the complementary strain (Figure 4a). In cells incubated at 50 MPa, the expression of sodB gene was remarkably increased by over 40-to 80-fold in all three strains. In addition, the expression of the two catalases (katE and katG) and the dyp genes were specifically induced in the mutant, while the dyp alone was slightly induced in the complementary strain (Figure 4b). To our surprise, the expression of the prx gene, which was clearly induced at 22 MPa in the ∆luxA, was not affected at 50 MPa.

Different ROS Elimination Rates in the Luminescent and Non-Luminescent Cells
To further understand how bioluminescence participated in bacterial coping with HHP and oxidative stress, we monitored the changes in ROS levels in the three strains cultured under pressures of 0.  We further examined the expression of ROS-scavenging enzymes in response to hydrogen peroxide, one of the common sources of oxidative stress. As shown in Figure 4c, sodB, dyp and prx were induced by H 2 O 2 in the mutant of ∆luxA, which was similar to the influence of pressure of 22 MPa, but to a greater level. In addition to that, the two genes coding for catalases were up-regulated by around 4-and 8-fold as well. On the other hand, Microorganisms 2023, 11, 1362 9 of 14 little influence was observed in the wild-type and the complementary strain, except for sodB which was induced by 3-to 4-fold.

Different ROS Elimination Rates in the Luminescent and Non-Luminescent Cells
To further understand how bioluminescence participated in bacterial coping with HHP and oxidative stress, we monitored the changes in ROS levels in the three strains cultured under pressures of 0.1 MPa and 22 MPa, or in the presence of H 2 O 2 at 0.1 MPa. As presented in Figure 1, all three strains had similar growth profiles at 0.1 MPa and 22 MPa, respectively (Figure 1a,b). In contrast, when grown with the presence of H 2 O 2 , different growth curves were observed. The c-∆luxA strain entered the exponential phase first, followed by the wild-type strain and the ∆luxA mutant grew most slowly. It was noted that although the duration of lag phase varied from different strains, same biomass was achieved by the stationary phase (Figure 1f).
The changes of specific ROS levels during growth were examined by sequential sampling. Three samples were collected from each culture for the quantification of the intracellular ROS level: at the early-exponential phase (named as phase I for short) when OD 600 nm reached approximately 0.1; the mid-exponential phase (named as phase II) with OD 600 nm of approximately 0.3; and the late-exponential phase (named as phase III) with OD 600 nm of approximately 0.5 (sampling points are indicated in Figure 1).
When cultivated at atmospheric pressure, all three strains had comparable ROS levels ( Figure 5). The ROS content decreased gradually from around 3 at phase I to around 2 at phase II and remained stable afterward. The application of 22 MPa tripled the intracellular ROS level in the ∆luxA mutant and doubled the content in the wild-type and the complementary strain at phase I (22 MPa versus 0.1 MPa). Within 2 h from phase I to phase II, the value decreased from 7.0 to 5.2 in the wild-type strain, from 7.6 to 5.1 in the complementary strain and from 9.5 to 6.5 in the non-luminescent mutant. By phase III, the ROS levels of the three strains were within the range of 4.5 to 4.8, and no significant difference was observed.
Addition of H 2 O 2 increased the ROS level by 4.3-and 17.5-fold in the wild-type and the ∆luxA mutant, respectively, but had little influence on the complementary strain (0.1 MPa + H 2 O 2 versus 0.1 MPa). The wild-type strain had lower ROS level at phase II than at phase I (5.3 versus 13.4) and stayed stable afterward. Meanwhile, the value in the non-luminescent mutant decreased rapidly from 61.6 to 9.6 during 5 h between phase I and phase II, and kept decreasing during phase II to phase III but at a relatively lower speed. By the stationary phase, the difference between the mutant and the wild-type strain was insignificant.
Taken together, the quantification of ROS at different growth stages confirmed that a pressure of 22 MPa or the presence of hydrogen peroxide induces the accumulation of intracellular ROS, and the effect was more significant in the non-luminescent mutant than in the luminescent cells. The elevated ROS generated from pressure of 22 MPa or H 2 O 2 could be eliminated during growth, and the ∆luxA mutant exhibited a higher ROS decreasing rate, which could be explained by the induced transcription level of ROS-scavenging enzymes. roorganisms 2023, 11, x FOR PEER REVIEW 11 of 15 Figure 5. The intracellular ROS level in the wild-type strain, ΔluxA and c-ΔluxA at different growth stages. Intracellular ROS level was represented by RFU/OD600nm. Panels (a-c), the intracellular ROS level of the three strains at early-exponential phase, mid-exponential phase and late-exponential phase, respectively. Blue, black and red bars and lines represent wild-type strain, ΔluxA mutant and c-ΔluxA complementary strain, respectively. The sampling points were indicated by arrows in Figure 1. Asterisks indicate significant differences: *** p < 0.001, ** p < 0.01, * p < 0.05 (data were analyzed by one-way ANOVA and the differences among the means were tested using Tukey's multiple comparison test). Figure 5. The intracellular ROS level in the wild-type strain, ∆luxA and c-∆luxA at different growth stages. Intracellular ROS level was represented by RFU/OD 600nm . Panels (a-c), the intracellular ROS level of the three strains at early-exponential phase, mid-exponential phase and late-exponential phase, respectively. Blue, black and red bars and lines represent wild-type strain, ∆luxA mutant and c-∆luxA complementary strain, respectively. The sampling points were indicated by arrows in Figure 1. Asterisks indicate significant differences: *** p < 0.001, ** p < 0.01, * p < 0.05 (data were analyzed by one-way ANOVA and the differences among the means were tested using Tukey's multiple comparison test).

Discussion
Previous studies in the deep-sea bacterium S. piezotolerans WP3 suggested that adaptation to HHP and oxidative stress are correlated [22]. In this study, by direct quantification of intracellular ROS using fluorescent probe DCFH-DA, we confirmed that incubation under elevated pressure led to oxidative stress in the deep-sea bacterium P. phosphoreum ANT-2200.
Moreover, compared to the luminescent strains, the induction of ROS by both HHP and hydrogen peroxide was more pronounced in the non-luminescent mutant, especially at the early exponential growth phase.
The qRT-PCR analysis revealed that the two luminescent strains had similar and relatively lower expression levels of antioxidative enzymes, and several kinds of ROSscavenging enzymes were specifically induced by HHP in the dark mutant. At the optimum pressure of 22 MPa, expression of sodB and dyp increased most remarkably in the ∆luxA mutant. When pressure increased to 50 MPa, the induction of sodB was observed in all three strains, while two catalases, a Dyp-type peroxidase and a peroxiredoxin were specifically up-regulated in the dark mutant. The different expression profile of antioxidative enzymes agrees with the changes in ROS content in that, while all three strains were capable of eliminating ROS gradually, the non-luminescent mutant with up-regulated ROS-scavenging enzymes exhibited a higher ROS elimination rate.
Superoxide dismutase, catalase and peroxidase are among the best-characterized antioxidant enzymes. During bacterial oxidative stress defense, superoxide dismutases catalyze the conversion of O 2 − into H 2 O 2 , which are then reduced by catalases and peroxidases into water and oxygen [29]. Peroxiredoxins have similar function as peroxidase, and in addition, participate in H 2 O 2 sensing and signaling, and maintaining peroxide level [30]. The participation of superoxide dismutase and catalase in bacterial coping with HHP stress has been reported in E. coli, Enterobacter sakazakii, yeast and deep-sea bacterium S. piezotolerans WP3 [22,[31][32][33]. Yet, it is the first time that the involvement of peroxidase and peroxiredoxin in HHP adaptation was demonstrated. It should be noted that ROS consists of several kinds of free radicals, such as the superoxide anion (O 2 − ), singlet oxygen ( 1 O 2 ), hydroxyl radicals (·OH) and non-radical hydrogen peroxide (H 2 O 2 ), etc. The DCFH-DA detects generalized free radicals instead of certain particular ROS [34]. It would be of great interest to identify the nature of ROS molecules generated by HHP and clarify reactions catalyzed by each ROS-scavenging enzyme during bacterial response to HHP. In addition, three putative superoxide dismutases were identified in the genome of ANT-2200; however, only one of them is up-regulated by elevated pressure, indicating different functions among the three iso-enzymes in coping with HHP-induced ROS.
We noticed that despite a relatively lower luminescent intensity, the complementary strain was more tolerant to HHP and oxidative stress than the wild-type strain. The quantification of lux genes showed that most of them had comparable expression levels in all three strains, and the only difference lies in the expression of luxA, which was doubled in the complementary strain compared to the wild-type strain ( Figure S2). Therefore, one possible explanation for the improved tolerance to oxidative stress is that LuxA binds to FMN during a bioluminescence reaction, whose overexpression might alter the intracellular redox homeostasis and enhance the antioxidant capacity through a yet-to-be-discovered mechanism.
Taken together, it seems that bioluminescence participates in the adaptation to HHP by functioning as a primary antioxidant system ( Figure 6). As deduced from the slightly increased ROS level and constitutive expression of ROS-scavenging enzymes in the wildtype strain grown under 22 MPa, a functional bioluminescence process would be sufficient to eliminate ROS generated under this condition. Further increased pressure would lead to a higher ROS level and ROS-scavenging enzymes, predominantly SodB, would be induced to cope with the excess ROS molecules. An impaired bioluminescence system induces the expression of several more ROS-scavenging enzymes, such as Dyp-type peroxidase and catalases, indicating their role as an alternative antioxidant system to maintain the homeostasis of intracellular ROS level.
Another possible physiologic function of bioluminescence is to activate DNA photolyase, which stimulates DNA repair by removing the DNA lesions formed by UV such as cyclobutane pyrimidine dimers [9][10][11]35,36]. It was once suggested that light-activated photolyase would be unnecessary in deep-sea microorganisms since sunlight is absent from the deep-sea environment [37]. However, the gene coding for photolyase was identified in the genome of ANT-2200 (PPBDW_v2_II0175), suggesting the capacity of photoreactivation in this deep-sea bacterium. We compared the survival rate of two luminescent (P. leiognathi W9 and W214) and a non-luminescent (P. angustum QY26) deep-sea Photobacterium strains after UV irradiation ( Figure S3). The results showed that similar survival rates were obtained for luminescent strains (ANT-2200, W9 and W214), regardless of incubation in the dark or in the light. However, the non-luminescent strain QY26 and the ∆luxA mutant of ANT-2200 had a significantly lower survival rate when incubated in the absence of light. We further demonstrated that self-emitted and external light had a similar effect on bacterial recovery from pressurization ( Figure S4). The growth of non-luminescent cells was slower than the luminescent ones after treatment at 100 MPa. However, if the cells were grown with exposure to light, same growth profiles were observed in all three strains. This result indicates that, as in shallow water bioluminescent bacteria, photons emitted by luciferase could activate photolyase activity and repair the UV damage in the deep-sea strain ANT-2200. But it remains uncertain if HHP causes similar DNA damage as the UV irradiation does and if the light itself facilitated bacterial recovery from HHP by stimulating photoreactivation.
Microorganisms 2023, 11, x FOR PEER REVIEW 13 of Figure 6. Schematic diagram shows the antioxidant systems in deep-sea bioluminescent bacteriu P. phosphoreum ANT-2200. Bioluminescence functions as the primary antioxidant system to elim nate ROS generated from HHP. When excess ROS was present, either generated from extreme high pressure or due to impaired bioluminescence, ROS-scavenging enzymes were induced to co trol the level of intracellular ROS.
Another possible physiologic function of bioluminescence is to activate DNA photo yase, which stimulates DNA repair by removing the DNA lesions formed by UV such cyclobutane pyrimidine dimers [9][10][11]35,36]. It was once suggested that light-activate photolyase would be unnecessary in deep-sea microorganisms since sunlight is abse from the deep-sea environment [37]. However, the gene coding for photolyase was ide tified in the genome of ANT-2200 (PPBDW_v2_II0175), suggesting the capacity of phot reactivation in this deep-sea bacterium. We compared the survival rate of two luminesce (P. leiognathi W9 and W214) and a non-luminescent (P. angustum QY26) deep-sea Phot bacterium strains after UV irradiation ( Figure S3). The results showed that similar surviv rates were obtained for luminescent strains (ANT-2200, W9 and W214), regardless of i cubation in the dark or in the light. However, the non-luminescent strain QY26 and th ΔluxA mutant of ANT-2200 had a significantly lower survival rate when incubated in th absence of light. We further demonstrated that self-emitted and external light had a sim lar effect on bacterial recovery from pressurization ( Figure S4). The growth of non-lum nescent cells was slower than the luminescent ones after treatment at 100 MPa. Howeve if the cells were grown with exposure to light, same growth profiles were observed in a three strains. This result indicates that, as in shallow water bioluminescent bacteria, ph tons emitted by luciferase could activate photolyase activity and repair the UV damage the deep-sea strain ANT-2200. But it remains uncertain if HHP causes similar DNA dam age as the UV irradiation does and if the light itself facilitated bacterial recovery fro HHP by stimulating photoreactivation.
In summary, we showed in this study that elevated pressure induced oxidative stre in a deep-sea bacterium. Bioluminescent cells could better maintain the ROS level und HHP, while cells with impaired bioluminescence require additional ROS-scavenging e zymes such as catalase and peroxidase to cope with the oxidative stress generated fro HHP. Therefore, bioluminescence in the deep-sea bacterium P. phosphoreum ANT-220 takes part in the adaptation to the deep-sea environment as an antioxidant system. The Figure 6. Schematic diagram shows the antioxidant systems in deep-sea bioluminescent bacterium P. phosphoreum ANT-2200. Bioluminescence functions as the primary antioxidant system to eliminate ROS generated from HHP. When excess ROS was present, either generated from extremely high pressure or due to impaired bioluminescence, ROS-scavenging enzymes were induced to control the level of intracellular ROS.
In summary, we showed in this study that elevated pressure induced oxidative stress in a deep-sea bacterium. Bioluminescent cells could better maintain the ROS level under HHP, while cells with impaired bioluminescence require additional ROS-scavenging enzymes such as catalase and peroxidase to cope with the oxidative stress generated from HHP. Therefore, bioluminescence in the deep-sea bacterium P. phosphoreum ANT-2200 takes part in the adaptation to the deep-sea environment as an antioxidant system. These observations expanded our understanding of an alternative strategy in deep-sea bacteria to adapt to the elevated pressure in a deep-sea environment.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/microorganisms11061362/s1, Table S1: Primers used for RT-qPCR analyses in this study; Figure S1: The luminescence of colonies of ANT-2200 wild-type, ∆luxA mutant and the complementary strain c-∆luxA; Figure S2: The transcription level of lux genes in different strains; Figure S3: The influence of UV irradiation on different Photobacterium strains. Figure S4: Growth recovery of wild-type strain, ∆luxA and c-∆luxA with exposure to light after HHP treatment.