Genetic redundancy in the naphthalene-degradation pathway of Cycloclasticus pugetii strain PS-1 enables response to varying substrate concentrations

Abstract Polycyclic aromatic hydrocarbon (PAH) contamination in marine environments range from low-diffusive inputs to high loads. The influence of PAH concentration on the expression of functional genes [e.g. those encoding ring-hydroxylating dioxygenases (RHDs)] has been overlooked in PAH biodegradation studies. However, understanding marker-gene expression under different PAH loads can help to monitor and predict bioremediation efficiency. Here, we followed the expression (via RNA sequencing) of Cycloclasticus pugetii strain PS-1 in cell suspension experiments under different naphthalene (100 and 30 mg L−1) concentrations. We identified genes encoding previously uncharacterized RHD subunits, termed rhdPS1α and rhdPS1β, that were highly transcribed in response to naphthalene-degradation activity. Additionally, we identified six RHD subunit-encoding genes that responded to naphthalene exposure. By contrast, four RHD subunit genes were PAH-independently expressed and three other RHD subunit genes responded to naphthalene starvation. Cycloclasticus spp. could, therefore, use genetic redundancy in key PAH-degradation genes to react to varying PAH loads. This genetic redundancy may restrict the monitoring of environmental hydrocarbon-degradation activity using single-gene expression. For Cycloclasticus pugetii strain PS-1, however, the newly identified rhdPS1α and rhdPS1β genes might be potential target genes to monitor its environmental naphthalene-degradation activity.


Introduction
Pol ycyclic ar omatic hydr ocarbons (PAHs) ar e classified as substances of concern (Envir onmental Pr otection Agency 1993 ) and are ubiquitous in marine environments where they bioaccum ulate and ar e toxic to sea life as well as humans (Landrum et al. 2003, Nikolaou et al. 2009, Murawski et al. 2014, González-Gaya et al. 2016, Stading et al. 2021, Zhang et al. 2021 ).The biodegradation of these chemicals, principally by microorganisms, is a crucial process in their oxidation, whic h ultimatel y mitigates their toxicity effects (National Research Council 2003, Genov ese et al. 2014, Dur an and Cr avo-Laur eau 2016, Ov erholt et al. 2016, González-Gaya et al. 2019 ).Micr oor ganisms ar e involv ed in PAH biodegradation, either by performing one step of the degradation pathway as part of a community or by using the complete pathway for the full degradation of one or more of these types of c hemicals (Dombr o wski et al. 2016, Jo y e et al. 2016, Gutierrez 2019a, Mahjoubi et al. 2021 ).Cycloclasticus spp.are known k e y PAH degr aders commonl y found in contaminated marine habitats and can completely oxidize PAHs like na phthalene, phenanthr ene and pyrene (Kasai et al. 2002, Cui et al. 2014, Wang et al. 2018, Bagi et al. 2022 ).Naphthalene, as the PAH with the highest water solubility, is often used as a model compound in PAH biodegradation studies, and its biodegradation pathway in Cycloclasticus spp.has been well described ( Fig. S1 ) (Wang et al. 2018(Wang et al. , 2021 ) ).
To date, multiple genes involved in naphthalene degradation ha ve been disco vered and more candidate genes likely exist in the genome of the model organism Cycloclasticus pugetii strain PS-1.Ho w e v er, it is important to consider that ther e ar e still knowledge gaps in our understanding of the naphthalene-and, mor e gener all y, PAH-degr adation pathways occurring in marine habitats .For example , Sieradzki et al. ( 2021 ) could not detect a complete na phthalene-degr adation pathway in whole comm unity metagenomes from PAH-contaminated surface water samples.Additionall y, ther e is a paucity of insight into the transcriptional behavior and the factors influencing the expression of functional genes.A r ecent pa per using metatr anscriptomics r e v ealed species-specific responses of two k e y hydr ocarbon degr aders, Marinobacter and Colwellia , to distinct exposure regimes resulting from additions of organic carbon derived from oil, synthetic dispersant, or oil and synthetic dispersant (Pena-Montenegro et al. 2023 ).In the envir onment, PAH emissions fr om anthr opogenic sources range from low diffusive inputs (e.g. through transportation and river runoff) to high amounts (e.g. through shipping, oil pipelines and platform/rig accidents) (National Research Council 2003, Dur an and Cr avo-Laur eau 2016, Ryther et al. 2021 ).Considering the wide range of emitted PAH loads, one factor overlooked so far in PAH biodegradation studies is the influence of environmental PAH concentrations in inducing gene expression for each of the biodegradation pathway steps.
One important application of investigating the PAHdegradation pathway is to predict and monitor a microbial ecosystem's response to PAH contamination.This response is dependent on the set of functional genes in the microbial community and the conditions that result in their expression.The genes encoding k e y enzymes in the PAH-degradation pathway could impact degr adation differ entl y, depending on their transcriptional behavior.Given the potential correlation between tr anscription and PAH-degr adation acti vity, functional mark er genes, whic h ar e tr anscribed dependent on av ailable substr ate, could help to de v elop a quantitative PCR (qPCR)-based tool (i.e.assays targeting the transcript-to-gene ratio) serving as a measur e for envir onmental PAH-degr adation activity (Wilson et al. 1999, Baelum et al. 2008, Brow et al. 2013, Tentori and Richardson 2020, Vogel et al. 2023b ).Ho w e v er, it is unknown if there is na phthalene-concentr ation-de pendent mark er-gene expression for functional genes in Cycloclasticus pugetii strain PS-1.
Genes that are well conserved in PAH-degrading organisms (Meynet et al. 2015, Liang et al. 2019 ) can be used as marker genes to identify k e y PAH degr aders in envir onmental comm unities through DNA-based analyses (Genovese et al. 2014, Bagi et al. 2022 ), e v en if they are transcribed independently of PAH availability.Because the expression of such substrate-independentlyexpressed genes does not reflect the PAH-degradation activity, tr anscriptomic r esults m ust be inter pr eted with caution.In a pr e vious study, we found that Cycloclasticus pugetii strain PS-1 expr essed thr ee functional genes involv ed in PAH degr adation independent of substrate availability (Vogel et al. 2023b ); howe v er, the tr anscriptional pattern of other genes of Cycloclasticus pugetii strain PS-1 is lacking.Furthermore, whether substrateindependent expression of functional PAH-degradation genes is a common strategy in Cycloclasticus pugetii strain PS-1 remains unconstrained.
Upon further examination of the genomes of Cycloclasticus spp., multiple genes encoding enzymes that are potentially capable of performing the same reaction in the naphthalene-degradation pathway exist, hinting at genetic redundancy (Wang et al. 2018, Bagi et al. 2022 ).Multiple layers of functional redundancy exist in micr oor ganisms (Ghosh and O'Connor 2017 ), and while functional r edundancy seemingl y counters selectiv e pr essur e (No w ak et al. 1997 ), it could lead to flexibility and thus be a benefit for the organism (Laruson et al. 2020 ).Ho w ever, it is unknown if there is genetic redundancy in the naphthalene-degradation pathway, if there is one preferred gene or a set of genes for each step of the na phthalene-degr adation pathway, and which conditions select these.In this study, we sought to determine if Cycloclasticus pugetii strain PS-1 possesses functional marker genes that are transcribed during naphthalene active degradation, or if substrateindependent expression of PAH-degradation genes is a common strategy in this organism.We also examined whether genetic redundancy for genes involved in the naphthalene-degradation pathway occurs in this organism.

Cycloclasticus pugetii strain PS-1 cell suspension experiments
A fr eeze-dried cultur e of the well-studied PAH-degr ading marine model organism Cycloclasticus pugetii strain PS-1 [American Type Cultur e Collection (ATCC) 51542], originall y isolated fr om deepsea sediments of the Pacific Ocean in Puget Sound (Dyksterhouse et al. 1995 ), was acquired from the ATCC (Virginia, USA).Strain PS-1 was r e viv ed according to the manufacturer's instructions and maintained in liquid culture with Marine-Bouillon 2216 (Sigma-Aldrich, USA), supplemented with 100 mg L −1 naphthalene .T he cultur es wer e confirmed as Cycloclasticus pugetii strain PS-1 with Sanger sequencing of the 16S rRNA gene, prior to conducting the RNA-sequencing experiment.
The experiments were conducted using a cell suspension in late log phase growth with a high cell density rather than a growing culture to eliminate growth as a parameter and thereby enable comparing the transcriptional response of Cycloclasticus with differ ent PAH concentr ations.To pr epar e inocula of cell suspensions, 700 ml of day 4 pr e-cultur e was pooled by centrifugation (5000 xg; 10 min), and then the cell pellet was washed twice with fresh medium prior to resuspending (b y v ortexing) to 70 ml, resulting in a highl y concentr ated suspension of str ain PS-1 cells.To set up the cell suspension experiment, 800 μL of the highly concentrated Cycloclasticus pugetii strain PS-1 inoculum was added to individual 20ml serum vials containing a total of 8 ml of carbon-and nutrientrich artificial seawater medium (Difco 2216, Sigma-Aldrich, USA), resulting in high cell densities of the order of 10 8 cells mL −1 ( Fig. S2 ).Naphthalene was provided at two loads, 30 and 100 mg L −1 , whic h wer e selected to r epr esent two concentr ations fr om near and above the solubility of this chemical in seawater [28.96 mg L −1 (Vogel et al. 2023a )].The first concentration, in low-N AP and pulse-N AP tr eatments, is near na phthalene's solubility in seawater and mimics diffusive low-concentration PAH input into marine en vironments .T he added amount was expected to dissolve in the culture medium, so contaminant consumption would lead to a continuous decrease of naphthalene concentrations up to its depletion.In the pulse-NAP treatments, a second low-concentrated pulse of 30 mg L −1 was added at 71 h after complete degradation of the initial naphthalene.In high-NAP conditions, the addition of naphthalene above its water solubility was expected to act as a substrate reservoir.As naphthalene was consumed, more would dissolve from crystals, thus maintaining exposure of cells to relatively constant naphthalene concentrations (similar to a steady state set-up) (Vogel et al. 2023a ).The reserv oir w as depleted once the total na phthalene concentr ation decreased below the water solubility limit.We considered that this experimental condition mimicked massive oil spills, which will infuse large amounts of hydrocarbons into seawater, often at concentr ations abov e water solubility.Ther efor e, 53 μL, 16 μL, or two times 16 μL of a highly concentrated naphthalene stock solution (15 227.87 mg L −1 dissolved in acetone) were added to final concentrations of 100 mg L −1 (high-NAP) and 30 mg L −1 (lo w-N AP), or as two pulses of 30 mg L −1 (pulse-NAP), r espectiv el y.The acetone e v a por ated immediatel y, leaving the na phthalene concentration near its solubility in seawater (28.96 mg L −1 ) and visible as undissolved crystals (Vogel et al. 2023a ).The same steps were undertak en to pre pare uninoculated vials with naphthalene as abiotic contr ols.Na phthalene concentr ations in the inoculated and uninoculated vials were quantified (described below) over the course of the experiment.
For pr epar ation of a PAH-fr ee contr ol, 200 μL of 0.1 M pyruv ate was added to individual 20-ml serum vials that contained 8 ml of medium.This addition was comparable with the molar amount of carbon (0.001 mol L −1 ) added in the high-NAP set-up.All biotic vials were set up as three sacrificial samples per timepoint for naphthalene quantification, and three additional sacrificial bottles for DN A and RN A extr action, whic h wer e all incubated in the dark on a rotary shaker (125 r/m; 18 • C).Sacrificial bottles were used to avoid mass losses of naphthalene due to volatilization.A liquid-liquid extraction of each bottle, with the strong solvent cyclohexane, was used to full y extr act the total naphthalene from both the sorbed phase and the aqueous phase.Naphthalene was quantified after inoculation, as well as after 12, 24, 48 and 168 h for high-concentr ation tr eatments, and once in the PAH-fr ee contr ols after 24 h.PAHs were quantified in low-concentration treatments after inoculation and after 12, 24 and 71 h.Pulsed treatments had identical conditions to lo w-N AP treatments betw een 0 and 71 h and, ther efor e, wer e onl y sampled after 73, 85, 97 and 168 h.Samples for RNA-sequencing were taken 2 h after inoculation for all na phthalene-containing tr eatments, including following the second pulse of naphthalene in the pulse-NAP set-ups, and at a later n utrient-de pleted ("starvation") time point when naphthalene was full y degr aded.

Quantification of hydrocarbons
Na phthalene concentr ations wer e quantified by gas c hr omatogr aphy coupled to mass spectrometry (GC-MS).A liquid-liquid extraction with cyclohexane of each complete sacrificial bottle was performed to avoid mass losses of naphthalene due to volatilization.As an internal standard, 12.6 mg L −1 of D 8 -naphthalene (dissolved in acetone) was added into the crimped vials before using 10 ml of cyclohexane as an or ganic solv ent to extr act na phthalene.Subsequently, the samples were shaken for 35 min at 270 r/m and the organic and aqueous phases were allo w ed to separate undisturbed for 48 h in the dark.The organic phase was subsequently extracted using a glass syringe, and diluted 1:100 with cyclohexane, befor e na phthalene was quantified using an Agilent 6890 N GC coupled to an Agilent 7973 inert MS.The GC-MS was equipped with an Agilent 7683 B autosampler with a J + W Scientific DB-5MS capillary column (30-m length; 0.025-mm ID; 0.25-μm film thickness) and was operated in single ion mode with splitless injection and a helium flow rate of 0.8 ml min −1 .

DNA and RNA extraction and processing
For quantification of the transcription of PAH-related genes during active naphthalene degradation (2 h after addition of naphthalene) and under starving naphthalene conditions, RNA sequencing was conducted.The timepoints for sampling were chosen based on pr e vious experiments (Vogel et al. 2023b ).Hence, for DN A and RN A analyses, the total v olume (8 ml) of each sacrificial bottle was filtered through sterile 0.22-μm Sterivex filters (Merc k Millipor e, Darmstadt, German y) on ice and stor ed at -80 • C until further processing.DNA and RNA were extracted using the Allprep mRN A/DN A kit (Quiagen, Hilden, Germany) according to the manufacturer's instructions with the exception that the extraction buffer was added directly to the Sterivex cartridges, and these were then vortexed at medium po w er for 4 min before removing the buffer with a 10-ml syringe and transferring it to the first spin column.Immediately upon completing these extraction steps, DN A w as stored at -20 • C and mRNA at -80 • C until further analysis.For RNA purification, the TURBO DNA-free kit (Thermo Fisher Scientific Inc., USA) was used to digest any remaining DNA, following the manufacturer's instructions .T he r esulting DNA-fr ee RN A w as submitted to the Institute for Medical Microbiology and Hygiene (University of Tübingen) for library preparation using Illumina str anded RNA pr e p, rRNA de pletion with Ribo-zero Plus, and sequencing using NextSeq 500 High Output Kit v2.5 (75 cycles, Illumina, San Diego, CA, USA).

Monitoring cell numbers-qPCR of functional genes rhd3 α and rhd2 α
To ensure the cell density remained constant over the course of the cell suspension experiment, w e follo w ed tw o functional genes-rhd3 α and rhd2 α-using qPCR.Both genes encode alpha subunits of aromatic ring-hydroxylating dioxygenases (RHD2 α and RHD3 α) and are present only once per genome in Cycloclasticus pugetii strain PS-1 (Vogel et al. 2023b ).Both qPCR methods and primers were already developed and used elsewhere: rhd3 α by Dionisi et al. ( 2011 ) (note that the gene is r eferr ed to as phnA1 in this study) and rhd2 α in our pr e vious study (Vogel et al. 2023b ).Primer and qPCR protocol information can be found in Vogel et al. ( 2023b ) (see also Tables S1 and S2 ).
To compare the differences in total gene expression between the different set-ups, multidimensional scaling (MDS) was conducted.MDS based on expression profile distances of the top 500 log2-fold changes between sample pairs with edgeR v3.26.5 was plotted for all treatments (Robinson et al. 2010 ).To assess the differences in the transcription levels of each gene, gene counts were used in differential abundance analysis for all treatments in R v4.1.1 (2021-08-10) with DESeq2 v1.34.0 (Love et al. 2014 ) using singularity container https://depot.galaxyproject.org/ singularity/ bioconductor-deseq2:1.34.0--r41h399db7b _ 0 .Significant differences were postulated for transcripts using the Benjamini and Hoc hber g-adjusted P -v alue ( P adj ) ≤ 0.05.Finall y, gene counts were transformed to transcript per million (TPM) to allow for the comparison of gene expression between treatments with StringTie2 v2.1.7 (K o vaka et al. 2019 ).
To identify k e y metabolic functions in the genome of strain PS-1, functional hidden Markov model profile-based KEGG orthology (KO) annotation and KEGG mapping (Kanehisa andGoto 2000 , Kanehisa et al. 2016a ,b ) was conducted using KofamKOALA v2022-06-02 (Kanehisa et al. 2016a,b , Aramaki et al. 2020 ) with a threshold E-value = 0.01, with release 102.0 ( https://www.genome.jp/tools/k ofamk oala/).All genes wer e gr ouped by functional categories using the KEGG database "modules" (le v el 3) while omitting any unannotated genes, and cumulative mean TPM of the biological triplicates were plotted per functional category.

Identification of the genes involved in PAH degr ada tion and defining transcription categories
To find candidate genes for all the reactions involved in the na phthalene-degr adation pathw ay, w e compiled a database containing 154 Cycloclasticus pugetii strain PS-1 genes related to PAH degradation ( Table S3, sheet A ) using the annotations of the NCBI database (August 2022) (Sc hoc h et al. 2020 ), the KEGG database (August 2022) (Kanehisa andGoto 2000 , Kanehisa et al. 2016a ,b ) and published liter atur e (Wang et al. 1996, Kasai et al. 2003, Wang et al. 2018, Liang et al. 2019, Wang et al. 2021, Bagi et al. 2022 ).
A sub-set of PAH-related genes was generated from the curated 154 PAH-gene liter atur e-database (described abov e; Table S3, sheet A ) using R (v3.6.0 (2019-04-26) and Rstudio 2022.07.2 + 576).Genes of inter est wer e selected based on high expression under naphthalene-containing conditions (mean TPM in at least one of the na phthalene-containing tr eatments within the 90th percentile, mean TPM ≥ 451).The resulting 43 genes ( Table S3, sheet B ) were assigned to the categories defined below, sorted by pathway step, and the TPMs, as a measure of expression, were plotted in a heatma p. Furthermor e, the significance between the naphthalene-containing conditions and the no-PAH control was highlighted (genes with -1 < log2-fold change < 1 and P adj ≤ 0.05).
To identify patterns in gene expression pertaining to naphthalene availability, four categories based on the genes' transcription in the presence or absence of naphthalene (NAP pos , N AP neg , N AP indep , no pattern) were defined ( Table S4 ), as follows: (i) significant upregulation under naphthalene-containing conditions and/or downregulation under naphthalene-starvation conditions compared with the no-PAH control (NAP pos ); (ii) significant downr egulation in na phthalene-containing conditions and/or upregulation under naphthalene-starvation conditions compared with the no-PAH control (NAP neg ); (iii) genes that sho w ed no significant upregulation or downregulation in the naphthalenecontaining treatments (NAP indep ) compared with the no-PAH control; and (iv) genes with no clear pattern in upregulation or downr egulation, irr espectiv e of the naphthalene concentration (no pattern).

Hydrocarbon degr ada tion
Naphthalene was fully degraded, regardless of the starting concentration, by cell suspensions of Cycloclasticus pugetii strain PS-1 over 168 h (Fig. 1 ), while cell numbers remained constantbetween 1.28 × 10 8 and 4.66 × 10 8 cells L −1 -over the course of the experiment ( Fig. S2 ).In high-NAP tr eatments, degr adation of 103.2 ± 0.93 mg L −1 naphthalene to 1.12 ± 0.70 mg L −1 was observed within 48 h.A maximum degradation rate of 4.16 mg L −1 h −1 was r eac hed within the first 12 h, follo w ed b y a decrease in degradation activity between 12 and 48 h (rates for each 12-h in- terv al wer e 1.28 and 1.53 mg L −1 h −1 , r espectiv el y).The r esidual na phthalene was full y degr aded during the r emaining incubation time.In both lo w-N AP (30.4 ± 0.58 mg L −1 ) and pulse-NAP (26.73 ± 0.33 mg L −1 ) treatments, the complete degradation of naphthalene occurred within 12 h of inoculation and after pulsing, with a degr adation r ate of 2.51 and 2.23 mg L −1 h, r espectiv el y.Na phthalene concentrations in all abiotic controls remained constant over the course of the experiment and no naphthalene was detected in PAH-fr ee contr ols ( Table S5 ).

Over all tr anscriptional activity in Cycloclasticus pugetii strain PS-1
To study gene expression profiles, we assessed similarities in log2-fold changes between samples.MDS between gene expression profiles sho w ed that the tr anscription intensity within tr eatments (conducted in biological triplicates) was m uc h mor e similar than between treatments ( Fig. S3 ).Naphthalene concentration in the samples and the time elapsed following the addition of PAH could explain the difference in transcription between the samples.Corr espondingl y, tr anscription in all samples where naphthalene had been completely consumed (starvation) was similar, regardless of the initial substr ate concentr ation.The 2-h lo w-N AP and pulse-NAP tr eatment tr anscriptomes also gr ouped together and the high-NAP samples after 2 h grouped separately from them ( Fig. S3 ).
To identify k e y metabolic functions, transcripts were grouped by functional categories.Among all tr anscripts, expr ession of genes related to aromatic hydrocarbon degradation was the highest of all KEGG modules (le v el 3) acr oss tr eatments (e v en in the no-PAH control), confirming that PAH degradation is an important metabolic feature of Cycloclasticus pugetii strain PS-1 (Fig. 2 ).Notabl y, onl y 39 genes were annotated in the KEGG database as related to PAH degradation.Ho w ever, w e identified a set of 154 genes potentially involved in PAH degradation by using available databases and liter atur e, indicating that there might be a larger group of genes related to PAH degradation, and the analysis based on KEGG modules might be underestimating the true activity of PAH-related genes in strain PS-1.).The expression is given in mean TPM per biological triplicate: 0 (white) to 2500 (dark blue).The number of genes in each functional category is shown in brackets.High and low concentration experiments received 100 and 30 mg L −1 of naphthalene at T 0 , r espectiv el y, wher eas the pulse tr eatments r eceiv ed 30 mg L −1 at T 0 and after 71 h.Transcription was determined after 2, 24 and 168 h for high-NAP treatments, after 2 and 71 h for lo w-N AP treatments (i.e.right before pulsing) and after 73 and 168 h for pulse-NAP tr eatments.PAH-fr ee contr ols with pyruv ate as carbon equiv alent wer e anal yzed after 24 h (no-PAH).

Transcriptional response of genes on the PAH-gene cluster
To further explore the transcriptional activity of Cycloclasticus pugetii strain PS-1 during PAH degradation, we examined the expression of a cluster of 33 genes, some of which were annotated in KEGG and all of which were previously identified as PAH-degradation genes in closely related Cycloclasticus spp.(Fig. 3 , Table S3, sheet C , and Table S6 ) (Kasai et al. 2003, Wang et al. 2018 ).This gene cluster was pr e viousl y described as "cluster E" and was differ entiall y expr essed in str ain P1 gr own on PAHs (na phthalene, phenanthrene and pyrene) compared with acetate-grown cells (Wang et al. 2018 ).Part of the cluster was published in 2003 as "phnA -cluster" (locus tag CYCPU_RS0111430 to CYCPU_RS0111480 in Table S6 ) and was described as enabling Cycloclasticus sp.strain A5 to degrade PAHs such as naphthalene and phenanthrene (Kasai et al. 2003 ).In the present study, the genes in the expression heatmap (Fig. 3 ) were sorted b y pathw ay step of naphthalene degradation and significant transcription was highlighted [genes with −1 < log2-fold change < 1 and adjusted P -value ( P adj ) ≤ 0.05].
The full PAH-gene cluster in strain PS-1 contains nine genes that encode RHD subunits, including rhd2 α and rhd3 α [also referred to as phnA1a in pr e vious publications (Kasai et al. 2003, McK e w et al. 2007, Dionisi et al. 2011 )].The RHD subunits were identified as functional marker genes for PAH degradation, encoding enzymes that catalyze the first step of phenanthrene and na phthalene degr adation, r espectiv el y (Wang et al. 2018 ).Further, CYCPU_RS0111455 was identified as putativ el y encoding an additional RHD β subunit based on its homology to other RHDs and the functional prediction of its active site (Paysan-Lafosse et al.We used the significance ( P adj ) and expression (log2-fold) change to define categories based on the transcriptional response of a gene to the presence or absence of naphthalene.Of the 33 genes in the P AH cluster , five genes-four encoding RHDs and one for a putative RHD-were assigned to the N AP pos category, meaning they w er e significantl y upr egulated in na phthalene-containing tr eatments and/or downr egulated under na phthalene-starv ation conditions compar ed with the no-PAH contr ol.Conv ersel y, thr ee RHD-encoding genes were assigned to the NAP neg category, demonstrating their lack of transcriptional upregulation in the presence of naphthalene and/or their upregulation under naphthalene-starvation conditions.A further 19 genes, like rhd3 α, wer e na phthalene-independentl y expr essed (NAP indep category), and their transcription did not change over time, e v en in the presence of different naphthalene concentrations.Ov er all, onl y fiv e of the 33 genes had low expr ession v alues, with a TPM below the 50th per centile (Fig. 3 , per centile definition see Fig. S4 ), demonstrating the high expression trend of genes within the PAH cluster in strain PS-1.More importantly, the expression of another five genes w as betw een the 90th and 95th percentile (mean TPM over all naphthalene-containing treatments between 451 and 878), and six were in the 95th percentile (mean TPM ov er all na phthalene-containing tr eatments equal and above 878), indicating very high expression (Fig. 3 ).These very highly expressed genes were annotated as encoding the small and large subunits of a RHD (CYCPU_RS0111490 and CYCPU_RS0111495, PAH-fr ee contr ols with pyruv ate as carbon equiv alent wer e measur ed after 24 h.r espectiv el y), a dihydr odiol dehydr ogenase/4-hydr oxythr eonine-4-phosphate dehydrogenase (CYCPU_RS0111480, second step of na phthalene degr adation), a ring-cleav a ge dio xygenase ( r cd ) (CY-CPU_RS0111460, third step of naphthalene degradation) and an (hydr oxyc hr omene-carboxylate) isomer ase [CYCPU_RS0111430, fourth step of na phthalene degr adation (Wang et al. 2018 )].The exceptionall y highl y expr essed genes CYCPU_RS0111490 and CY-CPU_RS0111495 are annotated to encode beta and alpha subunits of a RHD in the PAH cluster (Wang et al. 2018 ).Ho w e v er, the role of this RHD in naphthalene degradation is unknown as it remained untested in Cycloclasticus sp.strain P1 (Wang et al. 2018 ) and is unc har acterized in all other isolated Cycloclasticus sp.We, ther efor e, tentativ el y named them rhdPS1 β and rhdPS1 α, respectiv el y.Using the basic local alignment search tool-BLAST (Altschul et al. 1990, Zhang et al. 2000, Morgulis et al. 2008 )we confirmed that the genes with the highest nucleotide pairwise identity were aromatic ring-hydroxylating dioxygenase subunits alpha and beta from Cycloclasticus sp.strain P1 and Cycloclasticus zancles 78-ME with 98.01% and 99.79% for rhdPS1 α and rhdPS1 β, r espectiv el y.Pr otein functional analysis with InterPro (Paysan-Lafosse et al. 2023 ) confirmed that the genes encoded RHD alpha and beta subunits and that rhdPS1 α contains the Rieske [2Fe-2S] iron-sulphur domain.In the KO database (Kanehisa and Goto 2000 , Kanehisa et al. 2016a ,b ), rhdPS1 α and rhdPS1 β are listed (#K16320) as involved in aminobenzoate degradation.Curr ent liter atur e has inv estigated potentiall y similar genes in a Sphingomonas sp. and a Burkholderia sp.(Chang et al. 2003, Gai et al. 2010 ); ho w e v er, the r espectiv e genes ar e not highl y r elated: the nucleotide pairwise identities were 48.1% and 55.5% for rhdPS1 α and 49.3% and 50.2% for rhdPS1 β, r espectiv el y.The transcription of two other RHD-encoding genes in the PAH cluster ( rhd9 α-CYCPU_RS0111505 and rhd9 β-CYCPU_RS0111510) r esponded positiv el y to na phthalene av ailability.The mean TPM ov er all na phthalene-containing tr eatments for rhd9 α and rhd9 β, ho w e v er, w as betw een the 50th and 75th per centile, indicating lo w er expression and a potentially lesser role during naphthalene degradation.Two additional genes encoding for RHD subunits ( rhd3 α-CYCPU_RS0111470 and rhd3 β-CYCPU_RS0111465) were expr essed independentl y of na phthalene av ailability, while thr ee genes ( rhd2 α-CYCPU_RS0111555, rhd2 β-CYCPU_RS0111560 and phnA1b -CYCPU_RS0111475) were assigned to the NAP neg category.
We identified genes encoding all the enzymes prior to step 6 of the na phthalene-degr adation pathway (Wang et al. 2018 ), except for a hydratase aldolase, which is necessary for the fifth step.Ho w e v er, the two genes encoding for two alcohol dehydrogenases (sixth step) were not highly expressed (i.e.CYCPU_RS0111520 mean TPM NAP ≤ 50th percentile and CYCPU_RS0111545 mean TPM NAP between the 50th and 75th percentile), indicating that some PAH-related genes used in naphthalene degradation by strain PS-1 may be located elsewhere in the genome.
From further analysis of the significantly upregulated and do wnregulated genes betw een the high-N AP-2 h and the no-PAH control, we identified an additional 184 genes that were differen-tiall y expr essed ( Fig. S5 ).In total, 23 of those genes ( Fig. S5 -marked as black stars ) were found in the 154 PAH-related gene-database compiled from the literature ( Table S3, sheet A ), and only four of these 23 were part of the pr e viousl y described PAH cluster (Fig. 3 , Table S3, sheet C , and Table S6 ), indicating that manual curation of genes involved in PAH degradation is essential.

Transcription of genes highly involved in PAH degr ada tion
To identify genes for all the reactions actively involved in the na phthalene-degr adation pathway, a subset of genes that were highl y expr essed in the pr esence of na phthalene (i.e.TPM ≥ 90th percentile in at least one of the naphthalene-containing treatments) was selected from the curated literature database, containing 154 PAH-related genes, and further analyzed ( Table S3, sheets A and B ). Out of this subset of 43 selected genes (Fig. 4 , Table S3, sheet B ), 16 genes were already known as part of the pr e viousl y described PAH cluster ( Fig. S6 ).From these 43 genes, 12 fell into the NAP pos category ( Fig. S6 ) and eight of those were annotated as encoding for RHD subunits .T he only two genes in the NAP neg category (both already known from the PAH cluster, see Fig. S6 ) also encoded RHD subunits.Most of the remaining genes were expressed independently of naphthalene concentration (20 genes of the selection, three of them encoding RHD subunits, Fig. S6 ), while the final nine genes sho w ed no clear na phthalene-r elated pattern of expression.Further genes were potentiall y r ele v ant for the strain due to their high expression (15 genes-90th ≤ mean TPM NAP < 95th percentile) or very high expression (10 genes-mean TPM NAP ≥ 95th percentile).Although 14 genes, whic h wer e not part of the P AH cluster , were identified as highl y expr essed (90th ≤ mean TPM NAP < 95th percentile) and four as v ery highl y expr essed (mean TPM NAP ≥ 95th percentile), the genes with the highest TPM values were still rhdPS1 α, follo w ed b y rhdPS1 β, which fell into the NAP pos category .Additionally , a gene encoding a hydratase aldolase (CYCPU_RS0105800) that could potentially conduct the fifth step in the naphthalene-degradation pathway ( Fig. S1 ) had very high expression and was in the NAP pos category.Ov er all, 25 out of the selected 43 genes could be assigned to one of the steps from 1 to 7 ( Fig. S1 ) in the naphthalenedegradation pathway, and 14 of those genes coded for RHD subunits , potentially in volved in the initial step of the degradation pathway (Fig. 4 ).

Naphthalene-dependent transcription of functional marker genes
We identified genes encoding an unc har acterized RHD alpha and beta subunit (termed rhdPS1 α and rhdPS1 β) whose expression responded significantly to naphthalene ( Fig. S1 ).Although located in the pr e viousl y described P AH cluster , as illustrated in Fig. 5 (Kasai et al. 2003, Wang et al. 2018 ), RHD-PS1 has not been c har acterized in any Cycloclasticus species to date.Comparing the transcription with the expression of other RHD-encoding genes, the substantial expression of rhdPS1 α and rhdPS1 β in response to naphthalene availability and degradation activity of Cycloclasticus pugetii strain PS-1 suggests that the RHD-PS1 dominates the first step of naphthalene degr adation.The ne wl y described genes ar e, ther efor e, promising candidates for functional marker genes and could potentially be used for monitoring the na phthalene-degr adation activity of strain PS-1 with a qPCR-based method that quantifies the tr anscript-to-gene r atio (Baelum et al. 2008, Br ow et al. 2013, Ten-tori and Richardson 2020, Vogel et al. 2023b ), and could ultimately help to tr ac k PAH-degr adation activity in contaminated envir onments.Knoc k out m utant (Per ez-P antoja et al. 2009 ) as well as r ecombinant-pr otein-expr ession studies (Wang et al. 2018 ) would help to further determine the role of rhdPS1 α and rhdPS1 β in the na phthalene-degr adation pathway of Cycloclasticus pugetii strain PS-1.
Additionally, we identified four genes encoding RHD alpha and beta subunits that responded positively to naphthalene input and ar e likel y involv ed in the first step of na phthalene degr adation (Fig. 4 ).Two of the genes-rhd7 α (CYCPU_RS0104890) and rhd7 β (CYCPU_RS_0 104 895)-wer e highl y expr essed (90th ≤ mean TPM NAP < 95th percentile) and the resulting RHD-7 was pr e viousl y described as an enzyme that initializes fluoranthene degradation in Cycloclasticus sp.strain P1 (Wang et al. 2018 ).Further studies are needed to confirm the role of these other NAP pos RHDs in naphthalene degradation, ho w ever, the observed link between transcription and na phthalene-degr adation acti vity indicates the y may play a k e y role in Cycloclasticus pugetii strain PS-1.The highly, and na phthalene-dependent, expr essed genes might enable strain PS-1 to r a pidl y incr ease the number of enzymes (e.g.RHD-PS1 and RHD-7) and thereby quickly adapt to acute naphthalene contamination.
The transcription of three additional RHD-encoding genes among the PAH cluster (Fig. 5 ) depended significantly on the available naphthalene concentration.Ho w ever, the genes were downregulated in the presence and upregulated in the absence of naphthalene, so they were attributed to the NAP neg category.While rhd2 β was highl y expr essed (90th ≤ mean TPM NAP < 95th percentile) in the absence of naphthalene, we observed the substrateindependent transcription of rhd2 α in previous growth experiments with naphthalene and phenanthrene (Vogel et al. 2023b ).This could indicate that transcription of rhd2 α and rhd2 β are not only dependent on the availability of naphthalene, but also the cultivation conditions of the organism (growth experiment vs. cell-suspension experiment) or as a response to starvation (Vogel et al. 2023b ).Notably, these RHD-encoding genes might exhibit a differ ent tr anscriptional behavior for alternativ e PAHs and further studies should examine the transcription of the identified marker genes in response to other PAH substrates such as phenanthrene, biphenyl and naphthalene derivates.

Substr a te-independent tr anscription of PAH genes
No significant upregulation in na phthalene-ric h and/or downr egulation under na phthalene-starv ation conditions compar ed with the no-PAH controls indicated transcription independent of naphthalene availability (NAP indep category).
Two of those genes were encoding the ferredoxin and ferredoxin reductase (part of the P AH cluster , Fig. 5 ), which are important parts of the multicomponent RHD enzymes, and thus, rele v ant in the first step of PAH degradation.Ho w ever, the resulting components are often shared between different RHD enzymes (Wang et al. 2018 ).Given this, it is not surprising to find the genes as constitutiv el y or na phthalene-independentl y expr essed in an organism that has highly expressed genes for at least six different RHDs.
Out of 13 genes potentially encoding enzymes involved in the na phthalene-degr adation steps 2 to 6 ( Fig. S1 )-from the entire genome-10 were naphthalene-independently expressed, indicating that substrate-independent expression of genes in the PAH degradation pathway occurs regularly in strain PS-1.Although all the genes involved in the downstream steps are likely associated with PAH degradation, some of them could produce enzymes that ar e involv ed in other pathways (Hernaez et al. 2002 ).Ov er all, it r emains uncertain if the isomer ases, dehydr ogenases and hydratase aldolases are specific for PAH degradation.The non-exclusive use of the downstream enzymes in other metabolic pathways might explain the naphthalene-independent expression of some of the PAH-degradation genes; ho w ever, further investigation w ould be r equir ed to pr ov e this hypothesis.
RHDs, ho w e v er, ar e specific for the first step in PAH degradation (Gibson and P ar ales 2000, Singleton et al. 2012, Yesankar et al. 2023 ).Finding three genes encoding RHD subunits that are highly expressed (90th ≤ mean TPM NAP < 95th percentile) and naphthalene independent is, ther efor e, sur prising.The genes rhd3 α, rhd3 β and rhd5 β all fell into the NAP indep category, and while their highly expr essed natur e makes them potentiall y important for str ain PS-1, their transcription is likely not a response to acute naphthalene input.These results were corroborated in previous qPCR-based experiments wher e substr ate-independent tr anscription was observed for rhd3 α as well as for pahE (a hydratase aldolase; CY-CPU_RS0105800) in Cycloclasticus pugetii strain PS-1 (Vogel et al. 2023b ).
The reasons for substrate-independent transcription of PAHdegr adation genes, especiall y RHDs, ar e unknown.Giv en that PAH degr adation is centr al to the metabolism of Cycloclasticus pugetii strain PS-1, as shown by analyzing the transcription per functional category (Fig. 2 ), the corresponding enzymes might be essential for the lifestyle of this highly specialized or ganism.Constitutiv e-i.e.substr ate-independent-expr ession of functional genes has been observed in other hydrocarbondegrading bacteria (Cunliffe et al. 2006, Churchill et al. 2008 ), making it a potentially common-but overlooked-phenomenon.Further, substr ate-independent tr anscription of RHD-encoding genes could have practical reasons for this organism.Due to the PAH-independent expression of rhd3 α, rhd3 β, rhd5 β and potentially rhd5 α, RHD-3 and RHD-5 might be used as a "backgr ound" na phthalene-degr adation system for strain PS-1.The PAH-independent expression should allow a constant availability of these enzymes, which might lead to substrate-independent PAH-degr adation ca pacity.

Genetic redundancy in the PAH-degr ada tion pathway of Cycloclasticus pugetii strain PS-1
Genetic redundancy is a common phenomenon in all pathway steps, but particularly in the first step of na phthalene degr adation.Five RHDs (RHD-PS1, RHD-7, RHD-3, RHD-5 and RHD-2) were identified, for whic h tr anscription of at least one subunit-encoding gene was within the 90th percentile.S3, sheet C , and Table S6 ; the cluster was pr e viousl y described in closely related Cycloclasticus sp.strains A5 and P1 (Kasai et al. 2003, W Wang et al. 2018 ).Coloring indicates the percentile of expression level by mean TPM values over all naphthalene-containing experiments in our study ( Fig. S4 ).
The reasons for genetic redundancy in PAH-degradation genes, especially in genes encoding RHDs, are unknown.RHDs catalyze the first reaction, which is the rate-limiting step in PAHdegr adation, so the r ate of carbon and ener gy gain incr eases with an accelerated rate of the first step.Potentially, genetic redundancy of RHD-encoding genes could be beneficial for strain PS-1 by providing several enzymes for the same function and, ther eby, incr easing the rate of the first step in naphthalene degradation (Per ez-P antoja et al. 2009 ).Additionally, a fast consumption of naphthalene should lead to a steeper gradient between the bioav ailable dissolv ed na phthalene and the pur e compound.The steeper gradient in turn would accelerate the dissolution rate of naphthalene (Volkering et al. 1992, 1993, Vogel et al. 2023a ), leading to a faster substrate supply rate for the organism.
Genetic redundancy of RHD-encoding genes could, alternativ el y, indicate a highly specialized PAH-degradation strategy of Cycloclasticus pugetii strain PS-1.Potentially, two PAH-degradation enzymatic systems could be used, depending on the availability of PAHs: a steadily available "background" system and a specialized "r a pid-r esponse" system for acute PAH input.The "bac kgr ound" system would include RHDs that were encoded by substr ate-independentl y expr essed genes that would be constantl y av ailable and could degr ade c hr onic PAH contamination at trace amounts.Considering that most genes encoding these na phthalene-degr adation-pathw ay enzymes w er e expr essed independentl y of na phthalene addition (NAP indep category), it is reasonable to assume that the substrate-independent expression of highly relevant PAH-degradation genes is the default strategy for Cycloclasticus pugetii str ain SP-1.Mor eov er, all of the pathway steps other than the first step were shared for multiple PAHs (i.e .naphthalene , phenanthrene and pyrene) in a closely related Cycloclasticus sp.(Wang et al. 2018 ), which emphasizes how essential those genes are in this potential "background" system.
The "r a pid-r esponse" system, on the other hand, would include (mainly) RHDs that were upregulated and downregulated depend-ing on the availability of substrates and would be expressed in the case of a high-contamination e v ent to quic kl y degr ade substr ates at high concentrations .T his system would enable Cycloclasticus pugetii strain PS-1 to respond quickly and benefit from the sudden availability of high loads of substrate .Moreo ver, the enzymes encoded by genes in this system might be more substrate specific because they potentially only need to respond to one or two PAHs and convert them at high rates.RHD-3 and RHD-5 would, therefor e, potentiall y serv e as a "bac kgr ound" metabolism, whic h could be active in naphthalene-or PAH-free en vironments , whereas RHD-PS1 and RHD-7 might be responding to acute naphthalene input.

Cycloclasticus spp. in the environment
The hypothesized PAH-degradation strategy of Cycloclasticus pugetii strain PS-1 offers several potential benefits to the growth of Cycloclasticus spp. in the en vironment.En vir onmentall y occurring concentr ations ar e m uc h lo w er than those r outinel y used in the laboratory [e.g. the sum of dissolved concentrations of 64 PAHs in surface water samples was between 2 ng L −1 (Indian Ocean) and 3.5 ng L −1 (North Atlantic) (González-Gaya et al. 2016 )].Consequently, we hypothesize that Cycloclasticus spp., as part of a nativ e comm unity, ar e commonl y oper ating with the "bac kground" system and high-substrate affinity system of enzymes, for which the genes are substrate-independently expressed.Conv ersel y, the "r a pid-r esponse" system, with enzymes encoded by PAH-dependentl y expr essed genes, is likel y trigger ed by higher substr ate concentr ations and consists of low-substr ate affinity enzymes.
We propose that the "rapid response" system would be triggered by a naphthalene concentration lo w er than that of naphthalene solubility in seawater (28.96 mg L −1 ) given that envir onmental concentr ations of na phthalene ar e, in r eported cases, m uc h lo w er than 28.96 mg L −1 (Dier cks et al. 2010 , González-Gaya et al. 2016, Vogel et al. 2023a ).This "r a pid r esponse" en-zymatic system could, for example, potentially have operated in the contaminated hydrocarbon plume that formed after the Deepwater horizon oil spill, where PAH concentrations amounted to a maximal 189 μg L −1 (Diercks et al. 2010 ).Changing the PAH/naphthalene input and thereby substrate a vailability, hence , is an important environmental condition that can influence the PAH/na phthalene-degr adation r ate consider abl y (Mostafa et al. 2019, Bacosa et al. 2021 ).Further studies should investigate the thr eshold concentr ations r equir ed to activ ate the "r a pid r esponse" system.
Identifying genes of both enzymatic systems in our model organism Cycloclasticus pugetii strain PS-1 highlights the level of adaption to a PAH-degradation lifestyle.Considering Cycloclasticus pugetii strain PS-1 was isolated from Puget sound (Dyksterhouse et al. 1995 )-a habitat with man y natur al oil seeps-a high le v el of adaption to both chronic traces of PAH, as well as recurring highinput PAH pollution, is not surprising.We anticipate that Cycloclasticus spp.can outcompete other PAH-degrading bacteria that only have one step or a subset of steps in the PAH-degradation pathway.The adv anta ge of the Cycloclasticus spp.would arise fr om them having multiple genes for each step of the pathway and hypotheticall y a "r a pid-r esponse" system of (mainl y) RHDs (Sier adzki et al. 2021 ).The substrate-dependent transcription of (mainly) RHDs might give Cycloclasticus pugetii strain PS-1 the possibility to r a pidl y incr ease its degr adation r ate, whic h could be one explanation why Cycloclasticus spp.are detected ubiquitously and can be isolated from PAH-amended enrichment cultures (Wang et al. 2008, Gutierrez et al. 2013, Cui et al. 2014, Rizzo et al. 2019 ).

En vironmental implica tions
With the increasing amount of envir onmental meta genomic and metatranscriptomic studies, it is important to consider what the detection of genes and transcripts from the NAP pos and NAP indep categories would imply for the r espectiv e micr obial comm unities and en vironments .T he detection of genes from the NAP indep category, such as rhd3 α and rhd3 β, or genes encoding downstream enzymes, such as pahE , could be-given that the genes are truly PAH-pathway specific-used as marker genes for PAH-degrading or ganisms, as pr e viousl y suggested (Dionisi et al. 2011, Liang et al. 2019 ).The transcription of those genes, ho w e v er, does not imply that the organisms are actively degrading PAHs, but rather indicates a general metabolic activity.Further, an environment with micr obial comm unity members possessing NAP indep genes may not always be pristine (Gutierrez 2019b ), even if no apparent PAH contamination is detectable (Angelova et al. 2021 ).We, ther efor e, posit that some environments may be continuall y pur ged fr om constant trace inputs of PAHs without enriching PAH degraders or inducing PAH-degrading pathwa ys , given that a "background" system of genes is constantly transcribed.
By contr ast, genes fr om the NAP pos category, especiall y rhdPS1 α and rhdPS1 β, could potentially not only identify PAH-degrading organisms, but also be used as functional marker genes for high na phthalene-degr adation activity of Cycloclasticus pugetii strain PS-1.Determining the transcript-to-gene ratio of NAP pos genes of a microbial community could be a valuable tool to quantify the cell-n umber-inde pendent degr adation r ate of specific compounds and, thereby, assess the P AH-biodegradation performance after high-concentration contamination e v ents, like oil spills or in a laboratory experiment (Wilson et al. 1999, Baelum et al. 2008, Brow et al. 2013, Tentori and Richardson 2020, Vogel et al. 2023b ).Further, identifying organisms with NAP pos genes in an environmental community could indicate a faster environmental recovery from acute high-input contamination.
Nonetheless, genetic redundancy in k e y genes for PAH degradation makes it difficult to quantify the PAH-degradation activity in environments based on the expression of a single gene or set of genes.Detailed knowledge about the PAH-degrading community and the categorization of the involv ed PAH-degr adation genes would be necessary to select a suitable set of target genes for the in vestigated en vironment.Moreo ver, the expression of such identified target genes could vary between strains and could be sensitive to other en vironmental factors .Further studies like a global assessment and c har acterization of RHDs in all curr entl y av ailable metagenomes and transcriptomes are necessary before a robust set of genes for the quantification of in situ PAH-degradation rates can be proposed.

Se v er al open questions remain regarding genetic redundancy in
Cycloclasticus pugetii strain PS-1.Further studies, including knockout mutants and enzymatic assa ys , ar e r equir ed to inv estigate for molecular redundancy (Per ez-P antonja et al. 2009 ), identify which of the RHDs have an affinity for other PAHs, or to confirm these RHDs are performing the initial step in naphthalene degradation (Wang et al. 2018 ).Additional r esearc h is necessary to determine whether the expressed RHD-encoding genes are induced by naphthalene, but the corresponding enzymes are not produced and/or not used in naphthalene degradation.
Mor eov er, in a closely related Cycloclasticus sp., the genes encoding RHD-2, RHD-3 and five other genes associated with the PAH-degradation pathw ay w ere co-regulated b y the same regulator ( Table S6 ) (Wang et al. 2021 ).In Cycloclasticus pugetii strain PS-1, ho w e v er, these genes were expressed differently: some NAP neg and others were substrate independent.Our understanding of conditions and substrates influencing the regulation of PAHdegradation genes in Cycloclasticus spp.remains undefined.Further studies are needed to determine if an alternative regulation mechanism is used in Cycloclasticus pugetii strain PS-1, or if the genes are co-regulated but the mRNA of the apparent NAP neg genes is potentially degraded and, therefore, not substrateindependently detected.
Because the model organism in this study is an isolated and very well studied Cycloclasticus sp., further studies could investigate if the observed transcriptional patterns changed in the case when Cycloclasticus pugetii strain PS-1 would live as part of a PAHdegr ading comm unity.Further, Arctic Cycloclasticus spp.from a natur al comm unity wer e shown to have a different set of RHDs than expected from the genomes of the isolated Cycloclasticus spp.(Vogel et al. 2023b ).Inv estigating the tr anscription of the PAH genes in environmental Cycloclasticus spp.under PAH-free or e v en hydr ocarbon-fr ee conditions-giv en that some Cycloclasticus spp.can degrade alkanes (Rubin-Blum et al. 2017, Gutierrez et al. 2018 )-would be the next step.Because the cell numbers of Cycloclasticus spp.under hydr ocarbon-fr ee conditions ar e typically low and mostly not detectable, assessing the transcription of PAH-degradation genes will be c hallenging.Ther efor e, quantifying tr anscripts (thr ough qPCR) or conducting metatr anscriptomic studies following an environmental contamination event when all hydr ocarbons ar e consumed, similar to our starv ation conditions, could elucidate the role of Cycloclasticus spp. in an environmental micr obial comm unity.

Conclusion
The na phthalene-dependent tr anscription of m ultiple RHDs indicated that strain PS-1 is very well adapted to respond instantly to high-concentration inputs of PAHs .T his fast reaction can potentially be achieved by increasing the overall degradation rate through the maintenance of an enzymatic "rapid response" system.Curr entl y, it is not possible to deduce the degraded PAH or the envir onmental degr adation activity by tar geting the tr anscription of a single functional gene or set of genes .T he ne wl y described functional marker genes rhdPS1 α and rhdPS1 β, ho w e v er, ar e pr omising tar get-gene candidates to quantify naphthalenedegr adation activities thr ough DN A/RN A-based methods, because their transcription seems to correlate to naphthalene degradation in Cycloclasticus pugetii strain PS-1.Using these genes, the monitoring of PAH degradation could, in future, be conducted in a highthroughput manner by using molecular-based methods such as the TtG ratio.This in turn could facilitate more efficient PAH bioremediation because the measures or conditions could be adapted mor e r a pidl y when the monitoring pr ov es that the degr adation r ate is c hanging.Further , an additional set of P AH-degradation genes that were expressed independently of naphthalene was also identified.Those genes are involved in all reactions of the curr entl y known na phthalene-degr adation pathway in Cycloclasticus spp., indicating there might be another set of PAH-degrading enzymes that is potentially used as a "background" system for the degr adation of envir onmentall y occurring tr ace amounts of PAHs.
The observed genetic redundancy in PAH-related genesparticularly RHDs-along the naphthalene-degradation pathway and their varying levels of transcription under different conditions has not been r eported pr e viousl y and should be further studied.This genetic flexibility indicated by the hypothesized two enzymatic systems could enable PAH degraders to respond to fluctuating hydrocarbon inputs in a need-based way.Understanding the degradation pathway used by k e y PAH degraders under varying conditions, such as low vs. high PAH concentrations, is important to assess contamination scenarios corr ectl y.These assessments could be used to enhance bacterial PAH degradation in, for example , bioremediation scenarios .

Figure 1 .
Figure 1.Na phthalene concentr ation in mg L −1 ov er time, quantified by GC-MS.Squares show concentrations in high-NAP (100 mg L −1 ), lighter blue triangles illustrate concentrations in lo w-N AP (one dosage; 30 mg L −1 ) and darker blue diamonds give the concentrations in the pulse-NAP treatments (2 ×30 mg L −1 ).Pulsed treatments were treated equally to lo w-N AP treatments up until 71 h and, therefore, only sampled after 73, 85, 97 and 168 h.Error bars represent the standard deviation of the r espectiv e thr ee sacrificial samples and ar e sometimes within the mark er.Time points for RNA-sequencing are marked with stars and colored according to treatments.

Figure 2 .
Figure2.Expression per set of genes accumulated in KEGG modules (le v el 3).The expression is given in mean TPM per biological triplicate: 0 (white) to 2500 (dark blue).The number of genes in each functional category is shown in brackets.High and low concentration experiments received 100 and 30 mg L −1 of naphthalene at T 0 , r espectiv el y, wher eas the pulse tr eatments r eceiv ed 30 mg L −1 at T 0 and after 71 h.Transcription was determined after 2, 24 and 168 h for high-NAP treatments, after 2 and 71 h for lo w-N AP treatments (i.e.right before pulsing) and after 73 and 168 h for pulse-NAP tr eatments.PAH-fr ee contr ols with pyruv ate as carbon equiv alent wer e anal yzed after 24 h (no-PAH).

Figure 3 .
Figure 3. Expression in PAH cluster sorted by pathway step.Heatmap showing the expression (mean TPM per biological triplicate for each treatment)of genes associated with PAH degradation in the P AH cluster .Genes are sorted by pathway step (regulator, first step, second step, third step, fourth step, sixth step and other).Significant log2-fold changes to the no-PAH control are displayed ( P adj ≤ 0.05, 1 < log2-fold change > 1), 90th ≤ mean TPM NAP < 95th percentile ( + ), and mean TPM NAP ≥ 95th percentile ( ++ ) are indicated.High and low concentration experiments received 100 and 30 mg L −1 of na phthalene, r espectiv el y, at T 0 , wher eas the pulse treatments received 30 mg L −1 at T 0 and after 71 h.Transcription was determined after 2, 24 and 168 h for high-NAP treatments, after 2 and 71 h for low-NAP treatments (right before pulsing) and after 73 and 168 h for pulse-NAP treatments.PAH-fr ee contr ols with pyruv ate as carbon equiv alent wer e measur ed after 24 h.

Figure 4 .
Figure 4. Selected, highl y expr essed PAH degr adation genes.Heatma p showing the expr ession: mean TPM fr om 0 (white) to 4100 (dark blue) per biological triplicate.Genes were selected if TPM was within the 90th percentile in at least one naphthalene-containing treatment.90th ≤ mean TPM NAP < 95th percentile ( + ), and mean TPM NAP ≥ 95th percentile ( ++ ) are indicated.Significant log2-fold changes compared with the control (pyr.24h) are displayed ( P adj ≤ 0.05, 1 < log2-fold change > 1).Genes are sorted by pathway step (from top to bottom: first step, second step, third step, fourth step, fifth step, sixth step, eighth step and others).High-and low-concentration experiments received 100 and 30 mg L −1 of naphthalene, respectively, at T 0 , whereas the pulse treatments received 30 mg L −1 at T 0 and after 71 h.Transcription was determined after 2, 24 and 168 h for high-NAP treatments, after 2 and 71 h for lo w-N AP tr eatments (i.e.right befor e pulsing), as well as at 73 and 168 h for pulse-NAP tr eatments.PAH-fr ee contr ols with pyruv ate as carbon equivalent were measured at 24 h.Genes that are part of the PAH-cluster are indicated [W Wang et al. ( 2018 ) ( * ), Kasai et al. ( 2003 ) ( * * )].

Figure 5 .
Figure 5. Gene cluster of PAH degr adation-r elated genes within the Cycloclasticus pugetii PS-1 genome.Further information on the genes is provided in TableS3, sheet C , and TableS6; the cluster was pr e viousl y described in closely related Cycloclasticus sp.strains A5 and P1(Kasai et al. 2003 , W Wang et  al. 2018 ).Coloring indicates the percentile of expression level by mean TPM values over all naphthalene-containing experiments in our study ( Fig.S4).Color coding for na phthalene-concentr ation dependence: blue, NAP pos ; pink, NAP neg ; gray, NAP indep .PAH cluster spanning 37 146 bp; total length of the genome is 2 383 924 bp.