Tao Sun, Shubin Li, Guangsheng Pei, Lei Chen,*, Weiwen Zhang,3,*

1 Laboratory of Synthetic Microbiology, School of Chemical Engineering & Technology, Tianjin University, Tianjin 300072, China

2

Frontier Science Center for Synthetic Biology & Key Laboratory of Systems Bioengineering, Ministry of Education of China, Tianjin 300072, China

3 Center for Biosafety Research and Strategy, Tianjin University, Tianjin 300072, China

Keywords:

ABSTRACT

1. Introduction

High oil prices and growing concerns over the energy crisis as well as the global climate change are driving investment and innovation in the renewable biofuels [1]. Among them, alkanes, with better chemical properties like low water solubility and high energy density [2], have attracted significant attention in recent years.For example,biosynthesis of C15 and C17 long-chain hydrocarbons at 2.54 g?L-1and 1.47 g?L-1has been achieved in Escherichia coli (E. coli) and Yarrowia lipolytica, respectively [3,4]. In addition to the conventional sugar-based chassis, recent synthetic biology efforts have also led to the development of photosynthetic cyanobacteria as ‘‘a(chǎn)utotrophic cell factories” for biosynthesis of various biofuels,such as ethanol,butanol and alkanes directly from CO2[5,6]. For example,by overexpressing aldehyde decarboxylase encoding gene orf1593 in Synecchococcus elongatus (S. elongatus)PCC 7942, Kaiser et al. [7] achieved a production of alkanes at a level of 2.30 mg?L-1. More recently, Yunus et al. [8] introduced acyl-ACP reductase (AAR) and aldehyde deformylating oxygenase(ADO) of S. elongatus PCC 7942, thioesterase A (TesA) of E. coli and fatty acid photodecarboxylase (FAP) of Chlorella variabilis in Synechocystis sp. PCC 6803 (hereafter Synechocystis) and achieved a production of alkanes at a level of 110 mg?L-1.Although the current production level is still very low, these exciting progresses clearly demonstrated a feasibility of renewable production of alkanes directly from CO2.

Most alkanes are toxic to cyanobacterial cells [9]. Kamarainen et al. [10] evaluated the tolerance of two cyanobacterial hosts(i.e., Synechocystis and S. elongatus PCC 7942) to the intended end-products,and found that among the externally added alkanes,C6 alkane caused more inhibition than both C3 and C11 alkane.Our previous study also found that 50% of the cell growth of Synechocystis could be arrested by exogenous n-hexane at a concentration of only 0.8% (vol) [11], suggesting that low tolerance may represent a hurdle to further expand the application of engineered cyanobacteria cells. To determine the mechanism related to n-hexane tolerance in cyanobacteria, Liu et al. [11] previously applied an iTRAQ-LC-MS/MS quantitative proteomics to Synechocystis under exogenous n-hexane stress,and the results showed that a global change of cellular metabolism occurred after n-hexane stress; notably, a large number of transporters and membrane-bound proteins, proteins against oxidative stress,several signal transduction proteins, and proteins related to sulfur relay system and photosynthesis were induced.

It is known that microbes tend to employ multiple resistance mechanisms in dealing with stress of single biofuel product [12–16],and it could be challenging to achieve tolerance improvement by sequential multi gene modifications. Alternatively, it has been proposed to focus on regulatory genes for tolerance improvement against biofuels, as the proper manipulation of a regulatory gene might achieve simultaneous modification of many genes related to tolerance [17]. As a successful example, Kaczmarzyk et al. [18]found that overexpression of an RNA polymerase sigma factor sigB in Synechocystis led to an enhanced tolerance to butanol in a butanol-shock experiment. Our recent efforts of using functional genomics approach to study metabolic responses to various biofuels also led to discovery of two response regulator genes (i.e.,slr1037 and sll0039) related to butanol tolerance [19,20], several transcriptional regulators (i.e., sll0794, sll1392, sll1712 and slr1860) related to ethanol tolerance, and a non-coding sRNA nc117 with its positive regulating target slr0007 related to short chain alcohols tolerance in Synechocystis[21–23].In addition,overexpression of sll0039, nc117 and slr0007 in the wild type has resulted in a slightly increased tolerance to butanol and short chain alcohols in Synechocystis, respectively [20,23].

To identify gene targets to improve n-hexane tolerance in Synechocystis,in this work,we first applied RNA-seq transcriptomics to reveal the global responses of Synechocystis to n-hexane stress at transcriptional levels. Based on the transcriptomics analysis and the comparative growth assay of knockout mutants [22], a knockout mutant of slr0724 gene encoding an HtaR suppressor protein(⊿slr0724 mutant) was confirmed with enhanced tolerance to nhexane than the wild type Synechocystis. As a single gene deletion led to increased tolerance to n-hexane, the ⊿slr0724 mutant may serve as a valuable cyanobacterial chassis for the autotrophic biosynthesis of biofuels.

2. Materials and Methods

2.1. Bacterial growth conditions and biofuels treatment

Synechocystis was grown in BG11 medium(pH 7.5)under a light intensity of approximately 50 μmol photons?m-2?s-1in an illuminating incubator of 130 r?min-1at 30 °C (HNY-211B Illuminating Shaker, Honour, China) [14] and the Synechocystis knockout mutants used in this study was supplemented with 10 μg?L-1chloramphenicol.Cell density was measured on a UV-1750 spectrophotometer(Shimadzu,Japan)at OD730.For growth,40 μl fresh cells at OD730of 0.1 collected by centrifugation and were then inoculated into 200 μl of BG11 liquid medium in 96-well cultivation plates.To ensure accuracy of finding, the growth patterns were further confirmed by growth in 100 ml flasks, in which 5 ml fresh cells at OD730of 0.1 were collected by centrifugation and were then inoculated into 25 ml of BG11 liquid medium in a 100 ml flask.Culture samples (200 μl) were taken and measured at OD730every 12 h. For n-hexane treatment, 0.5%–0.6% (volume ratio) n-hexane was added at beginning of the cultivation. Growth experiments were repeated at least three times to confirm the growth patterns.

2.2. Transcriptome sequencing

RNA-seq transcriptome sequencing was carried out as described previously [24]: (i) Samples composition: the wild type Synechocystis grew under BG11 medium with or without 0.55% (vol)n-hexane were collected for the transcriptome analysis at 24 h,48 h and 72 h,respectively,each sample with two biological replicates; (ii) RNA preparation and cDNA synthesis: approximately 10 mg of cell pellets were frozen by liquid nitrogen immediately after centrifugation. Total RNA extraction was achieved using a miRNeasy Mini Kit (Qiagen, Valencia, CA, USA). 500 ng total RNA were subjected to cDNA synthesis using a NuGEN Ovation?Prokaryotic RNA-seq system according to manufacturer’s protocol(NuGEN, San Carlos, CA, USA); (iii) Next-generation sequencing:RNA 2x100 bp paired-end sequencing was performed using Solexa Genome Analyzer II (Illumina, USA) using the standard protocol.The image deconvolution and calculation of quality value were performed using Goat module (Firecrest v.1.4.0 and Bustard v.1.4.0 programs)of Illumina pipeline v.1.4.Sequenced reads were generated by base calling using the Illumina standard pipeline.

2.3. Transcriptomics data analysis

Prior to read mapping, sequence reads were pre-processed using NGSQCToolkit (Version: 2.3) to remove low-quality bases,and reads shorter than 20 bp [25]. For paired-end Illumina reads,both pairs were removed if either pair mapped to rRNA sequences.Remaining reads were mapped to the Synechocystis genome which was downloaded from NCBI using Bowtie(Version:2.0.0)with the default parameters [26]. Raw counts of reads that uniquely mapped to each gene region were calculated by HTseq (Version:0.6.1) [27]. Then reads counts were normalized to the aligned RPKM (Reads Per Kilobase per Million mapped reads) to obtain the relative expression levels. With raw counts of reads mapping to unique genes as input, differential expression analysis was performed using R packages of DESeq2 for comparisons between samples [28]. The resulting p-values were adjusted using Benjamini and Hochberg’s approach for controlling the false discovery rate(FDR). Genes with log2 fold change >1 and an adjusted p-value<0.01 were determined as differentially expressed between nhexane and control conditions.

2.4. qRT-PCR analysis

The identical samples of RNA-seq transcriptomics were used for qRT-PCR analysis.The qPCR reaction was carried out in 10 μl reactions containing 5 μl of UltraSYBR Mixture (CW Biotech, Beijing,China), 3 μl dH2O, 1 μl dilute template cDNA and 1 μl of each PCR primer, employing the StepOnePlusTMReal-Time PCR system(Applied Biosystems, Foster City, CA) [13]. Three technical replicates were performed for each sample. Data analysis was carried out using the StepOnePlus analytical software (Applied Biosystems, Foster City, CA) and the 2-ΔΔCTmethod [29]. Data was presented as ratios of the amount of normalized transcript in the treatment to that from the wild type control.

2.5. Construction and analysis of knockout mutants

A fusion PCR based method was employed for the construction of gene knockout fragments [30]. Briefly, for the gene target selected, three sets of primers were designed to amplify a linear DNA fragment containing the chloramphenicol resistance cassette(amplified from a plasmid pACYC184) with two flanking arms of DNA upstream and downstream of the target gene.The linear fused PCR amplicon was used directly for transformation into Synechocystis cells by natural transformation [22].

2.6. Complementation of Slr0724 in the ⊿slr0724

Gene expressing vector pJA2 was a broad host replicating vector which was kindly provided by Prof.Paul Hudson(KTH Royal Institute of Technology of Sweden)[31].The ORF of Slr0724 was amplified by PCR using the primer 5′-ATGGCGGGAATATTATCCCCAT-3′and 5′-CTATTCATCCTCGTCCAGCAATG-3′.The purified PCR product was phosphorylated by T4 polynucleotide kinase and the vector pJA2 was digested by XbaI and BamHI restriction enzymes; then the digested vector was dephosphorylated and blunted by alkaline phosphatase and T4 DNA polymerase, respectively. At last, T4 ligase was used to construct pJA-Slr0724.pJA-Slr0724 and a control vector were introduced back into the ⊿slr0724 mutant by electrotransformation. The ⊿slr0724-complementation strain and the control strain were named ⊿slr0724/pJA0724 and ⊿slr0724/pJAC,respectively.

2.7. Functional enrichment analysis

The COG (Cluster of Orthologous Groups of proteins) of the genes was classified according to the eggNOG functional classification system[32].The metabolic pathway analysis of the genes was conducted according to KEGG (Kyoto Encyclopedia of Genes and Genomes) Pathway Database. The significance of whether differently expressed genes were enriched in a given functional pathway or functional category was calculated by the Wallenius non-central hypergeometric test using the GOseq R package in which genelength bias was corrected [33]. Functional enrichment was assessed separately for differential expressed genes at different time points, respectively.

3. Results and Discussion

3.1. Effects of n-hexane stress on Synechocystis

In this study,effects of n-hexane on cell growth were measured with the wild type Synechocystis at an n-hexane concentration of 0.50%, 0.55% and 0.60% (vol), respectively (Fig. 1). The results showed that n-hexane at 0.55% (vol) concentration caused more than 50% growth decrease at 48 h, and the concentration was selected in the following study.For RNA-seq transcriptomic analysis, the wild type Synechocystis cells grown in BG11 media with or without 0.55%(vol)n-hexane were collected at 24 h,48 h and 72 h,respectively.

3.2.RNA-seq transcriptomics reveals metabolic responses to n-hexane

Fig. 1. Effects of n-hexane on Synechocystis. Growth curves of the wild type Synechocystis under control BG11 medium (WT) and medium with n-hexane at 0.5%–0.6% (vol). The error bar represents the calculated standard deviation of the measurements of three biological replicates.

Both n-hexane treated and control RNA samples collected at 24,48 and 72 h were subjected to RNA-seq analysis.A total of 244 million raw sequencing RNA (100 nt) reads was obtained, with average reads of 20 million reads per sample. After trimming off the 3′and 5′adapter sequences, removing poor low-quality reads(reads with more than 30% Phred scores smaller than 20 or reads shorter than 20 bp)and reads mapping to rRNA,a total of 157 million qualified mRNA-based sequence reads were used to map the Synechocystis genome (Table S1 in Supplementary Material).Although the effective sequencing depth may vary in different samples, the qualified sequence reads were able to match to all 3189 encoding genes in the Synechocystis genome, suggestive of good RNA-seq coverage.

To visualize the transcriptomic data and their repeatability, we normalized the raw reads counts to RPKM and then subjected them to a PCA analysis.The score plot showed that control and n-hexane treated samples were visibly separated well in the first principal component (Fig. 2), suggesting significant differences in terms of the metabolic responses to n-hexane stress at the transcript level.In addition, the plot showed that two biological replicates were clustered together for most of the conditions, with only exception for n-hexane treated replicate samples H721 and H722 of 72 h,which may be due to the cell aging and relatively poor RNA quality after long treatments of n-hexane.Moreover,a moving trend of the transcriptomic profiles along the second principal component seemed correlated with the growth phases of Synechocystis.

Differential gene expression analysis was performed using DESeq2 software [34]. Using a strict criterion as log2 fold change>1 and an adjusted p-value<0.01,a total of 797 genes were differentially expressed between n-hexane-treated and control conditions, among which a total of 318, 234, 252 genes were downregulated at 24,48 and 72 h,respectively;and 278,213,380 genes were up-regulated at 24, 48 and 72 h, respectively (Table S2 in Supplementary Material). Consistent to early studies [11,12], the differentially regulated genes were distributed in a wide range of metabolic pathways, suggesting a global metabolic response occurred upon n-hexane stress in Synechocystis (Fig. S1).

3.3. Signal transduction function enriched among differentially regulated genes by n-hexane

Fig.2. PCA plot of RNA-seq profiles of different time points.Each point on the graph represents one biological replicate sample;samples with different treatments were indicated by colors and shape (circles represent control sample, squares represent n-hexane treated samples). Proportions of the principal components 1st and 2nd accounting for 40.1% and 28.7% total variation, respectively.

To reveal metabolic responses to n-hexane, COG enrichment analyses were performed for differently regulated genes. The COG category enrichment result showed that several functional categories were enriched,among which‘‘[J]Translation,ribosomal structure and biogenesis” was significantly induced at all three time points; while ‘‘[S] Function unknown” changed notably at 48 h(Fig.3).Consistently,in our previous study using quantitative proteomics to investigate the stress response of Synechocystis to exogenous n-hexane, these two categories‘‘Translation, ribosomal structure and biogenesis” and ‘‘Function unknown” respectively took up 7.55% and 5.61% among the identified regulated genes[11]. In addition, ‘‘[T] signal transduction mechanism” was also enriched at all the time points due to its prominent changes(Fig. 3). As evidence has shown that signal transduction mechanism plays an important role in stress transduction and acclimation [35], and our previous work have also found that ‘‘signal transduction mechanism”take up as high as 10.9%of the identified regulated genes under n-hexane stress [11], which indicates that signal transduction genes be important to cope with n-hexane stress in Synechocystis.

3.4. Quantitative real-time RT-PCR (qRT-PCR) validation of transcriptomic analysis

To validate the quantitative results from the transcriptome analysis, a subset of 10 genes were randomly selected based on their expression levels for qRT-PCR analysis of samples collected from three time points (the gene ID and their related primer sequences used for real-time RT-PCR analysis were listed in Table S3). Among them, four genes (i.e., slr0146, slr0147, sll1239 and ssl2250) were significantly down-regulated and one gene slr0724 was slightly down-regulated at three time points; two genes (i.e., sll1599 and sll0671) were up regulated at three time points; and four genes without significant changes (sll1404 and ssl2749 slightly up regulated and down regulated in two time points, respectively, while ssr2802 unchanged). Comparative qRTPCR analysis between the n-hexane-treated and control samples showed a visible positive correlation with correlation coefficients greater than 0.80 using the Pearson correlation analysis (Fig. 4),suggesting an overall good quality of this transcriptome analysis.

3.5. Identification of regulatory genes involved in n-hexane tolerance

It is well known that regulatory genes are involved in correlating metabolic responses against complicated environmental stress in Synechocystis[17,35,36]. More recently,we have found that tolerance to biofuels such as butanol and ethanol were also regulated by various regulator genes in Synechocystis [20–23,37]. However,so far no study on the regulation of alkane tolerance has been reported. In our previous work, a library of knockout mutants for 34 putative transcriptional regulator-encoding genes were constructed[22],which included mutants for the regulatory genes differential regulated by n-hexane as revealed by transcriptomic analysis. We thus subjected the 34 mutants and the wild type Synechocystis for a comparative phenotype analysis under nhexane stress.The growth curves of the mutants and the wild type Synechocystis on BG11 media with or without n-hexane (0.55%(vol)) were determined. The results showed that a mutant of slr0724 encoding an HtaR suppressor protein homolog became more tolerant to n-hexane toxicity(0.55%(vol))than the wild type Synechocystis (Fig. 5(a)),and the enhanced tolerance became more significant when the n-hexane concentration was increased to 0.60% (vol) (Fig. 5(b)), while no visible growth difference was observed under normal growth medium, suggesting that slr0724 may function as a negative regulator to the tolerance against nhexane stress, and its deletion increased the survival and growth under n-hexane stress. Consistently, although slr0724 showed slightly down-regulated in the transcriptomic analysis result (approximately 0.84, 0.92 and 0.91 at 24 h, 48 h and 72 h, respectively; Table S2), the qRT-PCR result revealed that slr0724 was significantly down-regulated between the n-hexane-treated and control samples (approximately 0.49, 0.33 and 0.47 at 24 h, 48 h and 72 h, respectively).

Fig. 3. COG enrichment analysis of differentially expressed genes. The color in the heatmap represents the –log10 p-value of Wallenius non-central hypergeometric tests.

Fig.4. Correlations between qRT-PCR and RNA-seq analyses for selected genes(i.e.,slr0146,slr0147,sll1239,sll1404,ssr2802,sll1599,slr0724,ssl2749,ssl2250 and sll0671):(a)24 h; (b) 48 h; (c) 72 h. For horizontal coordinate, values represent the means of fold changes of gene expression in biological replicates under n-hexane stress condition compared with control condition,revealed by RNA-seq.For vertical coordinate,values represent the means of fold changes of gene expression in biological replicates under nhexane stress condition compared with control condition, revealed by qRT-PCR.Error bars represent standard deviation.Pearson correlation coefficients were shown on the top of each figure.

Fig. 5. Growth of the wild type (WT) and the ⊿slr0724 mutant under control and n-hexane stress conditions: (a) WT and ⊿slr0724 under normal medium or medium with 0.55% (vol) n-hexane. (b) WT and ⊿slr0724 under normal medium or medium with 0.60% (vol) n-hexane. The error bar represents the calculated standard deviation of the measurements of three biological replicates.

To further confirm the involvement of gene slr0724 in n-hexane tolerance,we constructed a slr0724-complementation strain in the⊿slr0724 mutant named ⊿slr0724/pJA0724 using a shuttle vector pJA2, and the effect of the complementation was confirmed by comparatively measuring growth time-curve of relative strains.The results showed that in the normal BG11 medium,three strains grew almost equally (Fig. 6(a)); while in the BG11 medium with 0.55% (vol) n-hexane, the growth of ⊿slr0724/pJA-Slr0724 was almost fully recovered to that of wild type,and significantly lower than the control mutant ⊿slr0724/pJAC (⊿slr0724 with a control pJA2 vector) under the same stress (Fig. 6(b)), demonstrating that slr0724 was involved in n-hexane tolerance in Synechocystis.

The slr0724 gene was previously reported to encode an AbrB/Maze/Mraz-like regulator [38,39]. Protein domain analysis using NCBI’s CDD(Conserved Domain Database)online software showed that Slr0724 contains two conserved protein domains [40], a PRK09974 domain (putative regulator PrlF, E-value 1.22×10-63)and an AbrB domain (regulators of stationary/sporulation gene expression, E-value 7.66×10-11). PrlF is part of ta type II toxinantitoxin system in E. coli, and toxin-antitoxin systems are generally considered as agents of the stress responses in prokaryotes[41]. Meanwhile, two AbrB regulators have been characterized in Synechocystis,Sll0359 is an activator of the expression of the operon hoxEFUYH that encodes the bidirectional Ni-Fe hydrogenase,while the deletion mutant of sll0822 gene was observed with significant decreases in growth rate and cellular pigment content, probably due to its role in deregulating partition of carbon between storage forms and soluble forms used for biosynthetic purposes[39,42,43].In addition,a CalA(AbrB1)protein,which belongs to the transcription regulator family AbrB, was involved in regulating iron superoxide dismutase, and a possible function against oxidative stress,in a nitrogen-fixing cyanobacterium Nostoc sp. PCC 7120 [44]. In addition,it and its isoenzymes CalB(AbrB1)regulate the formation of heterocysts by interacting with the promoter regions of two genes hetP and hepA required for heterocyst formation in Nostoc sp. PCC 7120 [45]. In the Synechocystis genome, two other genes(i.e., ssl1300, and ssr7040) were also annotated for encoding AbrBlike regulators; however, no functional information is currently available [39]. Up to now, little studies could be found to obtain more direct evidences for possible function of slr0724.Hernandez-Prieto and Futschik [46] developed CyanoEXpress: a web database for interactive exploration and visualisation of transcriptional response patterns in Synechocystis. CyanoEXpress currently comprises expression data for 3073 genes and 178 environmental and genetic perturbations obtained in 31 independent studies. From the CyanoEXpress database, slr0724 was found down-regulated under some oxidative stresses, while in another study, Ling et al. [47] investigated the cellular mechanisms of S.cerevisiae response to alkane biofuels using transcriptome analysis and found that 6 of 55 oxidative stress response genes were differently regulated under alkane stress, which indicates exogenous alkane cause oxidative stress.These results pointed to the possibility that the Slr0724 transcription regulator might be involved in nhexane stress response in Synechocystis.

Fig. 6. Growth of the wild type (WT), the ⊿slr0724/pJA0724 and ⊿slr0724/pJAC mutant under control and n-hexane stress conditions (0.55% (vol), n-hexane; 1.3% (vol)).(a) WT, ⊿slr0724/pJA0724 and ⊿slr0724/pJAC under control medium. (b) WT, ⊿slr0724/pJA0724 and ⊿slr0724/pJAC under medium with 0.55% (vol) n-hexane. The error bar represents the calculated standard deviation of the measurements of three biological replicates.

4. Conclusions

In this study, we determined the transcript-level responses to n-hexane stress in Synechocystis through RNA-seq transcriptome analysis. Further COG enrichment analysis showed that signal transduction systems were among the most regulated, which prompted us to screen a library of transcriptional regulators(TRs) mutants constructed previously for regulatory genes related to n-hexane tolerance. The efforts led to the identification of slr0724 gene encoding an HtaR suppressor protein that could improve the tolerance of Synechocystis against n-hexane toxicity.The mutant ⊿slr0724 tolerant to n-hexane toxicity could be a good chassis candidate for biofuels production in the future. This study also lay the foundation for better understanding the mechanism of tolerance to n-hexane in Synechocystis, which could contribute to the further engineering of n-hexane tolerance in cyanobacteria.

Author’s Contribution

Lei Chen and Weiwen Zhang conceived of the study. Shubin Li and Tao Sun carried out the transcriptomics, RT-PCR, mutant construction, and phenotypic analysis. Guangsheng Pei carried out bioinformatics analysis. Shubin Li, Tao Sun, Guangsheng Pei, Lei Chen and Weiwen Zhang drafted the manuscript. All authors read and approved the final manuscript.

Data Availability

Data will be made available on request.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was supported by grants from the National Key Research and Development Program of China (2020YFA0906800,2021YFA0909700, 2018YFA0903600 and 2019YFA0904600).

Supplementary Material

Supplementary material to this article can be found online at https://doi.org/10.1016/j.cjche.2022.11.015.