Hfq-dependent artificial regulatory small RNAs for targeted gene expression control in the nitrogen-fixing Pseudomonas stutzeri A1501

Abstract

Artificial trans-encoded small RNAs (atsRNAs), which repress translation by base pairing with target mRNAs, have proven to be powerful tools for regulation of gene expression for reconstructing cell factories and genetic circuits. Currently, rationally designed atsRNAs use mostly E. coli natural trans-acting sRNAs with an Hfq-binding structure and a Rho-independent terminator as scaffolds. However, owing to the structural differences in Hfq proteins between E. coli and other bacteria, the inhibitory efficacy of atsRNAs may be influenced by heterologous sRNA scaffolds, highlighting the necessity of screening native sRNA scaffolds in various bacteria to develop functional synthetic sRNAs. The root-associated bacterium Pseudomonas stutzeri A1501 is a well-studied strain regarding nitrogen fixation, yet post-transcriptional regulatory platforms for this species remain scarce. In this study, we describe the development of an atsRNA platform by employing the natural Hfq-dependent sRNA CrcZ scaffold from the nitrogen-fixing strain P. stutzeri A1501, which is conserved in the Pseudomonas strain. On the basis of the CrcZ scaffold, two atsRNAs targeting the master nitrogen-fixing regulatory complex encoded by the nifLA operon were developed and found to effectively suppress nifLA gene expression at the posttranscriptional level in P. stutzeri A1501. Further studies demonstrated that the 5′ untranslated region of target mRNAs and five Hfq-binding sites are optimal features for designing atsRNAs using CrcZ as a scaffold. The well-established effective design principle for synthetic sRNAs with the CrcZ sRNA scaffold will enable more rational and efficient engineering of synthetic gene networks in Pseudomonas strains.

1. Introduction

The synthetic small noncoding RNA (sRNA or ncRNA) expression platform is a rapid and efficient gene expression modification system that allows targeted control of gene expression at the translational level without direct modification of chromosomes [1,2]. Traditional knockout-based transcriptional control strategies can modulate a specific number of genes, but they are sometimes difficult to perform for genes that are essential for cell function [3,4]. To overcome the drawbacks of conventional genome engineering methods, a synthetic sRNA-based genome-wide knockdown strategy has been developed for Escherichia coli[[5], [6], [7], [8], [9], [10]]. This synthetic sRNA system was constructed on the basis of the abilities and characteristics of native trans-acting sRNAs in E. coli and is an advantageous tool for repression of target genes that are essential for cell function and cannot be knocked out or when fine downregulation is preferable to gene deletion[[11], [12], [13]].

As a class of posttranscriptional regulators, sRNAs play key roles in many physiological processes, such as in regulating cell homeostasis, regulating the stress response, and remodelling metabolism [14]. It has become increasingly evident that sRNAs play a significant and global role in bacterial gene regulation [15]. Bacterial regulatory sRNAs include mainly cis-encoded antisense RNAs and trans-encoded sRNAs [1]. In contrast to cis-encoded antisense sRNAs, trans-encoded sRNAs share only limited complementarity with their target mRNAs. They usually repress protein expression by binding to the 5′ untranslated region (UTR) of target mRNAs and most often occlude the ribosome-binding site (RBS), which is functionally analogous to eukaryotic miRNAs[[16], [17], [18]]. Additionally, most trans-acting sRNAs are Hfq dependent. The RNA chaperone Hfq may serve passively as a platform to allow sRNAs and mRNAs to sample potential complementarity, effectively increasing the local concentrations of sRNAs and mRNAs and stabilising the sRNA‒mRNA complex [8,19]. Thus, the Hfq-binding site is usually associated with the suppression activity of synthetic sRNAs. Most trans-acting sRNAs exhibit modularity-related characteristics, namely, target binding sequences and the ability to recruit Hfq-binding sequences, and a majority of synthetic sRNAs have been developed by using trans-acting sRNAs as scaffolds for the efficient regulation of gene expression in E. coli. However, cis-acting sRNAs cannot be used to design synthetic sRNAs because they must be inserted into the 5′ UTRs of chromosomal target genes to control the expression of the genes of interest.

In 2012, Kang et al. proposed and validated for the first time the application of sRNAs for biotechnological purposes [11]. Coleman et al. for the first time, constructed hybrid MicF sRNAs that inhibit target gene expression on the basis of the modularity of natural trans-acting sRNAs in gram-negative bacteria; however, no clear design principles for synthetic sRNAs, such as the base pairing sequences to be used, required degree of complementarity to RBS sequences and requirements for the coding regions of the targeted mRNAs, have been proposed [20]. Subsequently, Man et al. established principles for the semirational design of artificial trans-encoded sRNAs (atsRNAs) for specific gene silencing in E. coli; these sRNAs consist of an mRNA base pairing region, Hfq-binding site and typical Rho-independent terminator [2]. To fine-tune the expression of target genes, Sang Yup Lee and colleagues developed a general strategy to design atsRNAs that contain a native E. coli sRNA scaffold (MicC), a 24-nt targeting region within the translation initiation region (TIR), and a terminator [21]. These synthetic sRNA regulators with repression activities are created largely by replacing the base paring region of natural sRNAs and are frequently applied in metabolic engineering studies of E. coli. For example, the use of synthetic sRNAs constructed with MicC and SgrS as scaffolds improved the production of amino acids, fatty acids, 5-amino-levulinic acid and heterologous butanol in E. coli [6,[22], [23], [24]]. Synthetic sRNAs based on native E. coli sRNA scaffolds have also been applied for metabolic engineering in other bacteria [25]. However, as a crucial component of this well-developed and widely used Hfq-sRNA regulatory tool derived from E. coli, the Hfq protein, while conserved in a wide range of bacteria, varies in length and sequence among species. This diversity is well illustrated by the phylogenetically distinct Hfq of Pseudomonas stutzeri A1501, a divergence likely driven by niche-specific adaptations such as its nitrogen-fixing lifestyle (Fig. S1). Such host-specific variability therefore poses a key limitation for adapting current synthetic sRNA tools to diverse bacteria [26]. To overcome this barriers, it may be necessary to screen for native scaffolds that exhibit high affinity for Hfq in target bacteria and, based on such insights, design synthetic sRNAs for optimal efficacy [27].

Here, we report the development of a plasmid-based synthetic sRNA platform comprising a native Hfq-dependent sRNA (CrcZ) scaffold and Hfq from the nitrogen-fixing strain P. stutzeri A1501. Using this platform, we constructed two synthetic sRNAs, designated AnsR-O5 and AnsR-U5, to target the nifLA operon, which encodes the master regulatory system for nitrogen fixation in P. stutzeri. The expression of these two synthetic sRNAs significantly suppressed the nitrogen fixation capacity of P. stutzeri A1501 by repressing nifLA expression. Compared with AnsR-O5, which targets a 20-nt region of the nifLA coding sequence (CDS) encompassing the AUG codon and the following 17 bases, AnsR-U5 targets the 5′ UTR sequence from the putative ribosome-binding site to the AUG codon and exhibits greater repression efficiency. This suggests that base-pairing within the 5′ UTR may efficiently enhance synthetic sRNA-mediated repression of nifLA. Furthermore, we generated five truncated AnsR-U5 variants carrying different numbers of Hfq-binding sites. Among these, AnsR-U5 (with five Hfq-binding sites) showed a 2-fold higher repression efficiency than AnsR-U1 (with one Hfq-binding site), indicating that the number of Hfq-binding sites significantly influences the silencing ability of artificial sRNAs.

2. Materials and methods

2.1. Bacterial strains, plasmids, and growth conditions

A summary of the bacterial strains and plasmids used in this study is provided in Table 1. P. stutzeri A1501 and its derivatives were grown at 30 °C in Luria–Bertani (LB) medium or in modified minimal lactate-containing medium (medium K). When necessary, chloramphenicol (Cm), tetracycline (Tc), ampicillin (Amp), kanamycin (Km) and hygromycin (Hyg) were added at final concentrations of 34 μg mL⁻¹, 50 μg mL⁻¹, 50 μg mL⁻¹, 50 μg mL⁻¹, and 50 μg mL⁻¹, respectively. The absorbance (OD₆₀₀) was measured with a U-3010 (Hitachi) spectrophotometer.

Table 1.

Strains and plasmids used in this study.

Strain/plasmid	Relevant characteristics	Source
Plasmids
pRK2013	Helper plasmid for conjugation into P. stutzeri A1501, Kmr	Lab collection
pLAFR3	Broad-host-range plasmid, Tcr	Lab collection
pBBR1MCS2	Cloning vector, Kmr	Lab collection
pUX-BF13	Helper plasmid for transposition of mini-Tn7 element, Ampr	Lab collection
pXY2	Mini-Tn7 vector for a lacZ translational fusion, Ampr and Gmr	Lab collection
pXY2-PnifL	pXY2 containing the PnifLA⸬lacZ fusion, Ampr and Gmr	This study
pBBR-crcZ	pBBR1MCS2 derivative carrying the CrcZ under the control of CrcZ promoter, Kmr	This study
pBBR-AnsR-U5	pBBR1MCS2 derivative carrying the AnsR-U5 under the control of CrcZ promoter, Kmr	This study
pBBR-AnsR-O5	pBBR1MCS2 derivative carrying the AnsR-O5 under the control of CrcZ promoter, Kmr	This study
pLnifA	pLAFR3 derivative carrying the wild type nifA gene, Tcr	This study
pBBR-AnsR-U5m1	pBBR1MCS2 derivative carrying the mutated AnsR-U5 with 50% mutation of the base number within the pairing site region, Kmr	This study
pBBR-AnsR-U5m2	pBBR1MCS2 derivative carrying the mutated AnsR-U5 with 75% mutation of the base number within the pairing site region, Kmr	This study
pBBR-AnsR-U0	pBBR1MCS2 derivative carrying the mutated AnsR-U5 without Hfq binding site, Kmr	This study
pBBR-AnsR-U1	pBBR1MCS2 derivative carrying the mutated AnsR-U5 with one Hfq binding site, Kmr	This study
pBBR-AnsR-U2	pBBR1MCS2 derivative carrying the mutated AnsR-U5 with two Hfq binding sites, Kmr	This study
pBBR-AnsR-U3	pBBR1MCS2 derivative carrying the mutated AnsR-U5 with three Hfq binding sites, Kmr	This study
pBBR-AnsR-U4	pBBR1MCS2 derivative carrying the mutated AnsR-U5 with four Hfq binding sites, Kmr	This study
pET30a-Hfq	Expression vector of Hfq protein, Kmr	Lab collection
P. stutzeri strains
A1501	Wild type, Chinese Culture Collection CGMCC 0351	Lab collection
Δhfq	Hfq-Cm and Hyg deletion mutant, Cmr and Hygr	Lab collection
A1578	A1501 containing plasmid pBBR1MCS2, Kmr	This study
A1825	A1501 containing pBBR-AnsR-O5, Kmr	This study
A1736	A1725 containing pLnifA, Kmr and Tcr	This study
A1737	A1825 containing pLnifA, Kmr and Tcr	This study
A1720	A1501 containing pBBR-AnsR-U0, Kmr	This study
A1721	A1501 containing pBBR-AnsR-U1, Kmr	This study
A1722	A1501 containing pBBR-AnsR-U2, Kmr	This study
A1723	A1501 containing pBBR-AnsR-U3, Kmr	This study
A1724	A1501 containing pBBR-AnsR-U4, Kmr	This study
A1725	A1501 containing pBBR-AnsR-U5, Kmr	This study
A1726	A1501 containing pBBR-AnsR-U5m1, Kmr	This study
A1727	A1501 containing pBBR-AnsR-U5m2, Kmr	This study
A1579	A1501 containing pBBR-crcZ, Kmr	This study
A1580	A1501 containing pXY2-PnifL, Gmr	This study
A1585	A1580 containing pBBR-AnsR-U5, Gmr and Kmr	This study
A1586	A1580 containing pBBR-AnsR-U5m2, Gmr and Kmr	This study
A1732	A1721 containing pLnifA, Kmr and Tcr	This study
A1733	A1722 containing pLnifA, Kmr and Tcr	This study
A1734	A1723 containing pLnifA, Kmr and Tcr	This study
A1735	A1724 containing pLnifA, Kmr and Tcr	This study
A1577	A1501 containing pLAFR3	[28]
A1421	Δhfq containing pBBR-AnsR-U1, Kmr and Hygr	This study
A1422	Δhfq containing pBBR-AnsR-U2, Kmr and Hygr	This study
A1423	Δhfq containing pBBR-AnsR-U3, Kmr and Hygr	This study
A1424	Δhfq containing pBBR-AnsR-U4, Kmr and Hygr	This study
A1425	Δhfq containing pBBR-AnsR-U5, Kmr and Hygr	This study

2.2. Construction of a complementation plasmid carrying the nifA gene and various artificial sRNA expression vectors for RNAi

A 1614-bp DNA fragment containing the wild-type nifA gene along with its promoter and terminator regions was amplified by PCR. The fragment was double-digested with Hind III/Bam HI and then ligated into pLAFR3 to yield the plasmid pLnifA. The plasmid carrying the nifA gene was subsequently introduced into the constructed synthetic sRNA expression strains by triparental mating to generate the corresponding nifA-complemented expression strains (Table 1).

To construct the anti-nifLA synthetic sRNA-encoding plasmid, the artificial sRNA DNA fragments generated using CrcZ as a scaffold with the native nitrogen-fixation-related inducible promoter of CrcZ were synthesised by BGI Co., Ltd. Restriction enzyme sites (Hind III and Bam HI) were incorporated into the synthesised fragments, which were subsequently cloned and inserted into the pBBR1MCS2 vector, resulting in the following plasmids: pBBR-AnsR-U5 and pBBR-AnsR-O5, carrying AnsR targeting the nifLA gene; pBBR-AnsR-U4, carrying the truncated AnsR-U5 with four Hfq-binding sites; pBBR-AnsR-U3, carrying the truncated AnsR-U5 with three Hfq-binding sites; pBBR-AnsR-U2, carrying the truncated AnsR-U5 with two Hfq-binding sites; pBBR-AnsR-U1, carrying the truncated AnsR-U5 with one Hfq-binding site; pBBR-AnsR-U0, carrying the truncated AnsR-U5 without Hfq-binding site; pBBR-AnsR-U5m1, carrying a mutated AnsR-U5 showing a 50% match with the targeting sequence of nifLA mRNA; pBBR-AnsR-U5m2, carrying a mutated AnsR-U5 showing a 25% match with the targeting sequence of nifLA mRNA.

2.3. RNA Isolation for qRT-PCR

Total RNA was isolated from bacteria cultured under the described conditions using the innuPREP RNA Mini Kit 2.0 (Analytikjena, Germany) according to the manufacturer's instructions. Total RNA was quantified via microspectrophotometry (NanoDrop Technologies, Inc.). RNA integrity was measured via an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.). RNA samples with RNA integrity numbers (RINs) above 7.0 and threshold cycle (CT) values above 32 were used for qRT‒PCR.

2.4. Quantitative real-time PCR

The expression levels of selected genes were determined via qRT‒PCR with Power SYBR Green PCR master mix via an ABI Prism 7500 sequence detection system (Applied Biosystems, USA) according to the manufacturer's instructions. Primers were designed on the basis of sequences of selected genes, which were imported into OligoPerfect (Invitrogen), a primer design software program designed to generate primer pairs suitable for real-time PCR. The primers used for qRT‒PCR are listed in Table S1. For quantification of gene expression, total RNA was reverse transcribed using random primers and the High Capacity cDNA Transcription Kit (Applied Biosystems) according to the manufacturer's instructions. All qRT‒PCR assays were performed in triplicate using a 20 μL mixture containing cDNA, 1 × Brilliant SYBR Green Quantitative PCR Master Mix (Stratagene), and each primer (approximately 5 pmol). Amplification and detection of specific products were performed via the following procedure: 95 °C for 10 min, followed by 40 cycles of 95 °C for 30 s, 55 °C for 1 min, and 72 °C for 30 s and then by a dissociation curve analysis. The 16S rRNA gene was used as the endogenous reference control, and relative gene expression was determined via the 2^−ΔΔCT relative quantification method. To obtain a standard curve for real-time PCR (RT‒PCR), PCR was performed with each primer set by using calibrated amounts of chromosomal DNA; these reactions were performed at the same time as the qRT‒PCR. The amplification efficiencies (E values) of the primer pairs used in the study were calculated from the standard curve obtained from a five-point 10-fold serial dilution series of the cDNA template according to the following equation: E (%) = [10^(−1/slope) - 1] × 100. All the experiments were performed with three biological replicates.

2.5. Nitrogenase activity assays

Nitrogenase activity was determined with bacterial suspensions incubated to an OD₆₀₀ of 0.1 in N-free minimal medium (1.0% oxygen and 10% acetylene) at 30 °C [29]. Protein concentrations were determined via a standard protein assay (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard. The specific activity of nitrogenase is expressed as nmol of ethylene per hour per milligram of protein. Each experiment was repeated at least three times.

2.6. Construction of translational lacZ fusions and ß-galactosidase assay

To test the translation initiation repression function of AnsR-U5 through base pairing with the 5′ UTR sequence of nifLA mRNA, the DNA fragment carrying the nifLA promoter was amplified via the primers PnifL-F and PnifL-R (Table S1) and subsequently cloned into the plasmid pXY2, a modified version of pUC18-mini-Tn7-Gm-lacZ [30]. The resulting recombinant plasmid, pXY2-PnifL, was electroporated into P. stutzeri A1501 together with pUX-BF13, followed by integration of the mini-Tn7 element carrying the lacZ reporter fusion into the unique Tn7 site downstream of glmS [31].

For the measurement of β-galactosidase activity in P. stutzeri strains under nitrogen fixation conditions, cells from overnight cultures were collected via centrifugation and washed twice with nitrogen-free medium K containing 50 mM lactate. The OD₆₀₀ of the cell suspensions was subsequently adjusted to 0.1 with nitrogen-free medium K containing 50 mM lactate. Ten millilitres of the cell suspension were transferred to a 60 mL flask. The cell suspension was incubated for 6 h at 30 °C with shaking under nitrogen fixation conditions with 1.0% oxygen. After incubation, the cells were collected via centrifugation and resuspended in 1.0 mL of Z-buffer in 2.0 mL EP tubes. Two to three drops of chloroform were added to the tubes and mixed by inverting several times to lyse the cells. The tubes were placed with the tops open in an incubator at 30 °C for 40 min. Two hundred microliters of ONPG solution (4 mg/mL) was added to the tubes, and the time of colour change was recorded. The reaction was stopped with 500 μL of 1 M Na₂CO₃ solution. Then, 150 μL of the reaction mixture was added to a 96-well microplate. The OD₄₂₀ and OD₅₅₀ values were measured via a microplate reader. β-Galactosidase activity is expressed as Miller units [32].

2.7. Measuring the stability of the synthetic sRNAs

To assess the effect of Hfq protein on synthetic sRNA stability, 2 mL bacterial samples incubated for 6 h under nitrogen fixation conditions were collected at different times immediately after the addition of rifampin (400 mg/mL). Two volumes of RNAprotect (Sigma, USA) were added to each sample to prevent RNA degradation, and the samples were immediately frozen in liquid nitrogen. Total RNA was isolated from each sample for use in estimating mRNA levels via qRT‒PCR. The primers used for measurement of the half-life of the selected gene mRNAs are listed in Table S1 in the supplemental material, and qRT‒PCR was performed as described above. The data are presented as percent mRNA levels relative to time point zero.

2.8. Microscale thermophoresis measurements

Microscale thermophoresis (MST) experiments were performed according to Zhan et al. [33]. A modified overlap extension PCR method was used to amplify the full-length AnsR-O5, AnsR-U5 and its mutants (N-AnsR-U5m) via mutagenic primers. A set of full-length sRNAs (N-AnsR probes) for MST was then transcribed by GenePharma using a MAXIscript kit (Thermo Fisher) from PCR-generated templates (Table S1). In addition, another set of Cy3-labelled 70-nt single-stranded RNA (ssRNA) oligonucleotides containing different base pairing regions of nifL mRNA (N-nifL competitors) were synthesised by GenePharma. For the measurements of RNA‒RNA interactions, 4 μL of sample containing 200 nM labelled probe and increasing concentrations of a nonlabelled competitor (from 5 nM to 150 nM) were loaded on standard treated silicon capillaries (Monolith NT.115 series capillaries; catalogue no. MO-K002). Data analyses were performed via Nanotemper Analysis software (Nano-Temper Technologies). The purified Hfq protein was labelled with the Monolith NTTM Protein Labeling Kit RED-NHS according to the manufacturer's instructions (No. MO-L011, NanoTemper Technologies). For analysis of the interactions of Hfq and RNA, the concentration of the labelled Hfq (20 nM) was kept constant, whereas the concentrations of nonlabelled ssRNA oligonucleotides varied from 0.3 μM to 10 μM. The measurements were carried out via a Monolith NT.115 instrument (NanoTemper Technologies GmbH) at 26 °C in diethyl pyrocarbonate (DEPC)-treated water with 40% excitation power and medium MST-Power. The dissociation constants (Kd values) were calculated as described previously.

2.9. Expression and purification of hfq for MST measurements

The expression plasmid for the Hfq protein, pET31a-Hfq, was constructed previously in the laboratory [34]. The plasmid pET31a-Hfq was incubated in LB medium at 37 °C, and the Hfq protein was expressed and purified via the Strep-Tactin system (IBA Life Sciences). The protein concentration was determined via a Bio-Rad protein assay reagent kit (Quick Start Bradford, Bio-Rad).

3. Results

3.1. Selection of a native trans-acting sRNA as the scaffold for synthetic sRNA design in P. stutzeri A1501

In previous studies, most synthetic sRNAs were engineered by fusing a target-specific sequence to the MicC scaffold. This sRNA was chosen from 101 native E. coli sRNAs [22], but no MicC homologous sRNAs were found in P. stutzeri A1501. A search for sRNAs in the genome of P. stutzeri A1501 was previously performed, and CrcZ, which is inducible under nitrogen fixation conditions (Fig. 1A) and has a size of 340 nt, was identified [33]. CrcZ exhibits high affinity for the Hfq protein and its stability is regulated by the Hfq protein, reflecting a universal expression characteristic of trans-acting sRNAs [35].

Fig. 1.

The expression characteristics and secondary structure of CrcZ. (A) Quantitative RT‒PCR analysis of the relative expression levels of the natural ncRNA CrcZ under nitrogen fixation conditions in P. stutzeri A1501. The statistical significance of the difference was confirmed by t-tests. ∗∗∗: p ≤ 0.001. (B) The secondary structure of the natural ncRNA CrcZ. The AAN sequences predicted to bind with the Hfq protein are highlighted in red.

Further bioinformatics analysis revealed that CrcZ contains the poly-U tail of the Rho-independent terminator, which is essential for Hfq action, and its secondary structure, which contains five CA-rich motifs, represents a natural sRNA with the greatest number of Hfq-binding sites known at present (Fig. 1B). These characteristics, particularly its Hfq-dependent stability, meet critical design criteria for a synthetic sRNA scaffold, as they ensure that engineered sRNAs will benefit from Hfq-mediated stabilisation. These attributes led to its selection as an endogenous scaffold for constructing artificial sRNAs.

3.2. Design and characterisation of synthetic sRNAs targeting the nifLA operon in P. stutzeri A1501

P. stutzeri A1501 is the model strain for associative nitrogen-fixing bacteria. To evaluate the artificial non-coding RNA system, we choose the nitrogen fixation regulatory gene nifLA as the target. In P. stutzeri A1501, nifL and nifA are clustered within the same operon. NifA acts as the central positive regulator of nitrogen fixation, directly controlling the expression of the nitrogenase-encoding genes nifHDK [36]. Consequently, the repression of the nifLA operon by synthetic sRNAs inhibits NifA activity, leading to the systematic down-regulation of the entire nitrogen fixation gene network and a consequent significant decline in nitrogenase activity. In bacteria, most trans-acting sRNAs impair translation and/or mRNA stability through imperfect base pairing with the 5′ UTR or translation initiation region of target mRNAs. To determine the optimal base pairing region of anti-nifLA synthetic sRNAs, we used two target-binding sequences that were complementary to the 5′ UTR (that is, the sequence from the putative RBS to the AUG codon, 20 nt in total) and to the CDS (that is, the AUG and the following 17 bases, 20 nt in total). The two target-binding sequences were introduced upstream of the CrcZ scaffold to create two anti-nifLA synthetic sRNAs (artificial nitrogenase activity-silencing non-coding RNA, AnsR), namely, AnsR-U5 and AnsR-O5 (Fig. 2A). The secondary structures of AnsR-U5 and AnsR-O5 were predicted on the basis of minimum free energy via the program RNAstructure 5.1 (http://rna.urmc.rochester.edu/RNAstructure.html). The folded structures of AnsR-U5 and AnsR-O5 exhibited multiple stem loops; these sRNAs were highly structured.

Fig. 2.

Design and characterisation of two atsRNAs, AnsR-U5 and AnsR-O5, which target the nifLA operon. (A) Schematic design of two anti-nifLA synthetic sRNAs, AnsR-U5 and AnsR-O5, in P. stutzeri A1501. The putative RBS of nifLA mRNA is highlighted in blue. Two target-binding sequence regions of nifLA mRNA encompassing the sequence from the putative RBS to the AUG (20 nt in total) or the AUG and the following 17 bases (20 nt in total) were chosen and introduced upstream of the CrcZ scaffold, yielding two atsRNAs, namely, AnsR-U5 and AnsR-O5. (B) Quantitative RT‒PCR analysis of the expression levels of AnsR-U5 expression strain A1725 (A1501 harbouring pBBR-AnsR-U5) and the AnsR-O5 expression strain A1825 (A1501 harbouring pBBR-AnsR-O5) under nitrogen fixation conditions. One-way ANOVA was applied for significance analysis. ∗: p ≤ 0.05, ∗∗∗: p ≤ 0.001. (C, D) Growth curves of the wild-type strain A1501, the empty-vector control strain A1578 (A1501 harbouring pBBRMCS-2), the AnsR-U5 expression strain A1725 (A1501 harbouring pBBR-AnsR-U5), and the AnsR-O5 expression strain A1825 (A1501 harbouring pBBR-AnsR-O5) in medium K or LB medium.

To construct an inducible customised synthetic sRNA expression plasmid functional in P. stutzeri A1501 under nitrogen fixation conditions, AnsR-U5 and AnsR-O5 were cloned and inserted into the pBBR1MCS-2 plasmid, yielding pBBR-AnsR-U5 and pBBR-AnsR-O5. In this plasmid, the expression of AnsR-U5 and AnsR-O5 is driven from the native nitrogen fixation-related inducible promoter of CrcZ. We analysed the transcription levels of the two atsRNAs in expression strains (A1725 and A1825) under both nitrogen fixation and non-nitrogen fixation conditions (Fig. 2B) via quantitative real-time PCR (qRT‒PCR). The results revealed that the transcription of both AnsR-U5 and AnsR-O5 was induced (1.26-fold and 1.51-fold, respectively) under nitrogen fixation conditions, suggesting that AnsR-U5 and AnsR-O5 are indeed expressed in P. stutzeri strains.

To further examine whether the expression of AnsR-U5 or AnsR-O5 affects the growth of P. stutzeri strains, we compared the growth of the AnsR-U5 and AnsR-O5 expression strains (A1725 and A1825) with that of the wild-type strain A1501 in modified minimal lactate-containing medium (medium K) and Luria–Bertani (LB) medium. No difference in growth was observed among the wild-type strain A1501, the empty-vector control strain A1578 (A1501 harboring pBBR1MCS-2), and the two atsRNA expression strains (A1725 and A1825) in both medium K (Fig. 2C) and LB medium (Fig. 2D), suggesting that expression of AnsR-U5 and AnsR-O5 did not affect cell growth.

3.3. Synthetic sRNAs AnsR-U5 and AnsR-O5 effectively repress nitrogen fixation in P. stutzeri A1501

Because the nitrogen fixation ability of P. stutzeri A1501 is regulated by the nifLA operon, nitrogenase activity was assayed to assess the inhibition efficiency of AnsR-U5 and AnsR-O5 under nitrogen fixation conditions (nitrogen-free and microaerobic conditions). As shown in Fig. 3A, the expression of both AnsR-U5 and AnsR-O5 resulted in a significant reduction in nitrogenase activity. Specifically, AnsR-U5 expression strain A1725 caused a dramatic suppression of nitrogenase activity, reducing it to approximately 22% of the activity of the wild-type strain A1501, even though the CrcZ backbone itself has only a minor impact on wild-type enzyme activity. Quantitative real-time PCR (qRT‒PCR) revealed that expression of both AnsR-U5 and AnsR-O5 significantly decreased the expression levels of nifLA and nifHDK to varying degrees compared with those in wild-type A1501, and it was inferred that the two anti-nifLA synthetic sRNAs mediated the reduction in nitrogenase activity by inhibiting nifA expression (Fig. 3B). To test this hypothesis, we constructed two nifA-complemented expression strains derived from the AnsR-U5 and AnsR-O5 expression strains (designated A1736 and A1737, respectively) and assayed nitrogenase activity in these strains. The results show that introduction of the nifA gene restored nitrogenase activity to the wild-type level under nitrogen fixation conditions (Fig. 3C). These data indicate that the expression of AnsR-U5 and AnsR-O5 decreased NifA protein levels, as indicated by the reduced nitrogenase activity.

Fig. 3.

Inhibitory effects of two atsRNAs, AnsR-U5 and AnsR-O5, on the nitrogen fixation ability of P. stutzeri A1501. (A) Nitrogenase activity in the wild-type and the ncRNA-expressing strains. WT: wild-type strain A1501; WT-CrcZ: CrcZ scaffold control strain A1579 (A1501 harbouring pBBR-crcZ); WT-AnsR-U5: AnsR-U5 expression strain A1725 (A1501 harbouring pBBR-AnsR-U5); WT-AnsR-O5: AnsR-O5 expression strain A1825 (A1501 harbouring pBBR-AnsR-O5). (B) Quantitative RT‒PCR analysis of the inhibitory effects of AnsR-U5 and AnsR-O5 on the expression of nif genes. WT: wild-type strain A1501; WT-AnsR-U5: AnsR-U5 expression strain A1725 (A1501 harbouring pBBR-AnsR-U5); WT-AnsR-O5: AnsR-O5 expression strain A1825 (A1501 harbouring pBBR-AnsR-O5). Differences were considered significant at p ≤ 0.05. (C) Nitrogenase activity in the wild-type and the nifA-complemented expression strains. WT: wild-type strain A1501; WT-AnsR-U5-nifA: nifA-complemented expression strain A1736 (AnsR-U5 expression strain A1725 harbouring pLnifA); WT-AnsR-O5-nifA: nifA-complemented expression strain A1737 (AnsR-O5 expression strain A1825 harbouring pLnifA). One-way ANOVA was applied for significance analysis. The experiments were performed in triplicate, and the mean values are shown. ∗: p ≤ 0.05, ∗∗: p ≤ 0.01 and ∗∗∗: p ≤ 0.001.

3.4. Synthetic sRNAs AnsR-U5 and AnsR-O5 translationally repress nif gene expression in P. stutzeri A1501

Like natural sRNAs, synthetic sRNAs also exert their inhibitory function by binding the target sequence. To examine the binding efficiency of AnsR-U5 and AnsR-O5 with the corresponding base pairing sequence of nifLA mRNA, microscale thermophoresis (MST) was used to determine the dissociation constant (Kd) in vitro using a set of 379-nt full-length AnsR-U5 (N-AnsR-U5) and 70-nt nifLA mRNA oligonucleotides containing the sequence from the RBS to the AUG (N-nifL mRNA1) or 379-nt full-length AnsR-O5 (N-AnsR-O5) and 70-nt nifLA mRNA oligonucleotides containing the AUG and the following 17 bases (N-nifL mRNA). As expected from the MST experiments, both AnsR-U5 and AnsR-O5 notably interacted with the corresponding nifLA mRNA, with Kd values of 13 ± 1 and 29 ± 0.63 nM, respectively (Fig. 4A and B).

Fig. 4.

AnsR-U5 and AnsR-O5 decreased the nitrogen fixation capacity of P. stutzeri A1501 by efficiently binding to nifLA mRNA. (A, B) Determination of the binding affinity of labelled AnsR-U5 or AnsR-O5 for nifLA mRNA by microscale thermophoresis. (C) Schematic representation of point mutations in AnsR-U5 variants (AnsR-U5m1 and AnsR-U5m2). The base pairing sequences of nifLA mRNA and AnsR are shown in black. Point mutations introduced into synthesised oligonucleotide derivatives are shown in red. (D, E) Determination of the binding affinity of labelled AnsR-U5 mutants for nifLA mRNA via microscale thermophoresis. (F) Nitrogenase activity in the wild-type, AnsR-U5-expressing, and point mutant strains. WT: wild-type strain A1501; WT-AnsR-U5: AnsR-U5 expression strain A1725 (A1501 harbouring pBBR-AnsR-U5); WT-AnsR-U5m1: AnsR-U5 point mutant strain A1726 (A1501 harbouring pBBR-AnsR-U5m1); WT-AnsR-U5m2: AnsR-U5 point mutant strain A1727 (A1501 harbouring pBBR-AnsR-U5m2); WT-CrcZ: CrcZ scaffold control strain A1579 (A1501 harbouring pBBR-crcZ) served as a control. The experiments were performed in triplicate, and the mean values are shown. One-way ANOVA was applied for significance analysis. ∗∗: p ≤ 0.01, ∗∗∗: p ≤ 0.001.

As mentioned above, AnsR-U5, targeting the 5′ UTR of nifLA mRNA and reducing nitrogenase activity to 22%, exhibited greater repression activity than AnsR-O5, indicating that the target-binding sequence region of nifLA mRNA encompassing the sequence from the RBS to the AUG was the optimal region. To further validate the role of the base pairing sequences for AnsR-U5-mediated regulation, the effects of AnsR-U5 mutation on base pairing with nifLA mRNA were investigated via two site-directed mutagenesis-derived mutants of AnsR-U5 (AnsR-U5m1 and AnsR-U5m2) (Fig. 4C). The data indicated that N-AnsR-U5m1, which contains a point mutation within a stem-loop region of AnsR-U5 that reduces its base-pairing complementarity to the target to 50%, exhibited weaker affinity for nifLA mRNA than did N-AnsR-U5 (100% complementarity), with dissociation constants of 56.6 ± 15.4 nM (Fig. 4D). Notably, no obvious binding signal was detected with N-AnsR-U5m2 (25% match) (Fig. 4E). Furthermore, we assessed their inhibitory effects on nitrogenase activity and found that the two-point mutants of AnsR-U5 exhibited weaker nitrogenase inhibition activity than the AnsR-U5 expression strain did. Notably, the inhibitory effect of AnsR-U5m2 (25% match) expression strain (A1727) was not significantly different from that of the CrcZ scaffold expression strain A1579 (Fig. 4F).

To further demonstrate that AnsR-U5 inhibits nifLA mRNA translation initiation in a sequence-specific manner, we constructed a translational PnifL⸬lacZ fusion in the wild-type A1501 background, generating strain A1580 (Fig. 5A). This strain carries the complete, native 5′ UTR of nifLA mRNA but does not express the artificial sRNA AnsR-U5, thus serving to establish the basal expression level. Subsequently, A1580 was transformed with either the AnsR-U5 expression plasmid (pBBR-AnsR-U5) or the mutant derivative (pBBR-AnsR-U5m2), yielding strains A1585 and A1586, respectively. Under nitrogen-fixing conditions, β-galactosidase assays revealed that lacZ expression in the PnifL⸬lacZ/AnsR-U5 co-expression strain A1585 was reduced to 27% of the level observed in the control strain A1580 (a 73% reduction), whereas the co-expression strain (A1586) carrying the AnsR-U5m2, with a mutation in the anti-nifLA-binding site, exhibited a weaker inhibitory effect (Fig. 5B). These results suggest that AnsR-U5 represses lacZ expression via base pairing with the 5′ UTR of nifLA mRNA at the post-transcriptional level, further supporting its role in modulating nifLA mRNA translation initiation.

Fig. 5.

Inhibitory effect of AnsR-U5 on the β-galactosidase activity of the PnifL::lacZ fusion under nitrogen fixation conditions. (A) Schematic representation of PnifL::lacZ. The 5′ UTR of nifLA, which pairs with AnsR-U5, was cloned upstream of lacZ in pXY2. The PnifL::lacZ translational fusion was integrated into a single highly conserved location positioned downstream of glmS on the chromosome of A1501, yielding strain A1580. The AnsR-U5 expression plasmid, pBBR-AnsR-U5, was introduced into A1580, yielding strain A1585. The AnsR-U5m2 expression plasmid, pBBR-AnsR-U5m2, was introduced into A1580, yielding strain A1586. The β-galactosidase activity represents the posttranscriptional expression of the reporter gene lacZ driven by the nifLA promoter. (B) β-Galactosidase activity of PnifL::lacZ translational fusion in the wild-type, AnsR-U5 expression, and mutant strains. WT-PnifL::lacZ: the control strain A1580 (A1501 harbouring pXY2-PnifL); WT-PnifL::lacZ-AnsR-U5: PnifL::lacZ/AnsR-U5 co-expression strain A1585 (A1580 harbouring pBBR-AnsR-U5); WT-PnifL::lacZ-AnsR-U5m2: PnifL::lacZ/AnsR-U5m2 co-expression strain A1586 (A1580 harbouring pBBR-AnsR-U5m2). The experiments were performed in triplicate, and the mean values are shown. The statistical significance of the difference was confirmed by one-way ANOVA. ∗∗∗: p ≤ 0.001.

3.5. Silencing efficiency of synthetic sRNAs scales linearly with the hfq-binding site count

Hfq binding is a crucial component of the atsRNA design strategy [2,8]; however, the effect of the number of Hfq-binding sites on the inhibitory activity of atsRNAs is still unclear. To further verify the involvement of the Hfq protein in AnsR-mediated gene repression, we constructed expression vectors for five truncated AnsR-U5 variants containing different numbers of Hfq-binding sites, while ensuring that all variants maintained their normal secondary structure and the same anti-nifLA binding sites. These vectors were: pBBR-AnsR-U0 (expressing a truncated AnsR-U5 without Hfq-binding site), pBBR-AnsR-U4 (with four sites), pBBR-AnsR-U3 (with three sites), pBBR-AnsR-U2 (with two sites), and pBBR-AnsR-U1 (with one site). We analysed the growth of strains harbouring these plasmids in LB medium or medium K and compared their nitrogenase activities with that of the wild-type strain A1501 under nitrogen fixation conditions. The results showed no growth difference between any of the truncated variant-expressing strains and wild-type A1501 in either medium (data not shown). However, in contrast to the AnsR-U5 expression strain A1725, which contains five Hfq-binding sites and exhibits approximately 28% of wild-type nitrogenase activity, the truncated variants showed 43–60% of the wild-type level (Fig. 6A). This result indicates that the ability of AnsR-U5 to inhibit nitrogenase activity in wild-type A1501 significantly decreases as the number of Hfq-binding sites is reduced. Furthermore, we performed complementation by introducing the nifA gene in these truncated AnsR-U5-expressing strains and assaying nitrogenase activity. The result showed that the expression of the nifA gene restored the nitrogenase activity of these truncated AnsR-U5-expressing strains to the wild-type level under nitrogen fixation conditions, whereas the empty-vector control strain A1577 (A1501 harboring pLAFR3) exhibited no significant alteration in activity (Fig. 6B).

Fig. 6.

Effect of Hfq on the repression capacity of AnsR-U5. (A) Nitrogenase activity in the wild-type, AnsR-U5 expression strains, and truncated AnsR-U5 variants. WT: wild-type strain A1501; WT-AnsR-U0: AnsR-U5 truncated strain A1720 (A1501 harbouring pBBR-AnsR-U0); WT-AnsR-U1: AnsR-U5 truncated strain A1721 (A1501 harbouring pBBR-AnsR-U1); WT-AnsR-U2: AnsR-U5 truncated strain A1722 (A1501 harbouring pBBR-AnsR-U2); WT-AnsR-U3: AnsR-U5 truncated strain A1723 (A1501 harbouring pBBR-AnsR-U3); WT-AnsR-U4: AnsR-U5 truncated strain A1724 (A1501 harbouring pBBR-AnsR-U4); WT-AnsR-U5: AnsR-U5 expression strain A1725 (A1501 harbouring pBBR-AnsR-U5); WT-pBBR: empty-vector control strain A1578 (A1501 harbouring pBBR1MCS2). (B) Nitrogenase activity in WT A1501 and the nifA-complemented strains of mutant AnsR-U5 variants. WT- pLAFR3: empty-vector control strain A1577 (A1501 harbouring pLAFR3); WT-AnsR-U1-nifA: nifA-complemented expression strain A1732 (truncated AnsR-U5 expression strain A1721 harbouring pLnifA); WT-AnsR-U2-nifA: nifA-complemented expression strain A1733 (truncated AnsR-U5 expression strain A1722 harbouring pLnifA); WT-AnsR-U3-nifA: nifA-complemented expression strain A1734 (truncated AnsR-U5 expression strain A1723 harbouring pLnifA); WT-AnsR-U4-nifA: nifA-complemented expression strain A1735 (truncated AnsR-U5 expression strain A1724 harbouring pLnifA); WT-AnsR-U5-nifA: nifA-complemented expression strain A1736 (AnsR-U5 expression strain A1725 harbouring pLnifA). (C, D) Analysis of the half-lives of AnsR-U1, AnsR-U2, AnsR-U3, AnsR-U4 and AnsR-U5 in P. stutzeri A1501 under nitrogen fixation conditions. (E) Analysis of the half-lives of the AnsR-U1, AnsR-U2, AnsR-U3, AnsR-U4 and AnsR-U5 transcripts in the hfq mutant strain (Δhfq) under nitrogen fixation conditions. (F–H) The binding affinities of AnsR-U3, AnsR-U4, and AnsR-U5 for labelled Hfq determined via microscale thermophoresis. Error bars represent the SD, and one-way ANOVA was applied for significance analysis. ∗: p ≤ 0.05, ∗∗: p ≤ 0.01, ∗∗∗: p ≤ 0.001 and ns: non-significant.

Most trans-acting sRNAs require the RNA chaperone Hfq to form stable sRNA-target‒mRNA hybrids or to maintain their own stability [37]. To test whether this function holds for atsRNA AnsR-U5, we examined the half-life of AnsR-U5 and these truncated AnsR-U5 transcripts in wild-type strain A1501 and the hfq mutant Δhfq. qRT‒PCR was performed on cells grown under nitrogen fixation conditions. Rifampin was added to inhibit RNA synthesis. Consistent with expectation, reducing the number of Hfq-binding sites resulted in shorter half-lives for the atsRNA transcripts. After the addition of rifampin to the cells, the half-lives of the AnsR-U1, AnsR-U2, AnsR-U3, AnsR-U4 and AnsR-U5 transcripts were 11, 17, 19, 21 and 22.5 min, respectively (Fig. 6C and D). Furthermore, in the hfq mutant, the half-lives of the AnsR-U1, AnsR-U2, and AnsR-U3 transcripts were reduced to approximately 5 min, whereas those of AnsR-U4 and AnsR-U5 were reduced to 6 min (Fig. 6E). To determine whether Hfq directly participates in regulating atsRNA stability, we chose three variants (AnsR-U5, AnsR-U4, and AnsR-U3) as representatives, since their half-lives in the hfq mutant covered the observed range (6 min and 5 min) and their numbers of Hfq-binding sites differ substantially. Moreover, the results of MST experiments revealed that AnsR-U3, AnsR-U4 and AnsR-U5 were capable of binding to the Hfq protein, with dissociation constants of 16 ± 0.33 nM, 13 ± 1.27 nM, and 8 ± 0.42 nM, respectively (Fig. 6F–H), indicating that binding affinity decreased as the number of Hfq-binding sites was reduced. Together, these data suggest that the Hfq protein enhances the ability of atsRNAs to suppress gene expression by increasing their stability.

These results highlight the role of the Hfq protein in atsRNA gene repression and the optimal prototype for the CrcZ-based atsRNA expression platform containing the target recognition sequence in the 5′ region, from the putative RBS to the AUG, with five Hfq-binding sites in the middle and a terminator hairpin in the 3′ region.

4. Discussion

For cell engineering and genetic network reconstruction, many modified gene expression systems have been developed, including systems that function at the transcriptional level[36,[38], [39], [40], [41], [42], [43], [44], [45]] and those that affect mRNA stability and translation [[46], [47], [48], [49], [50]]. Unlike conventional gene knockout strategies, the use of atsRNAs does not require chromosomal modifications, which enables high-throughput studies on the effects of knockdowns on host cell physiology [6]. Therefore, the use of the prokaryotic plasmid-based synthetic sRNA system has become popular in the past decade and has become an efficient and convenient approach to address current challenges in metabolic engineering and synthetic biology [8]. The root-associated bacterium P. stutzeri A1501 is a rare example of a Pseudomonas strain with nitrogen fixation ability and is important for nitrogen supply in agricultural production [51]. Nevertheless, the nitrogen fixation control system in A1501 is highly intricate, partly because it manages the balance between carbon and nitrogen metabolism. Recent studies on native non-coding RNAs in A1501, such as CrcZ and NfiS [33,34], have elucidated the post-transcriptional regulatory network governing these processes. However, the functional repertoire of its endogenous sRNAs remains largely uncharacterised, and systematic tools for targeted, tuneable gene repression are lacking. This limitation hinders the precise rewiring of its complex regulatory circuits. Thus, developing a synthetic sRNA system will enable us to improve the engineering of synthetic gene networks of A1501.

In the past decade, rational design strategies for atsRNAs have been developed in E. coli. Rationally designed synthetic sRNAs that consist of the E. coli MicC scaffold and a replaceable target-binding sequence have been applied to engineer a wide range of microorganisms [5,21,30]. The use of natural trans-coding sRNA scaffolds, which are known to recruit the cellular proteins involved in gene silencing, such as Hfq, is the key element of the design strategy for synthetic sRNAs. Previous work on synthetic sRNA engineering was performed on the basis of different native trans-acting sRNA scaffolds in E. coli, such as SgrS, MicF, MicC, Spot42 and SibC [8,21]. Unlike mammalian microRNAs, which share common and intuitive sequence characteristics [52], bacterial sRNAs are highly diverse in their sequences and in their interactions with their targets [53,54]. Therefore, an effective RNA silencing tool based on other natural sRNA scaffolds for prokaryotes remains to be developed. The regulatory RNA CrcZ is conserved across all sequenced Pseudomonas strains, where it functions as a central component of the Hfq/Crc-mediated carbon catabolite repression (CCR) network [55,56]. CrcZ possesses five conserved Hfq-binding motifs that enable strong interactions with Hfq, and its regulatory mechanism relies primarily on sequestering Hfq as a molecular decoy rather than on extensive base-pairing with target mRNAs. This protein-binding-dominant property significantly reduces sequence-dependent off-target effects compared with those of traditional antisense RNA designs, making CrcZ an advantageous scaffold for synthetic regulatory RNAs. In this system, Hfq recognises and binds target mRNAs, while Crc enhances the stability of Hfq/Crc/mRNA tripartite complexes despite lacking intrinsic RNA-binding activity; importantly, Crc does not interfere with the scaffold function of CrcZ, which is primarily determined by its secondary structure and Hfq interaction. Our previous studies have further demonstrated that by sequestering Hfq, CrcZ acts as a natural antagonist that relieves repression of non-preferred catabolic pathways and modulates carbon utilisation hierarchy in P. stutzeri A1501 [35]. Native CrcZ functions in a condition-dependent manner: it is active when the preferred carbon source is depleted but plays a negligible role when the preferred carbon source is abundant. In this study, experiments were performed under nitrogen-fixing conditions with lactate as the sole carbon source, which are conditions not conducive to the activity of the CrcZ–Hfq–mRNA regulatory system. Consequently, introduction of the atsRNA (AnsR-U5 or AnsR-O5) is unlikely to significantly affect native CrcZ targets, consistent with our observations of normal cell growth and specific recovery of nitrogenase activity. Collectively, these attributes ensure that synthetic sRNAs built on the CrcZ scaffold function independently of endogenous Crc and CrcZ activity, with minimal off-target effects.

In this study, we used P. stutzeri A1501 CrcZ as a scaffold and constructed a plasmid-based synthetic sRNA system. With this system, two anti-nifLA synthetic sRNAs, AnsR-U5 and AnsR-O5, were designed, and these sRNAs efficiently repressed the expression of the nifLA operon in P. stutzeri A1501, indicating that this system could suppress the expression of multiple genes without affecting the cellular burden in the same cell. The general strategy for designing synthetic sRNAs in E. coli targets the translation initiation region, including the Shine-Dalgarno sequence and the downstream 20–30 nucleotides of the target mRNA, and this principle has been successfully extended to other bacteria [6,21]. In this work, we also investigated how different target mRNA-binding regions affect atsRNA-mediated repression. AnsR-U5 and AnsR-O5 were designed to bind either the 5′ UTR or CDS of nifLA mRNA. The anti-nifLA sRNA-targeting regions overlapping with the 5′ UTR presented the greatest repression (>80%), whereas targeting the CDS region (AnsR-O5) resulted in relatively inefficient repression. These results indicate that the 5′ UTR is more effective target region for atsRNAs in P. stutzeri.

In addition to the target-binding region, the recruitment of Hfq to the sRNA scaffold is another key determinant of sRNA function. Hfq binding stabilizes the sRNA and facilitates its hybridization with the target mRNA, making it an essential step in atsRNA-mediated gene repression [16]. Although the Hfq proteins of E. coli and P. aeruginosa show functional complementarity [57], our phylogenetic analysis revealed that Hfq from P. stutzeri A1501 occupies a distinct position, which likely reflects its niche-specific adaptations such as nitrogen fixation. Such divergence, potentially involving variations on the RNA-binding surface, could compromise the efficacy of synthetic sRNAs designed based on E. coli models. Moreover, unlike the well-established synthetic sRNA system in E. coli, in which the scaffold normally contains only one Hfq-binding site, the P. stutzeri CrcZ scaffold has five A-rich Hfq-binding motifs. To assess whether increasing the number of Hfq-binding sites would improve the artificial sRNA repression capacity, we further constructed five truncated variants of AnsR-U5 and found that increasing the number of Hfq-binding sites could indeed enhance the interactions between atsRNAs and target mRNAs, resulting in greater gene silencing by atsRNAs. Although base pairing is believed to be the most effective determinant of target discrimination by atsRNAs [8], our results also highlight the importance of the chaperone protein Hfq in determining specificity. This study provides guidance for optimising the rational design principles for atsRNAs that can be used to fine-tune the expression of chromosomal genes.

5. Conclusions

In this study, we developed an atsRNA platform based on the Pseudomonas-conserved CrcZ sRNA scaffold derived from P. stutzeri A1501. Two designed atsRNAs targeting the nifLA operon effectively suppressed nifLA translation in the native host. We demonstrate that optimal silencing requires both accessibility of the 5′ UTR of the target mRNA and the presence of five Hfq-binding sites embedded within the CrcZ scaffold. These findings will offer a straightforward and generalisable strategy for targeted gene repression across Pseudomonas species.

CRediT authorship contribution statement

Yueyue Han: Writing – original draft, Methodology, Investigation, Data curation, Conceptualization. Yuhan Zhang: Writing – original draft, Data curation, Conceptualization. Chao Li: Methodology, Investigation. Min Lin: Project administration, Funding acquisition. Fanyang Lv: Resources, Methodology. Chen Qi: Resources, Data curation. Yongliang Yan: Writing – review & editing, Funding acquisition. Yuhua Zhan: Writing – review & editing, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

The authors declare no competing financial interest or personal relationships that could have influenced the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: