The ribosome assembly GTPase EngA is involved in redox signaling in cyanobacteria

Photosynthetic organisms must cope with environmental challenges, like those imposed by the succession of days and nights or by sudden changes in light intensities, that trigger global changes in gene expression and metabolism. The photosynthesis machinery is particularly susceptible to environmental changes and adaptation to them often involves redox-sensing proteins that are the targets of reactive oxygen species generated by photosynthesis activity. Here we show that EngA, an essential GTPase and ribosome-assembly protein involved in ribosome biogenesis in bacteria and chloroplasts, also plays a role in acclimatization to environmentally relevant stress in Synechococcus elongatus PCC7942 and that PipX, a promiscuous regulatory protein that binds to EngA, appears to fine-tune EngA activity. During growth in cold or high light conditions, the EngA levels rise, with a concomitant increase of the EngA/PipX ratio. However, a sudden increase in light intensity turns EngA into a growth inhibitor, a response involving residue Cys122 of EngA, which is part of the GD1-G4 motif NKCES of EngA proteins, with the cysteine conserved just in the cyanobacteria-chloroplast lineage. This work expands the repertoire of ribosome-related factors transmitting redox signals in photosynthetic organisms and provides additional insights into the complexity of the regulatory interactions mediated by EngA and PipX.

PII is in 14-fold excess over PipX, which is in a roughly similar excess over its regulatory targets NtcA and EngA (Guerreiro et al., 2014;Labella et al., 2016). However, it is not known whether the EngA/PipX ratio changes under environmentally relevant conditions such as cold stress, where PipX counteracts EngA activity.
The aim of this work was to gain additional insights into the functions and peculiarities of cyanobacterial EngA and its regulatory connections with PipX. We have shown that upregulation of EngA levels, with a concomitant increase of the EngA/PipX ratio, takes place during acclimatization to environmentally relevant conditions such as low temperature or high light. In silico analysis of EngA revealed Proposed model for PipX regulation of EngA. Top panel: Schematic representation of the signaling pathway from PII to NtcA and EngA with indication of the relevant protein-protein interactions and alternative scenarios for PipX-EngA interactions. Circle areas are proportional to protein abundance (PII trimers; PipX: trimers, dimers and monomers from inner to outer circles, respectively; NtcA dimers; and EngA monomers). The red hammer head lines and the black arrow indicate inactivation by sequestration and co-activation, respectively. Middle panel "Bound or no PipX": under standard culture conditions PipX is involved in partner swapping with PII and NtcA while EngA is active. Bottom panel "Free PipX": when the levels of free PipX raise, it also binds to EngA, interfering with its ribosomeassembly function. PDB models: PII-PipX from 2XG8; NtcA-PipX from 2XKO; closed EngA predicted with SWISS model using as query S. elongatus protein sequence based on the PDB 4DCS model; open EngA from Escherichia coli model 3J8G. Re-published from Jerez et al. (2021).
Frontiers in Microbiology 03 frontiersin.org distinctive structural features conserved in the cyanobacteriachloroplast lineage. One of these cyanobacterial signatures is the presence of a conserved cysteine at the GD1-G4 motif (NKXES in bacteria) of cyanobacterial EngA proteins, which mediates a growth inhibition response triggered by high light in S. elongatus. This work expands the repertoire of ribosome-related factors transmitting redox signals in photosynthetic organisms and provides additional insights into the complexity of the regulatory interactions mediated by EngA and PipX.
To monitor the growth, optical density at 750 nm (OD 750nm ) was measured using a Ultrospec 2,100 pro UV-Vis Spectrophotometer (Amersham). For growth in liquid, 30-50 mL of cultures were adjusted to an initial OD 750nm of 0.1 and grown until they reach 0.5-0.6 (timepoint 0). Cold conditions (18°C) were performed in a FOC 200I Connect Cooled Incubator (VELP Scientifica) with cool white fluorescent lights (40 μmol photons m −2 s −1 ). Low (LL) and moderate light (ML) conditions (30°C and 2-3 or 70 μmol photons m −2 s −1 , respectively) were provided by cool white fluorescent lights, and high light conditions (HL, 1000 μmol photons m −2 s −1 ) by a MASTER SON-T PIA Plus 250 W E40 lamp.
For growth on solid media, exponentially growing cultures (mix of approximately the same amount of biomess from three clones) were adjusted to 0.5 before dropping 5 μL of the cell suspensions and serial dilutions (5 −1 , 10 −1 , and 10 −2 ) onto BG11 plates. Isopropyl β-D-1thiogalactopyranoside (IPTG) was added as indicated. Cells were grown at 30°C under standard conditions, high light intensity (400 μmol photons m −2 s −1 , provided by cool white LED lights), or exposed to a very high light pulse (VHL, 850 μmol photons m −2 s −1 , provided by a MASTER SON-T PIA Plus 250 W E40 lamp) for 20 min and then grown under standard illumination for 5 days. To induce EngA expression before a VHL pulse, 3 N Ptrc-EngA and 3 N Ptrc Osym -EngA/ΔengA cultures were grown in liquid BG11 for 2 h with the addition of 1 mM IPTG, washed with BG11, and adjusted to OD 750nm 0.5 before doing the drops in BG11 plates, grown at moderate light. To maximize the differences, photographs were always taken after 5 days growing using a Nikon camera at the default parameters. Pictures were analyzed using ImageJ. Circular regions of interest (ROI) were manually generated for each drop, and the average pixel intensity in the red channel was retrieved. An empty ROI was generated to obtain the background noise value of each picture which was subtracted from the other ROI average values. The growth of each strain relative to the control strain was calculated as the minimum value of the ratios between the measured biomass values of drops of the same dilution. Wilcoxon rank sum test with Bonferroni correction versus the control strain was performed with the RStudio program (RStudio Team, 2020).

Plasmid and strains construction
All plasmids constructed in this work (Table 1) were analyzed by automated dideoxy DNA sequencing. Primers used are described in Supplementary Table S1.

Protein extraction, immunodetection, and band quantification
To perform immunodetection of PipX, EngA, and PII in S. elongatus, 10 mL of cultures from the growth curves were harvested by centrifugation (7,300 g for 6 min at 4°C) at different times. The pellets were lysed in 100 μL of lysis buffer (50 mM Tris/HCl pH 7.4, 4 mM EDTA, 0.5 mM PMSF, 0.5 mM benzamidine, 1 mM DTT) with the addition of 0.1 μm glass beads, following the procedure described in Labella et al. (2016). Cells were subjected to three cycles of 60 s disruption at a speed of 5 m/s in a high-speed homogenizer Minibeadbeater and 60 s of resting at 4°C. The lysates were centrifuged (5,500 g for 5 min at 4°C) and the supernatant fractions were extracted, To separate the proteins, the extracts were charged in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE; 10-20% polyacrylamide linear gradient) using two independent gels (one for detecting EngA and PII and the other for PipX and PII), followed by immunoblotting in 0.1-μm polyvinylidene fluoride membranes (from GE Healthcare), maintaining the temperature at 4°C. The membranes were blocked with Tris-Buffered Saline (TBS; 20 mM Tris/HCl pH 7.5, 500 mM NaCl) solution containing 5% nonfat dried milk for 30 min at room temperature and then incubated overnight in TBS solution containing 2% nonfat dried milk and the primary antibody (diluted at 1:5000 for PipX and EngA, and 1:10000 for PII). Rabbit antisera against EngA were obtained from Pineda Antikörper Service (Berlin, Germany) using ∼ 2 mg of pure recombinant His-EngA as an antigen and following a 60-day immunization protocol. The antiserum against PipX and PII proteins was donated by K. Forchhammer (Univ. Tübingen, Germany). The membranes were then incubated at room temperature for 1.5 h with a 1:150,000 dilution of ECL rabbit IgG, HRP-linked F (ab′)2 fragment (from donkey; GE Healthcare). The signal was detected adding the SuperSignal WestFemto reagent (Pierce) and recorded in a Biorad ChemiDoc Imager using the automatic exposure mode, avoiding pixel saturation.
Protein band intensities were quantified using the ImageJ software, with the "rectangle" function and the "wand" tool used to measure the area plot corresponding to the signal intensity. The area from the corresponding immunodetection (EngA or PipX) was normalized using the area of PII or the area of an inner band from the corresponding gel, as indicated. Data were represented as ratios of stress/standard, mutant/WT alleles or directly as signal intensity, always referred to the timepoint 0 h. Wilcoxon rank sum test with Bonferroni correction versus the control strain was performed with the RStudio program (RStudio Team, 2020).

Computational methods
EngA homologous sequences, labeled as COG1160 proteins, were obtained from the EGGNOG database v5.0. Sequences were split into cyanobacterial or non-cyanobacterial sequences according to their NCBI ID and the resulting sets of protein sequences were used as queries for MEME searches. MEME discriminative mode (Bailey and Elkan, 1994) was applied with two different criteria, both using the default parameters but reducing the maximum motif length to 12 or 18 residues. Cyanobacterial and non-cyanobacterial sequences were used as primary and control input sequences, respectively. Weblogos were generated using the raw aligned version of the sequences from EGGNOG. 1 EngA homologous sequences in the Streptophyta phylum were also retrieved from the EGGNOG database v5.0 using Arabidopsis thaliana homolog as query (AT3G12080). Raw alignment sequences 2 were realigned together with S. elongatus sequences using ClustalW with default parameters.
In all cases, to simplify the visualization of the weblogos, the alignments were trimmed for all positions in which Synpcc7942_2340 presented a gap and then the weblogo3 online tool 3 was used to generate the image using the color scheme "Chemistry. "

Results and discussion
EngA accumulates during cold stress in S. elongatus Despite the essential involvement of EngA in ribosome biogenesis, a crucial and highly regulated process in all living cells, little is known about the control of EngA expression in model bacterial systems and there are no reports on the regulation of EngA protein levels at low 1 http://eggnogapi5.embl.de/nog_data/text/raw_alg/COG1160 2 http://eggnogapi5.embl.de/nog_data/text/raw_alg/ENOG5037JM3 3 http://weblogo.threeplusone.com/create.cgi  (Farewell and Neidhardt, 1998;Gottesman, 2018;Zhang et al., 2018;Cheng-Guang and Gualerzi, 2021). Cold sensitivity, presumably related to the temperature dependence of rRNA folding and the role of EngA as an rRNA chaperone, has been reported in Der-depleted strains of Escherichia coli (Bharat and Brown, 2014) and in S. elongatus. In this cyanobacterium, EngA has a stimulatory role on growth rate at low temperatures that is counteracted by PipX while overexpression of EngA has no impact during growth in standard conditions (Jerez et al., 2021).
To determine whether EngA and/or PipX levels are affected by growth at low temperatures, we grew S. elongatus cultures at standard conditions before splitting them into two cultures to be grown at standard (30°C) or at low (18°C) temperature, two conditions in which growth rates are clearly different ( Figure 2A). Subsequent immunodetection analysis of EngA, PipX and PII proteins revealed that while at 30°C all three protein levels remained constant during the length of the experiment, at 18°C the levels of EngA progressively increased ( Figure 2B). To quantify the changes, we used the immunodetection signal for PII, indistinguishable amongst the different samples, as an internal control to normalize and calculate the 18°C/30°C ratios of the signal intensity for EngA or PipX bands. As shown in Figure 2C, there was a constant increase of EngA levels after transfer to 18°C, already apparent at the 3 h timepoint, and reaching almost 4-fold at the end of the 32 h long experiment. In clear contrast, PipX levels were independent of the temperature.
Temperature is a key environmental parameter, with low temperatures having major effects on biomass productivity and growth rate. The increase of EngA levels in response to cold stress supports the involvement of this ribosome-assembly factor in acclimatization to low temperatures, raising questions on whether it is a widespread phenomenon across bacteria or just a cyanobacterial peculiarity.
The increase of the EngA/PipX ratio provides additional parallelism between NtcA-PipX and EngA-PipX interactions in the context of transcription and translation regulation during nitrogen deficiency or cold stress, respectively. Since NtcA levels increase during nitrogen starvation (Espinosa et al., 2014), both PipX partners are upregulated under environmental conditions in which they are most required and this per se may increase the levels of the corresponding PipX complexes. However, while we know that 2-OG and high ATP/ADP are required to maximize NtcA-PipX complex formation, the molecular details governing EngA-PipX interactions during cold stress remain to be elucidated.

EngA also accumulates during high light stress in S. elongatus
Light is the most relevant environmental parameter for photosynthetic organisms. It has a major impact on biomass productivity and growth rate, but excess light also damages the photosynthesis machinery and can inhibit growth. Therefore, EngA levels under cold and high light stress in S. elongatus. (A) Growth curves (OD 750nm ) of S. elongatus cultures grown in parallel at standard (30°C) and low temperature conditions (18°C) or at low light (LL, 2-3 μmol m −2 s −1 ) and high light (HL, 1000 μmol m −2 s −1 ). (B) Representative immunodetection of PipX, EngA and PII proteins and fast green staining as a loading control. (C) Relative levels of the indicated proteins, previously normalized by PII and referred to the timepoint 0 h, shown as ratios of the values for the indicated culture conditions. The timepoint 0 always correspond to cultures that were growing at 30°C or at LL and then transferred to the same or different conditions, as indicated. Data are presented as means and error bars (standard deviation) of three biological replicates.
Frontiers in Microbiology 06 frontiersin.org acclimatization to strong light is a complex process requiring the implementation of a variety of protective mechanisms to reduce photoinhibition (Bailey and Grossman, 2008;Keren and Krieger-Liszkay, 2011;Muramatsu and Hihara, 2012). S. elongatus cultures that are acclimated to high light often show the yellowish appearance typical of stressed cultures undergoing chlorosis (Llop et al., 2023) but can nevertheless grow as fast or even faster than at the moderate light intensities used as standard conditions. To gain further insights into the environmental regulation of EngA and a possible correlation between EngA levels and growth rates, we next compared the levels of EngA in S. elongatus cultures growing under two very different light intensities, defined here as low (LL, 2-3 μmol m −2 s −1 ) or high light (HL, 1000 μmol m −2 s −1 ) conditions. As shown in Figure 2A, these non-standard culture conditions resulted in slower (LL curve) or a similar growth rate (HL curve) than control cultures (30° curve) at the moderate (standard) light conditions (70 μmol m −2 s −1 ) used in our laboratory.
Western analyses were carried out for 5 h following the transfer to high light, after which, due to important protein degradation, quantification became increasingly difficult from the chlorotic high light cultures. The results showed that EngA levels increased in extracts from cultures transferred to high light, increasing 3-fold just after 5 h ( Figures 2B,C). Therefore, the results suggested that the enlargement of the EngA pool correlates with acclimatization to high light.
While the temperature dependence of rRNA folding explains the importance of an enlarged EngA pool during growth at low temperatures, it is intriguing that increases in the EngA pool also occur under high light, an environmental condition that should not compromise rRNA folding.
Promoter sequences upstream engA are required for environmental regulation To investigate the level at which upregulation of EngA levels takes place we next determined EngA levels in strain 3 N Ptrc-EngA/ΔengA (see Figures 3A,B for details on strain construction), in which the ectopic engA gene is expressed from the Ptrc promoter while cat coding sequences, providing resistance to chloramphenicol-acetyltransferase, precisely replace those of engA. The corresponding allele (Ptrc::engA) is within a cassette encoding LacI repressor for control of gene expression and a selection marker (Nt R for nourseothricin-resistance). To provide recombination sites at the S. elongatus chromosome for allelic replacement the cassette is flanked by sequences from the neutral site NS3 (see details in Figure 3A). Therefore in this strain the only source of EngA, the Ptrc::engA allele, contains just the engA coding sequences under the control of the Ptrc promoter.
3 N Ptrc-EngA/ΔengA and the wild-type control were analyzed in parallel. Cultures were grown at standard (30°C) conditions before splitting them into two cultures to be grown at the same (30°C) or at low (18°C) temperatures and subsequently analyzed by Western blots at different time points. As shown in Figure 4, S. elongatus cells expressing engA from the Ptrc::engA allele failed to increase EngA levels upon transfer to 18°C, in contrast to wildtype cells encoding the native engA locus. Therefore, cis-acting sequences upstream of the engA gene are required for upregulation in response to cold, suggesting that there is control at least at the transcriptional level.

Engineering a S. elongatus strain with low and IPTG-inducible levels of EngA
Since engA is an essential gene that cannot be inactivated (Jerez et al., 2021), we attempted the construction of a conditional null strain in which the expression of engA was under the control of an efficiently repressed and inducible promoter, for which we tried to decrease the promoter leakage of the Ptrc::engA construct.
To increase the affinity of the LacI repressor for the lac operator controlling Ptrc::engA we introduced mutations to increase operator symmetry as described (Sadler et al., 1983 and Figure 3A). The resulting plasmid derivative (pUAGC87) and parental control (pUAGC77) were used to generate Nt R transformants that were subsequently PCR-analyzed to verify homozygosis for the corresponding Ptrc Osym ::engA and Ptrc::engA alleles ( Figure 3B, top), generating strains 3 N Ptrc Osym -EngA and 3 N Ptrc-EngA, respectively. To confirm that the changes did not significantly affect the induced levels of expression from Ptrc in S. elongatus, we performed western blots to detect EngA from extracts of 3 N Ptrc-EngA or 3 N Ptrc Osym -EngA cultures induced with 1 mM IPTG ( Figure 3C). As expected, no significant differences were obtained at the 3 or 24 h timepoints, and thus the lac operator mutation does not impair IPTG-induction.
Next, the ΔengA allele was transformed in parallel into 3 N Ptrc-EngA and 3 N Ptrc Osym -EngA. Subsequent PCR analysis of chloramphenicol-resistant (Cm R ) clones confirmed the allelic replacement of the S. elongatus engA locus by cat, giving strains 3 N Ptrc-EngA/ΔengA and 3 N Ptrc Osym -EngA/ΔengA ( Figure 3B, lower panels). However, rapid segregation of the inactive allele independently of IPTG also occurred for strain 3 N Ptrc Osym -EngA/ΔengA, indicating that there was still enough promoter leakage from Ptrc Osym ::engA to complement EngA essential functions.
Western analysis of protein extracts from strains WT, 3 N Ptrc-EngA/ΔengA and 3 N Ptrc Osym -EngA/ΔengA detected small but reproducible differences between them, with both engineered derivatives producing slightly higher levels of EngA ( Figure 3D). Importantly, although there was still promoter leakage, repression was more efficient at Ptrc Osym than at Ptrc, in line with the idea that Ptrc can be more efficiently regulated in S. elongatus (Geerts et al., 1995) than in Synechocystis (Camsund et al., 2014). Thus, by obtaining Ptrc Osym we expanded the repertoire of IPTG-inducible promoters that can be used for gene expression studies in S. elongatus.
3 N Ptrc Osym -EngA/ΔengA, expressing at least as much EngA protein as the wild-type strain under standard culture conditions but devoid of its natural regulation, was used alongside other relevant strains to gain further insights into the physiological relevance of EngA upregulation and into the contribution of EngA or the EngA/PipX ratio to growth during high light stress. Information concerning EngA or PipX protein levels, PipX/EngA ratios, and regulation of relevant engA alleles of this panel of six S. elongatus strains (including the control strains WT and Ptrc) is summarized in  Table 3 are shown in Figure 5A. In high light 3 N Ptrc Osym -EngA/ΔengA grew slower than the 3 N Ptrc control strain, a result supporting the importance of upregulation of EngA levels for growth in these stress conditions. However, strain 3 N Ptrc-EngA, either in the absence or in the presence of IPTG, also grew (slightly) slower than the control in high light. Since strain 3 N Ptrc-EngA maintains the engA locus intact and should therefore be capable of upregulating EngA levels, it appears that in this strain the extra levels of EngA provided by the ectopic Ptrc::engA allele are not stimulating growth in high light but rather slowing it down. Thus, these results suggest that while the relatively small increase in EngA levels produced by upregulation of the engA locus may be important for acclimatization and growth in high light stress, additional or unregulated levels of EngA can also interfere with growth under high light conditions. It is worth noting that high light is the first experimental condition in which increased EngA levels have rather negative effects on S. elongatus growth and that IPTG-induced overexpression of EngA has opposite effects on growth under cold or high light stress, two environmentally relevant conditions in which EngA levels are upregulated. The implication is that the molecular mechanisms and details by which EngA mediates acclimatization to stress must differ between the two studied conditions. The negative role of EngA in the context of high light stress suggests that cyanobacterial EngA may have acquired specific functions related to the photosynthetic lifestyle.
Physiological levels of PipX are required for maximal growth of S. elongatus at high light Interestingly, both ΔpipX and 1 S Ptrc-PipX strains grew worse than their controls in high light ( Figure 5A). The finding that either eliminating PipX or increasing its levels some 4-fold slowed down growth at high light indicates that maximal growth on high light conditions requires physiological levels of PipX, supporting the idea that significant alterations of the PipX/EngA ratio have negative impacts on S. elongatus growth at high light. This is, to our knowledge, the first report of an environmental condition in which wild type S. elongatus has growth advantage over the ΔpipX mutant, implying that PipX plays a positive role specifically under conditions of high light stress. Given the ability of PipX to bind to EngA, presumably to sequester it, it is tempting to propose that, in addition to interference with EngA activity and growth, PipX may also interfere with the EngA-mediated inhibition of growth observed in high light.

Growth of S. elongatus at high light versus growth recovery after a pulse of very high light
High light damages the photosynthesis machinery, leading to photoinhibition of photosystem II (PSII), which is a protein-pigment complex particularly susceptible to photodamage. The repair of photodamaged PSII requires de novo synthesis of proteins, particularly of PSII subunit D1, a process which is also sensitive to oxidative stress caused by an accumulation of ROS (Reactive Oxygen Species; Nishiyama et al., 2001Nishiyama et al., , 2004. However, during the acclimatization of Synechocystis PCC6803 (hereafter Synechocystis) to very strong light, protein synthesis is accelerated, particularly in the case of protein D1 (Jimbo et al., 2019). In Synechocystis growth rate was correlated with the rate of synthesis of the D1 protein under different light intensities (Jimbo et al., 2019).
ROS inhibits the synthesis of D1 and of almost all proteins at the elongation step of translation (Nishiyama et al., 2001(Nishiyama et al., , 2004 and increases the levels of psbA mRNA (encoding D1) that is not associated with ribosomes, suggesting that the initiation of translation may be a target of ROS (Nishiyama et al., 2001). Interestingly, a connection between EngA and the PSII repair cycle has already been reported in A. thaliana thylakoids (Kato et al., 2018), suggesting to us that EngA may be regulating translation initiation and, as a result, growth under high light and that the sign of this regulation, positive or negative, would depend on the time that cultures have been exposed to high light. That is, we reasoned that the role of EngA would be different after a drastic increase in light intensity than later on, when multiple acclimatization responses are already operating.
To test this idea with an independent type of drop-plate assay, and hopefully to increase phenotypic differences amongst the studied strains, we next analyzed the growth recovery of S. elongatus cultures after a relatively short pulse of very strong light. In particular, a 20′ pulse of 850 μmoles photons m −2 s −1 was applied to culture drops before the plates were incubated under the moderated light of standard conditions ( Figure 5B). When it was pertinent to increase the levels of EngA, IPTG was added to cultures 2 h before drop plating.

EngA and PipX are both involved in delaying growth recovery after a very intense pulse of high light
Comparison between the two types of experiments shown in Figures 5A,B indicated that the growth recovery assay provided higher sensitivity than the growth assays under continuous high light, strengthening the previously observed negative impact of PipX or EngA excess under continuous high light. In fact, the negative impact of the EngA excess was already observed with non-treated cultures of 3 N Ptrc-EngA or 3 N Ptrc Osym -EngA/ΔengA and further increased in IPTG-treated cells, thus confirming the relevant role of EngA in promoting high light-dependent growth delay.
Therefore, independent assays confirmed that in addition to its growth-promoting function, EngA can also inhibit growth. Given the evolutionary conservation and complexity of interactions involved in the ribosome-assembly function of EngA, the role of EngA on growth inhibition upon high light exposure would presumably be the result of interfering with translation initiation.
Interestingly, the ΔpipX strain recovered faster than the control, indicating that PipX delayed growth recovery after the high light pulse. This result, apparently at odds with the slower growth of ΔpipX under continuous high light illustrates that the two types of assays give complementary information in relation to the response to high light Promoter sequences upstream of engA are required for up regulation. Representative immunodetection of EngA and PII and relative protein levels from strains 3 N Ptrc (engA allele) or 3 N Ptrc-EngA/ΔengA (Ptrc::engA) transferred to 18°C. Protein intensity levels were normalized by PII and referred to the timepoint 0 h. Data are presented as means and error bars (standard deviation) of three biological replicates.  Figure 3D).
Frontiers in Microbiology 09 frontiersin.org stress. While PipX stimulated growth specifically under continuous high light it interfered in a dosage-dependent manner with growth recovery after a high light pulse, suggesting that PipX enhances the inhibitory role of EngA upon sudden high light exposure.

The distinctive signatures of cyanobacterial EngA proteins include a redox motif
The inferred role of EngA as a sensor of high light stress and its physical interaction with PipX, a cyanobacterial hallmark protein, suggested that cyanobacterial EngA has acquired specific functions related to the photosynthetic lifestyle. If that was the case, these differences may encompass specific structural features distinguishing cyanobacterial EngA proteins from their prokaryotic homologs. With this in mind, we searched for cyanobacterial signatures based on the protein sequence.
Preliminary searches using ScanProsite and InterProScan with the Synpcc7942_2340 protein sequence just detected the already wellknown motifs of GTPases. To search for specific motifs that may be present exclusively in cyanobacterial EngA proteins, we compared cyanobacterial and non-cyanobacterial EggNOG EngA/COG1160 sequences with MEME discriminative mode (Bailey and Elkan, 1994).
The results of two independent searches with a motif maximum length of 12 or 18 amino acids agreed in the top 5 e-value hits, identifying five regions of 12 amino acids significantly overrepresented in cyanobacteria which are distributed alongside the protein sequence ( Figure 6; Supplementary Figure S1; Supplementary Table S2).
The first sequence (TRDRTYXXXXWX) overlaps with the GD1-G2 motif and includes two additional highly conserved aromatic residues. The second sequence (AVNKCESXXXGX) expands the GD1-G4 motif. It includes, at the non-conserved (X) position of the canonical NKX[D/E] G4 motif, a cysteine (Cys122 in S. elongatus) which is invariant in cyanobacteria and plants, with the notable exception of Gloeobacter, a cyanobacterium without thylakoids. The third sequence (GXEFFXINRXFK) is located between motifs GD2-G3 and GD2-G4 and contains a highly conserved submotif (INRXF). The fourth sequence (XWXSPPXXRXGX) is at the beginning of the KH domain and shares two consecutive prolines with the non-cyanobacterial sequences. The fifth sequence (GTPXRLXWRGKX), located close to the C-terminus of the protein is characterized by a highly conserved tryptophan.
The second signature sequence was very interesting since the cysteine, which is a hallmark of the AVNKCESXXXGX cyanobacterial signature, is often found at the X position of the G4 motif of RAS GTPases and subjected to posttranslational modification in the context of redox signaling (Lander et al., 1997;Hobbs et al., 2013;Messina et al., 2019). Since photosynthesis is a major source of redox agents (Hamilton, 2019) and these increase with light intensity, the redox-mediated inactivation of EngA would provide a mechanism to transfer redox information to the translation machinery. It would also explain the inhibitory role of EngA, triggered by exposure to drastic increases in light intensity, on the growth of S. elongatus cultures.
In summary, this analysis, showing distinctive structural features characteristic of the cyanobacterial-chloroplast lineage, provides a working hypothesis for the rather complex regulatory role of EngA in the context of high light stress. In particular, the inhibitory role of EngA would be associated with the oxidation of the conserved residue Cys122.

Cys122 at EngA is not essential in S. elongatus
To investigate the in vivo importance of Cys122 in cyanobacterial EngA we independently introduced mutations to encode Ala or Ser at the engA codon for Cys122 in plasmid pUAGC87. The resulting plasmid derivatives (pUAGC37 and pUAGC38) were then used to deliver Ptrc Osym ::engA C122A or Ptrc Osym ::engA C122S sequences, respectively, to the S. elongatus chromosome (Fig. S2), using the same procedure described above for the parental gene fusion. Construction of strains 3 N Ptrc Osym -EngA, 3 N Ptrc Osym -EngA C122A , and 3 N Ptrc Osym -EngA C122S was carried out in parallel and verified first by PCR to detect the presence of the ectopic alleles from the Nt R transformants, and then by RFLP analyses with AluI (C122A) or Hyp188I (C122S) to detect restriction sites incorporated with the mutations. Most of the AluI and all of the Hyp188I analyzed clones carried the corresponding mutation in homozygosis (Fig. S2). Verified clones with the C122A or C122S were named as 3 N Ptrc Osym -EngA C122A or 3 N Ptrc Osym -EngA C122S , respectively.
Strains 3 N Ptrc Osym -EngA C122A and 3 N Ptrc Osym -EngA C122S , in parallel with control strain 3 N Ptrc Osym -EngA were then used to inactivate engA by allelic replacement exactly as before. PCR confirmed the complete segregation of the ΔengA allele in the absence of IPTG and subsequent restriction analysis confirmed that only the point mutant alleles were present in the Cm R selected clones ( Figure 7A). Since both 3 N Ptrc Osym -EngA/ΔengA and 3 N Ptrc Osym -EngA C122A /ΔengA strains were viable, we concluded that Cys122 is not essential in S. elongatus. Furthermore, the growth of strains 3 N Ptrc Osym -EngA/ΔengA, 3 N Ptrc Osym -EngA C122A /ΔengA, or 3 N Ptrc Osym -EngA/ΔengA C122S on drop-plate assays under standard conditions ( Figure 7B, ML panel) was indistinguishable.
Substitutions at Cys122 of EngA prevent growth inhibition upon exposure to very intense light As shown in Figure 7B, growth tests performed under continuous high light did not reveal significant differences Distinctive features of cyanobacterial EngA proteins. SWISSmodel prediction of EngA structure with GD1, GD2 and KH domains colored green, orange and blue, respectively. A GDP molecule binding to GD2 is also shown. Prediction is based on Bacillus subtilis EngA structure, which is superimposed using Pymol and shown in grey. Combined results of two MEME discriminative analyses of cyanobacterial versus non-cyanobacterial EngA sequences are shown and numbered from N-to C-terminal direction. Positions corresponding to the two sets of results (12 and 18 length motifs, red and dark colors, respectively) are shown and EngA weblogos are indicated close to their location in the structure. G1-4 motifs in GD1 and GD2 are indicated as darker bars.
Frontiers in Microbiology 11 frontiersin.org between strains 3 N Ptrc Osym -EngA/ΔengA, 3 N Ptrc Osym -EngA C122A /ΔengA, or 3 N Ptrc Osym -EngA C122S /ΔengA. In contrast, recovery from a pulse of very high light was dramatically faster in the two mutant strains, with no appreciable differences between them, thus indicating that the Cys122 residue of EngA was involved in the growth delay triggered by exposure to the very strong light and that the response was suppressed by the conservative mutations C122A or C122S.
Therefore, the results suggest that the conserved NKCES motif is mediating redox signaling and that Cys122 is required to inhibit growth because of the ability to sense the rapid and dramatic increase in ROS triggered by drastic increases in light intensity. The comparison of the outcomes from the two types of drop-plate assays in the context of high light also illustrates the challenges of gaining information on novel signaling pathways and mechanisms from in vivo or genetic approaches, emphasizing the importance of using appropriated or diverse experimental designs.

Redox regulation of EngA, a new target of ROS in cyanobacteria
The involvement of ribosome-related factors in redox signaling was first reported in the context of translation elongation in Synechocystis, where ROS inhibits the synthesis of D1 and of almost all proteins at the elongation step of translation (Nishiyama et al., 2001(Nishiyama et al., , 2004. The elongation factors EF-Tu and EF-G are direct targets of ROS, they are inactivated via oxidation of conserved cysteine residues in Synechocystis and in the plant model system A. thaliana (Ejima et al., 2012;Yutthanasirikul et al., 2016;Jimbo et al., 2018;Toriu et al., 2023). Our results, showing the involvement of Cys122 from EngA in the growth inhibition of S. elongatus triggered by high light, add the ribosome-assembly protein EngA to the repertoire of ribosome-related factors transmitting redox signals generated by photosynthesis activity. Therefore, ribosome assembly would also be a key target of ROS in oxygenic photosynthetic organisms, where Cys at the GD1-G4 motif NKCES is a hallmark of EngA.

Model for redox regulation of EngA and contribution to translation and growth
To integrate previous literature with the information gained in this work we propose a model for redox regulation of EngA, summarized in Figure 8.
During moderate light growth conditions, simplified here as ML, EngA is expressed at basal levels. The EngA pool would be mainly in the reduced and active form, so basal levels would be sufficient to promote ribosome assembly, therefore allowing translation and culture growth. Immediately upon exposure to high light (HL start), ROS would trigger the oxidation of EngA at Cys122, inactivating most of the EngA pool, causing interference with EngA functions in translation initiation and slowing down growth. Whether this inactivation is also accompanied by self-aggregation, as it is the case with EF-Tu or EF-G, cannot be excluded. In this context, it is tempting to speculate that the ability of EngA to self-interact in bacterial two-hybrid assays for protein-protein interactions (Jerez et al., 2021), could reflect a tendency to self-aggregation that could also increase upon oxidation.
The drastic increase in ROS upon exposure to high light would trigger multiple transcriptional and posttranscriptional responses, including upregulation of the levels of ribosome-related factors such as EngA (Figure 2) or EF-Tu (Jimbo et al., 2019) and activation of redox-active enzymes such as thioredoxins (Mallén-Ponce et al., 2022). Together these responses would facilitate subsequent acclimatization to high light.
During long-term exposure to high light (HL acclimatization) and despite the continuous generation of ROS, the enlarged EngA pool would be in an active equilibrium between reduced and active states thanks to the action of thioredoxins. As a result of this and other protective mechanisms, EngA would stimulate translation and growth.
What is the role of PipX in this model? On the one hand, it appears that PipX toxicity, that is, the negative effect of PipX levels on S. elongatus growth, is accentuated in high light, conditions in which EngA activity is compromised by oxidative stress and a relatively low EngA/PipX ratio would interfere with EngA functions and growth. This is inferred from the impaired growth of strain 1 S Ptrc-PipX on high light and its slower recovery after a high light pulse, as well as by the faster recovery of the null pipX mutant after a high light pulse. On the other hand, the slightly impaired growth of the pipX mutant in high light indicated that during long-term exposure to high light PipX also plays a positive role and here it is tempting to propose that binding of PipX to part of the enlarged EngA pool favors the active form of EngA in detriment of the inactive one. Whether the formation of PipX-EngA complexes prevents oxidation of EngA and/or facilitates reactivation by assisting interactions with thioredoxins are two possibilities worth investigating.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions
AC designed research. AL, SB, RC, and PS performed research. AC wrote the paper. All authors analyzed data, contributed to manuscript revision, read, and approved the submitted version.

Funding
This work was supported by grant PID220-118816GB-I00 from the Spanish Government (MICINN) and grants VIGROB22-126 and VIGROB23-126 from the University of Alicante. SB is supported by a Model for redox regulation of EngA under high light and impact on translation and growth. Left panel: Under moderate light conditions, most EngA molecules are active (reduced), stimulating translation and growth. Middle panel: A drastic increase in light intensity (HL) increases ROS leading to massive oxidation of EngA, that becomes inactive and interferes with translation and growth. Right panel: The increase in ROS upon the initial exposure to HL triggered multiple protective transcriptional responses, including up regulation of the levels of EngA and activation of thioredoxins. This displaces the redox equilibrium towards reduction of EngA, that in turn is able to sustain translation and growth. A number of PipX-EngA complexes (represented by the X on the EngA molecules) would form in each of the three scenarios, with different outcomes. When EngA is at its basal levels, PipX binding would mainly interfere with EngA functions at the ribosome, but when EngA is abundant and ROS generation important binding to PipX would mainly protect it from EngA inactivation. For simplicity only PipX binding to the active form of EngA has been represented (see text for additional details).
Frontiers in Microbiology 13 frontiersin.org National Grant from the Algerian Ministry of Higher Education and Scientific Research.