Phototropin connects blue light perception to starch metabolism in green algae

Article https://doi.org/10.1038/s41467-025-57809-3 Phototropin connects blue light perception to starch metabolism in green algae Yizhong Yuan 1,11 , Anthony A. Iannetta 2 , Minjae Kim 3 ,PatricW.Sadecki 2 , Marius Arend 4,5,6 ,AngelikiTsichla 1,12 ,M.ÁguilaRuiz-Sola 1,13 , Georgios Kepesidis 1,14 , Denis Falconet 1 , Emmanuel Thevenon 1 , Marianne Tardif 7 ,SabineBrugière 7 , Yohann Couté 7 , Jean Philippe Kleman 8 ,IrinaSizova 9 , Marion Schilling 1 , Juliette Jouhet 1 , Peter Hegemann 9 , Yonghua Li-Beisson 3 , Zoran Nikoloski 4,5,6 , Olivier Bastien 1 ,LeslieM.Hicks 2 &DimitrisPetroutsos 1,10 In photosynthetic organisms, light acts as an environmental signal to control their development and physiology, as well as energy source to drive the conversion of CO 2 into carbohydrates used for growth or storage. The main storage carbohydrate in green algae is starch, which accumulates during the day and is broken down at night to meet cellular energy demands. The signaling role of light quality in the regulation of starch accumulation remains unexplored. Here, we identify PHOTOTROPIN-MEDIATED SIGNALING KINASE 1 (PMSK1) as a key regulator of starch metabolism in Chlamydomonas reinhardtii. In its phosphorylated form (PMSK1-P), it activates GLYCERALDEHYDE3-PHOSPHATE DEHYDROGENASE (GAP1), promoting starch biosynthesis. We show that blue light, perceived by PHOTOTROPIN, induces PMSK1 dephosphorylation that in turn represses GAP1 mRNA levels and reduces starch accumulation. These findings reveal a previously uncharacterized blue lightmediated signaling pathway that advances our understanding of photoreceptor-controlled carbon metabolism in microalgae. Photosynthetic microalgae convert light into chemical energy in the form of ATP and NADPH, which fuel CO 2 fixation in the Calvin–Benson cycle. This process of CO 2 fixation is initiated by the activity of the CO 2 - assimilating enzyme ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco)1. In eukaryotic algae, such as the model photosynthetic green microalga Chlamydomonas reinhardtii (hereafter Chlamydomonas), concentrated CO 2 is delivered to Rubisco within a specialized microcompartment in the chloroplast called the pyrenoid2. Light is also a spatiotemporal signal; red light is detected by bilincontaining phytochromes, while blue light is perceived by flavinReceived: 5 April 2024 Accepted: 21 February 2025 Check for updates 1 Université Grenoble Alpes, CNRS, CEA, INRAE, IRIG-LPCV, Grenoble, France. 2 Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. 3 Institute de Biosciences et Biotechnologies Aix-Marseille, Aix Marseille University, CEA, CNRS, BIAM, Saint Paul-Lez-Durance, France. 4 Bioinformatics Department, Institute of Biochemistry and Biology, University of Potsdam, Potsdam, Germany. 5 Systems Biology and Mathematical Modeling Group, Max Planck Institute of Molecular Plant Physiology, Potsdam, Germany. 6 Bioinformatics and Mathematical Modeling Department, Center of Plant Systems Biology and Biotechnology, Plovdiv, Bulgaria. 7 Université Grenoble Alpes, INSERM, CEA, UA13 BGE, CNRS, CEA, FR2048, Grenoble, France. 8 Université Grenoble Alpes, CNRS, CEA, IBS, Grenoble, France. 9 Institute of Biology, Experimental Biophysics, Humboldt-Universität zu Berlin, Berlin, Germany. 10 Department of Organismal Biology, Uppsala University, Uppsala, Sweden. 11 Present address: Department of Plant Biology, Carnegie Institution for Science, Stanford, CA, USA. 12 Present address: Department of Organismal Biology, Uppsala University, Uppsala, Sweden. 13 Present address: Instituto de Bioquímica Vegetal y Fotosíntesis, Consejo Superior de Investigaciones Científicas-Universidad de Sevilla, Sevilla, Spain. 14 Present address: Sandia National Laboratories, Livermore, CA, USA. e-mail: di[email protected] Nature Communications | (2025) 16:2545 1 1234567890():,; 1234567890():,; containing cryptochromes and/or PHOTs, the latter characterized by two photosensory light, oxygen, or voltage (LOV) domains3.Chlamydomonas notably lacks phytochromes in its genome4, nevertheless, this microalga has evolved a rich repertoire of photoreceptors, including a single-copy PHOT, four cryptochromes, eight rhodopsinlike proteins, and the UV-B photoreceptor UVR85. Through this network of specialized photoreceptors, Chlamydomonas regulates important cellular functions, including: gene expression, sexual life cycle, phototaxis, and photoprotection6–13. Fixed CO 2 in combination with nitrogen is used to synthesize amino acids—the building blocks of proteins that drive biochemical reactions. It is also employed in the synthesis of cellular reserves that ensures carbon and energy supply during the waning period. The most abundantcarbonreservein themodelphotosyntheticgreenmicroalga Chlamydomonas is starch; its synthesis occurs during the day and its degradation is triggered at night to sustain energy-demanding cellular functions14. Starch also serves another critical role as starch sheaths encasing the pyrenoid act as a barrier, reducing CO 2 leakage from this structure15. While actively dividing, Chlamydomonas cells accumulate starch mostly around the pyrenoid. In contrast, when subjected to a stress such as nutrient limitation, starch granules are massively accumulating in the chloroplast stroma16. Little is known about the molecular mechanisms underlying the control exerted by light on carbon storage in microalgae, and current knowledge is limited to factors impacting starch accumulation under adverse environmental conditions, such as nitrogen17 and phosphorus18 limitation. A link between light perception and starchaccumulation has been suggested in vascular plants. For instance, Arabidopsis mutants devoid of the red/far-red photoreceptor phytochrome B have impaired carbon partitioning and although they have reduced CO 2 uptake, they over-accumulate daytime sucrose and starch at the expense of growth19. Further, the blue light receptor PHOTOTROPIN (PHOT) has been found to mediate starch degradation in guard cells in the light, thus energizing stomatal opening in Arabidopsis20. Despite early findings reporting that light quality impacts carbohydrate accumulation and metabolism in green algae21,22, the underlying molecular mechanism connecting light perception to starch accumulation remains unexplored. Here we address this gap by combining genetics, proteomics, and phosphomimetics to unveil a signaling cascade linking blue light perception by PHOT with starch accumulation in Chlamydomonas. Phot-dependent de-phosphorylation at a single serine residue of phototropin-mediated signalling kinase 1 (PMSK1) represses GAP1 (also known as GAPDH; glyceraldehyde-3-phosphate dehydrogenase), which we found to act as an enhancer of starch metabolism. Results The phot mutant is a starch hyperaccumulator We compared starch levels in wild-type CC125 cells (WT) exposed to white and red light and found that red light favored starch accumulation, in agreement with prior work21. Interestingly, when low fluence blue light was superimposed on red light, the beneficial effect of red light was lost and cells accumulated starch levels similar to those of cells exposed to white light (Supplementary Fig. 1). Therefore, we reasoned that blue light acts as a repressor of starch accumulation likely via a blue light receptor. To test this hypothesis, we measured starch content in WT and mutant cells lacking different blue-light receptors, including: acry, lacking the ANIMAL-TYPE CRYPTOCHROME6, generated in this study, pcry lacking the PLANT-TYPE CRYPTOCHROME23, generated in this study, the double acrypcry mutant, generated in this study, and the phot mutant, devoid of PHOTOTROPIN (generated in ref. 24). When grown asynchronously under continuous light, the starch content of the cryptochrome mutants was indistinguishable from that of the WT (Fig. 1a) and the same was true when cultures were synchronized under 12 h light/12 h dark cycles (Supplementary Fig. 2), favoring cryptochrome accumulation in Chlamydomonas6,23.Incontrasttothe cryptochrome mutants, the phot mutant accumulated approximately three times more starch than the WT under continuous illumination (Fig. 1a). This phenotype was rescued by ectopic expression of the WT PHOT gene; it was fully rescued in strain phot-C125 (Fig. 1b), which accumulates PHOT protein at WT levels (Supplementary Fig. 3a), and partially rescued in strain phot-C2 (Fig. 1b), which accumulates PHOT protein but to a lesser extent than the WT (Supplementary Fig. 3a). Complete rescue of the phenotype was also achieved by complementation of a permanently active PHOT mutant lacking the LOV sensory domains (phot-kin strain; Fig. 1b and Supplementary Fig. 3a, b); aphot-kin strain in a different genetic background has previously been shown to exhibit PHOT kinase activity in a light-quality-independent manner13. The accumulation of high levels of starch in the phot mutant was also confirmed by transmission electron microscopy (TEM), revealing striking changes in starch deposition patterns compared to the wild type. As indicated by as the arrows in the TEM images in Fig. 1c, the chloroplasts of the phot mutant are filled with starch granules and its pyrenoid, is surrounded by abnormally thick starch sheaths. In contrast, in the chloroplasts of the WT and phot-C1 strains starch is predominantly localized as thin sheaths around the pyrenoid (Fig. 1c and Supplementary Fig. 4). We further explored the link between PHOT and starch accumulation in synchronized cultures under white, blue or red light. Starch accumulated during the light phase and degraded during the dark phase, in accord with14,26, under all light qualities and in all different strains tested (Fig. 1d). We found that starch levels in WT are higher under red light, where PHOT is inactive, than under white or blue, wherePHOTis active(Fig. 1d). In the phot mutant, starch levels are high in all three light qualities, while the fully complemented phot-C1 behaves like WT and the partially complemented phot-C2 is intermediate between WT and phot.Incontrasttothephot mutant, starch levels in strain phot-kin,inwhichPHOTisactive,arelowandunaffected by the light quality (Fig. 1d). These data (Fig. 1) indicate that PHOT acts as suppressor of starch accumulation in Chlamydomonas. We provided further evidence for this claim by using another set of phot mutants in a different genetic background, the cw15-3028,13.In accordance with the data in Fig. 1, we found that the cw15-302 phot mutant accumulated more starch than cw15-302 (Supplementary Fig. 3c). This phenotype was only partially rescued in the complemented pphot strain, which accumulates lower levels of PHOT protein than the cw15-302 WT strain13, and was fully rescued in the cw15-302 phot mutant expressing the PHOT kinase domain (pkin strain; Supplementary Fig. 3c). Conversely, complementation of cw15302 phot cells with truncated gene carrying only the photosensory domains LOV1 and 2 (plov strain) did not rescue starch accumulation. Importantly,we observed a similar phenotype when the cw15-302 phot mutant was complemented with the dead kinase PHOT (pkin-D strain; Supplementary Fig. 3c), demonstrating that the suppression of starch accumulation by PHOT requires its kinase activity. Growth (Supplementary Fig. 5a) and photosynthesis (Supplementary Fig. 5b) were not affected by the overaccumulation of starch in phot (Fig. 1c). Finally, phot showed no difference to WT in terms of total protein (Supplementary Fig. 6), lipid content (Supplementary Fig. 7) and composition (Supplementary Fig. 8). GAP1 is a key regulator of starch metabolism and is controlled by PHOT To gain more insight into the molecular mechanism underlying the PHOT-mediated light perception and starch accumulation, we applied mass spectrometry (MS)-based quantitative proteomics to compare WT and phot cells grown in asynchronous photoautotrophic conditions under continuous white light. Gene ontology enrichment analyses revealed that carbohydrate metabolic processes are upregulated in the phot mutant (Fig. 2a). We notably found Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 2 that GLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE (GAP1) was expressed 27.5-fold higher in phot compared to WT (Fig. 2a, Supplementary Data 2). GAP1 is a chloroplast isoform of GAPDH in Chlamydomonas, a Calvin-Benson cycle enzyme predominantly localized in the stromal region surrounding the pyrenoid27,28. The significant accumulation of GAP1 in phot was in accord with the high accumulation of GAP1 mRNA (Fig. 2b). While mRNA accumulation of all tested starch-related genes was high in the phot mutant in the middle of the light phase (Supplementary Fig. 9), GAP1 was the most highly expressed, prompting further investigation into the link between PHOT and GAP1. We found that GAP1 mRNA overaccumulation phenotype of phot was completely rescued in phot-C1 and phot-kin and partially rescued in phot-C2 (Fig. 2b), in accord with the PHOT expression levels in the different complemented lines (Supplementary Fig. 3a, b), suggesting that PHOT acts as suppressor of GAP1 upon illumination. Similar to the PHOT-mediated suppression of starch (Supplementary Fig. 3c), the PHOT-dependent suppression of GAP1 requires the kinase activity of PHOT (Supplementary Fig. 10). The pkin-D strain, a phot mutant expressing a dead kinase PHOT with point mutations in the ATP binding site8and the plov strain expressing only the sensory domains LOV, both accumulate GAP1 mRNA at the level of phot. Complementation of the phot mutant with the full-length PHOT (phot strain) or its kinase domain (pkin strain) rescues GAP1 mRNA to WT levels (Supplementary Fig. 10). We also measured GAP1 mRNA in cells synchronized to a 12/12 light-darkcycle. GAP1 mRNA in WT was found to be strongly influenced by the diurnal cycle, starting at very low levels after the initial onset of white light, rising to maximal levels at the end of the light phase, and then progressively and slightly declining during the night phase (Fig. 2c); this is in accordance with data from previous studies29,30 on the impact of diel cycle on the genome-wide gene expression (replotted in Supplementary Fig. 11). GAP1 mRNA accumulation profile was identical in white and blue light conditions. However, under red light, expression levels at the beginning of the day were significantly bigger and reached approximately 45 times higher levels at the midpoint of the light phase (Fig. 2c), suggesting that the low levels of GAP1 mRNA after the initial onset of light are a result of suppression by blue light. Indeed, in all light qualities tested, GAP1 mRNA levels in the synchronized phot mutant(Fig.2d)wereidenticaltothoseoftheWTin red light (Fig. 2c). The fully complemented phot-C1 behaved like WT (Fig. 2e) while the partially complemented phot-C2 only partially WT phot-C1phot ab WT acry pcry acrypcry phot 20 WT phot phot-C1 phot-C2 phot-kin Starch content (μg/million cells) WT phot phot-C1 phot-C2 phot-kin 0122436 0 5 10 15 20 25 0122436 d c 01224 36 Starch content (μg/million cells) Starch content (μg/million cells) 0 5 10 15 20 ✱✱✱✱ ✱✱✱✱ ✱ ✱✱✱✱ 0 5 10 15 ✱✱✱✱ ns ns ns Time (h) Fig. 1 | PHOT inhibits starch accumulation in Chlamydomonas reinhardtii. aStarch content of WT, acry, pcry, acrypcry lines under continuous light. bStarch content of WT and variously phot-complemented lines under continuous white light. cTransmission electron microscopy pictures of WT, phot and phot-C synchronized to a12/12 dark/light cycle. Samples were collected at the end of the light phase. Red arrows indicate starch granules. Representative images from three replicates were shown. dStarch content of WT and phot-complemented lines synchronized to 12/12 dark/light cycle. Red triangles indicate sample collection time. The dark phase is indicated by blackbars above the graphs; the light phase by white, blue or red bars, depending on thelight quality used.In some cases, the error bars are smaller than the data point symbols. Data are represented as mean±SD (n = 3 biologically independent samples). The statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons tests (a,b)and two-way ANOVA with Dunnett’s multiple comparisons tests (d). Asterisks indicated the p-values (*p < 0.05; ****p < 0.0001; ns, not significant). In some cases, the error bars are smaller than the data point symbols. Detailed statistical analyses are presented in the Source Data File. Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 3 rescued the observed phenotype (Fig. 2f). Notably, the phot-kin strain in which PHOT is devoid of the photosensory LOV domains and is always active regardless of light quality13, accumulated very low levels of GAP1 mRNA (Fig. 2g; note the log scale). However, in phot-kin cells grown under red light, GAP1 mRNA levels are higher at 6h compared to cells grown in white or blue light (Fig. 2g). Despite, the increased GAP1 mRNA in red light this was not sufficient for overaccumulation of starch (Fig. 1d). This indicates that additional regulators, beyond PHOT, may be involved in repressing GAP1 in a blue light-dependent manner. Our results so far show a strong association between starch and GAP1 mRNA accumulation in the different phot mutants (Figs. 1band2b). WT, phot-C1 and phot-kin accumulate low amounts of starch (Fig. 1b) and low mRNA GAP1 (Fig. 2b), phot accumulates high Rel. GAP1 mRNA abundance 10 -1 10 0 10 1 10 2 10 3 WT phot phot-C1 phot-C2 phot-kin ab de f Carbohydrate metabolism Photosynthesis Generation of precursor metbolites Photosynthesis light reaction Photosynthesis light harvesting ✱✱✱✱ ✱✱✱✱ ✱ ✱✱✱✱ WT gap1 oe-1 gap1 oe-2 gap1 oe-3 phot phot gap1i-1 phot gap1i-2 phot gap1i-3 0 5 10 15 20 Starch content (μg/million cells) ✱✱ ✱ ns ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ WT gap1 oe-1 gap1 oe-2 gap1 oe-3 phot phot gap1i-1 phot gap1i-2 phot gap1i-3 10 -1 10 0 10 1 10 2 10 3 Rel. GAP1 mRNA abundance ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱ ✱✱✱✱ ✱✱ ns WT phot phot-C1 phot-C2 phot-kin c 10 -2 10 -1 10 0 10 1 10 2 10 3 Rel. GAP1 mRNA abundance 10 -2 10 -1 10 0 10 1 10 2 10 3 Rel. GAP1 mRNA abundance 061218 24 061218 24 Time (h) g hi Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 4 levelsofstarch(Fig.1b) andhighGAP1 mRNAlevels(Fig.2b), and finally the partially complemented phot-C2 (Supplementary Fig. 3a) accumulates intermediate levels of starch and GAP1 mRNA (Fig. 1band Fig. 2b). To further confirm the positive connection between GAP1 expression level and starch amount, we generated WT strains overexpressing GAP1 (Supplementary Fig. 12). Strains gap1-oe1 and gap1oe2 accumulated approximately 15-fold more GAP1 mRNA than WT (Fig. 2h), resulting in 1.5-fold higher starch content (Fig. 2i). The strain gap1-oe3, with a more modest overexpression of GAP1 mRNA (2.5-fold compared to WT; Fig. 2h), had starch content at WT levels (Fig. 2i). For comparison, we included the phot mutant in these analyses, which expressed 80-fold more GAP1 and accumulated 3-fold more starch than the WT. We also downregulated GAP1 in the phot mutant and generated strains phot-gap1-i1,2and 3that accumulated 1.6-, 3.1and 5.5-fold less GAP1 mRNA (Fig. 2h) and 1.3-, 1.5and 2-fold less starch, respectively, compared to phot (Fig. 2i). In conclusion, overexpression of GAP1 in WT led to overaccumulation of starch whereas downregulation of GAP1 in phot decreased starch accumulation (Fig. 2h, i). These results (Fig. 2) strongly suggest that GAP1 plays a key role in starch metabolism in Chlamydomonas, acting under the control of PHOT. PHOT alters phosphorylation state of PHOTOTROPINMEDIATED SIGNALLING KINASE 1 To identify missing components in the phototropin-mediated signaling pathwaythat suppressstarch accumulation in Chlamydomonas,we compared the phosphoproteome of WT and phot cells after an overnight dark acclimation and a 5-min exposure to blue light. We identified 1119 phosphopeptides, belonging to 747 phosphoproteins, and applied unsupervised hierarchical clustering of z-transformed relative abundance of phosphopeptides to obtain an overview of the condition-specific and phototropin-dependent phosphorylation changes (Supplementary Fig. 13 and Supplementary Data 3). Phosphopeptides falling within cluster “E”were highly phosphorylated in the dark and became de-phosphorylated after blue light illumination, but only in the case of WT; in the phot mutant they remained highly phosphorylated after blue light illumination. One such peptide was the ADGVSpSPHELTR, phosphorylated at serine120 (S120), which belongs to the gene product of Cre16.g659400 encoding a Ser/Thr protein kinase (Fig. 3b), localized in the cytosol, plasma membrane and flagella (Supplementary Fig. 14). We named this kinase phototropin-mediated signalling kinase 1 (PMSK1). To validate the phosphoproteomic findings, we generated WT and phot lines expressing PMSK1 fused to a FLAG epitope. These lines (WT/PMSK1-FLAG and phot/PMSK1-FLAG)weredarkacclimatedprior exposure to red orblue light and samples were taken at 5minintervals for a period of 20 min to assess the PMSK1-FLAG phosphorylation status using Phos-tag SDS-PAGE, followed by immunodetection against FLAG (Supplementary Fig. 15). In accordance with the role of PHOT in the light-dependent dephosphorylation of PMSK1 detected in our phosphoproteomic analyses (Fig. 3a), the Phos-tag data revealed a rapid dephosphorylation of PMSK1-FLAG (Fig. 3c) upon exposure to light. In the WT background, dephosphorylation occurred only under blue light (Fig. 3c). In contrast, in the phot background, dephosphorylation was barely observed, even under blue light (Fig. 3c). Furthermore, no dephosphorylation was observed under red light in either strain (Fig. 3c). These data (Fig. 3a–c) support our hypothesis that dephosphorylation of PMSK1 requires blue light-activated PHOT. Interestingly, the observed blue light dependent mobility shift of PMSK1-FLAG (Fig. 3c) was abolished when S120 was substituted by an alanine (A) or an aspartic acid (D) residue (Fig. 3d). These data indicate thatphosphorylationatresidueS120isessentialfortheshift.However, this shift may not be solely attributed to S120 phosphorylation, but rather to additional phosphorylation events that depend on the initial phosphorylation of S120. Phosphorylation state of serine 120 of PMSK1 controls starch accumulation We next set out to investigate the functional significance of the phosphorylation status of S120 in vivo, taking advantage of the phosphomimetic mutation S120D and the non-phosphorylatable mutation S120A in PMSK1-FLAG, expressed in WT and in phot (Supplementary Fig. 15). We measured GAP1 mRNA and starch accumulation in WT, phot and all above-mentioned generated mutants, synchronized under white, blue or red light in a 12 h light/12 h dark regime. Overexpression of PMSK1S120D-FLAG in WT, resulted in an enhanced GAP1 expression level and increased starch content under all three light qualities, exceeding those recorded in the phot mutant (Fig. 4a, b and Supplementary Figs. 16 and 17). Overexpression in phot of PMSK1S120D-FLAG resulted in an even further higher GAP1 expressionlevel and increased starch content as compared to phot (Fig.4a,b and Supplementary Fig. 17). Conversely, overexpression inphot orWT of PMSK1S120A-FLAG, resulted in low, WT-level or even lower, GAP1 mRNA and starch levels, across all three light quality tests (Fig. 4a, b and Supplementary Figs. 16 and 17). We also analyzed the impact of overexpression of the unmodified PMSK1-FLAG in both WT and phot mutant (Fig. 4a, b and Supplementary Figs. 15 and 17); WT and WT/ PMSK1-FLAG behaved very similarly in all light colors with respect to GAP1 mRNA and starch accumulation (Fig. 4a, b and Supplementary Fig. 17). However, phot/PMSK1-FLAG showed increased GAP1 mRNA and accumulated more starch (Fig. 4a, b and Supplementary Fig. 17). This increase is likely due to the absence of PHOT, causing both endogenous and transgenic PMSK1-FLAG in the cells to remain phosphorylated (Fig. 3c). Taken together, our data (Fig. 4) reveal that the phosphorylation state of S120 of PMSK1 controls starch metabolism through regulation of GAP1 mRNA levels.To investigate if PMSK1’s role is dependent on its kinase activity, we used a kinase-inactive version ofPMSK1, where Asp442 was substituted with Asn (D442N) to disrupt the ATP-binding site, alongside the phosphomimetic S120D or the non-phosphorylatable S120A mutations. Neither mutation (S120D or S120A) affected starch content (Supplementary Fig. 18), indicating that PMSK1 fulfills the observed regulatory role on starch metabolism through its kinase activity. Fig. 2 | GAP1 is inhibited by blue light via PHOT and plays key role in PHOTdependent starch metabolism in Chlamydomonas. aGene ontology enrichment analyses for differentially abundant proteins in phot compared to WT based on whole cell proteomics data. LFC, log2 fold change (enriched GO terms were only found for proteins with increased abundance in phot in comparison to WT); color code indicates adjusted p-value of GO set enrichment (null hypothesis: number of differentially abundant proteins in GO set is hypergeometric random distributed. P-value were adjusted according to Benjamini-Hochberg procedure bGAP1 relative mRNA abundance of WT and phot-complemented lines under continuous white light. c–grelative mRNA abundance of GAP1 in WT and phot-complemented lines synchronized to a 12/12 dark/light cycle. Phases are indicated by white and gray shading. Line colors indicated light qualities. Black, white light; Red, red light; Blue, blue light. The GAP1 transcription level (h) and starch content (i)ofWT,GAP1 overexpression, and photgap1 knockdown lines under continuous light. Data are presented as mean± SD (n = 3 biologically independent samples). The statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisonstests(b,h) and two-way ANOVA with Dunnett’smultiple comparisonstests (c–g) of log10 transformed mRNA data as indicated in the graphs. The statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons tests (i). Asterisks indicated the p-values (*p < 0.05; **p< 0.01; ****p< 0.0001; ns, not significant). In some cases, the error bars are smaller thanthe data point symbols. Detailed statistical analyses are presented in the Source Data File. Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 5 PMSK1 acts downstream of PHOT to regulate starch metabolism in response to blue light We applied CRISPR-CAS9 to disrupt PMSK1 in the WT and in phot,thus generating the single pmsk1 and the double phot pmsk1 mutants (Supplementary Fig. 19). In comparison to WT, the pmsk1 mutant accumulated 3-, 4and 10-fold lower GAP1 mRNA levels in white, blue and red light, respectively (Fig. 5a; when comparing GAP1 mRNA levels at the middle of the day). Nevertheless, GAP1 expression in pmsk1 remained dependent on light-quality used, reaching higher expression levels under illumination with red light (Fig. 5a). However, this overaccumulation of GAP1 mRNA in red light was not sufficient for overaccumulation of starch (Fig. 5b and Supplementary Fig. 20), as in the case of the gap1 oe-3 line, which slightly overaccumulated GAP1 mRNA but starch levels remained comparable to WT levels (Fig. 2handi).Asa result, we concluded that starch accumulation in pmsk1 was lightquality-independent. The double mutant phot pmsk1 accumulated 3-fold less GAP1 mRNA and 2-fold less starch as compared to phot (Fig. 5a, b and Supplementary Fig. 20). Taken together, our results show that PMSK1 plays a critical role in transducing the PHOT-mediated blue light signal to regulate starch metabolism in Chlamydomonas.Indeed,thepmsk1 mutant has been instrumental in getting a clear picture of this distinct light-signalling pathway, summarized graphically in Fig. 6. Based on our data, we propose that PMSK1 is an activator of GAP1 mRNA accumulation, and its activity is regulated by the phosphorylation status of S120 in response to light quality. When PMSK1 is phosphorylated, it becomes active and promotes GAP1 mRNA accumulation. Conversely, when it is not phosphorylated, PMSK1 is inactive and does not further promote GAP1 mRNA accumulation. In addition to this regulation, GAP1 expression is influenced by other signals, possibly circadian, photosynthetic, or diurnal rhythms, responsible for the expression pattern observed over the course of the day. Red light drives the phosphorylated form of PMSK1 (PMSK1-P; Fig. 3c), resulting in high GAP1 mRNA accumulation (Fig. 4a). However, in white and blue light, dephosphorylation of active PMSK1-P results in an accumulation of un-phosphorylated PMSK1 (PMSK1-U; Fig. 3c), reducing the relative abundance of PMSK1-P within the total PMSK1 pool. This shift toward PMSK1-U decreases the proportion of PMSK1-P, leading to reduced GAP1 mRNA levels (Fig. 4a). In the pmsk1 mutant, GAP1 mRNA levels (Fig. 5a) are similar to those in WT/ PMSK1-S120A (unphosphorylated form; Fig. 4a), showing lower expression at 12 h. However, there is still regulation in red light, which is abolished in the phot pmsk1 double mutant (Fig. 5a), suggesting the existence of other regulators of GAP1 that are controlled by blue light independently of PMSK1, although PMSK1 appears to play a major role. Our model (Fig. 6), which suggests that PMSK1-P acts as an activator of GAP1 mRNA accumulation, explains why the PMSK1 mutation has the greatest impact on GAP1 mRNA levels in red light (Fig. 5a). In red light, PMSK1-P is most abundant in WT, making the mutation’s PMSK1 585 aa (ca. 63.9 kDa) 3670 bp S120 ADGVSpSPHELTR Kinase domain 500bp P U ATPB WT 0’ -P control WT phot 0’ 5’ 10’ 20’ 0’ 5’ 10’ 20’ P U phot 0’ -P control PMSK1FLAG ATPB Phospho-level (log2) 50 75 50 75 kDa a b d c PMSK1FLAG WT D WT BL phot D phot BL 7 8 9 10 11 ✱✱✱ ns -P control 0’ 20’ PMSK1 0’ 20’ S120A 0’ 20’ S120D PMSK1FLAG ATPB P U 50 75 kDa Fig. 3 | Identification of PMSK1 as a key protein on PHOT-dependent starch metabolism in Chlamydomonas reinhardtii. aChanges in the phosphorylation level of S120 of PMSK1 in response to BL in WT and phot,quantified by phosphoproteomics. Samples were collected after 24h of acclimation to darkness and 5 min after the start of blue light. D dark, BL blue light. Data are presented as log2transformed phospho-levels (n = 5 biologically independent samples). Error bars indicate mean±SEM. Box plots represent the median (center line), interquartile range (bounds of the box), minima, and maxima (whiskers) with individual data points overlaid. The statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons tests. Asterisks indicated the p-values (****p < 0.0001; ns not significant). bUpper: Genomic structures of PMSK1. Black boxes and lines indicate exons and introns, respectively. Lower: Schematic structures of PMSK1. Red lines indicate the phosphorylated residue in PMSK1. The orange box indicates the kinase domain. cPhosphorylation level of PMSK1-FLAG as afunctionoftime inWT/PMSK1-FLAG and phot/PMSK1-FLAG, exposedtoblueorred light after 24 h acclimation to darkness (indicated as t = 0 in the graph). Detection was performed by Phos-tag SDS-PAGE; ATPB was used as a loading control. Phosphatase-treated WT/PMSK1-FLAG and phot/PMSK1-FLAG samples were also loaded on the gels. A representative experiment from at least three independent experiments is shown. dChanges in the phosphorylation level of PMSK1-FLAG in WT/PMSK1-FLAG, WT/PMSK1S120A-FLAG and WT/PMSK1S120D-FLAG lines. Samples were collected after24 h ofacclimation todarkness (t = 0′)and20′afterexposuretoblue light (100 μmol photons m−2s−1). Detection was performed by Phos-tag SDS-PAGE; ATPB was used as a loading control. Phosphatase-treated WT/PMSK1-FLAG sample was also loaded on the gels. “U”and “P”indicate the unphosphorylated and phosphorylated PMSK1-FLAG respectively. The blue or red bar to the left of the immunoblots indicates the light quality used. A representative experiment from at least three independent experiments is shown. Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 6 effect more pronounced. In contrast, in white or blue light, PHOT promotes PMSK1 dephosphorylation, reducing PMSK1-P levels and diminishing the mutation’s impact in these conditions. Discussion In this work we report the discovery of an unrecognized lightsignalling pathway linking blue-light perception by PHOT with starch accumulation in Chlamydomonas. Our data solve a four-decades-long question about the reasons for starch accumulation in green algae predominantly under red light31. We showed that PHOT regulates the phosphorylation of a specificserineresidue(S120)onayetuncharacterized kinase, PMSK1. In its phosphorylated state, PMSK1 transduces a signal that controls the accumulation of GAP1 mRNA (Fig. 6). Since GAP1 is involved in the generation of phosphorylated sugars, as precursors of starch synthesis, our findings demonstrate how this light-signalling pathway leads to improvement of starch metabolism. Our work establishes PMSK1 as a kinase involved in PHOTdependent signaling in algae, highlighting its distinct role in this pathway. Yet, PHOT may not be the only blue-light responsive protein suppressing starch accumulation in Chlamydomonas; as our data show, application of red light in the phot mutant results in higher accumulation of starch compared to white or blue illuminated cells (Fig. 1d), suggesting the presence of additional blue-light responsive protein(s) repressing starch accumulation, as illustrated in our model (Fig. 6). How PHOT, a kinase, is involved in the dephosphorylation of PMSK1 is an intriguing question that needs further investigation. One of the possibilities is that a protein phosphatase (indicated PPase in Fig. 6)isa missing component in our proposed model; PHOT may mediate the activity of this PPase to dephosphorylate S120 of PMSK1. In Arabidopsis, which encodes two phototropins, Phot1 and Phot2, blue light induces the phosphorylation of NPH3 (NON-PHOTOTROPIC HYPOCOTYL3) at serine 744 (S744) in a Phot1-dependent manner. This phosphorylation creates a 14-3-3 binding site enabling NPH3 to associate with 14-3-3 proteins. Subsequently NPH3 gets dephosphorylated32,33.Whether Chlamydomonas PHOT operates via a similar mechanism—interacting with and phosphorylating PMSK1 prior to its dephosphorylation at S120—remains to be determined. PMSK1 is found to belong to a family of Serine/threonine-protein kinases (named PMSK-like family, see material and methods) which is conserved in green algae and vascular plants. When searching for PMSK1-like sequences in the Arabidopsis genome in the NCBI database34,HIGHLEAFTEMPERATURE1(HT1) 35 and CONVERGENCE OF BLUE LIGHT AND CO2 1/2 (CBC1/2)36 are found in addition to the PMSK-like Arabidopsis members (Supplementary Figs. 21–24 and Supplementary Text), proteins that have been shown to mediate responses to CO 2 and blue light in Arabidopsis. Yet, while Arabidopsis CBC1/2/HT1 acts to stimulate stomatal opening by inhibiting S-type anion channels36,ourfindings demonstrate that the regulatory function of its Chlamydomonas counterpart PMSK1 controls starch metabolism through the transcriptional regulation of GAP1. WT phot WT/PMSK1 -FLAG S120A WT/PMSK1 -FLAG S120D phot/PMSK1 -FLAG S120A phot/PMSK1 -FLAG S120D WT/PMSK1-FLAG phot/PMSK1-FLAG ns ✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ns ns ns ns ns ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱ 10 -2 10 -1 10 0 10 1 10 2 10 3 Rel. GAP1 mRNA abundance 10 -2 10 -1 10 0 10 1 10 2 10 3 Rel. GAP1 mRNA abundance WT phot WT/PMSK1-FLAG 0 10 20 30 Starch content (μg/million cells) 0 6 12 18 24 0 6 12 18 24 0 6 12 18 24 0 6 12 18 24 WT/PMSK1 -FLAG S120A WT/PMSK1 -FLAG S120D phot/PMSK1-FLAG phot/PMSK1 -FLAG S120A phot/PMSK1 -FLAG S120D a b Time (h) Fig. 4 | PMSK1 phosphorylation status controls starch metabolism in Chlamydomonas reinhardtii. GAP1 transcription level (a) and starch content (b)of various PMSK1 overexpression lines synchronized to a 12/12 light dark cycle under different light qualities. Data are presented as mean ± SD (n = 3 biologically independent samples).Phases are indicated bywhite and gray shadingin(a). Line colors indicated light qualities. Black, white light; Red, red light; Blue, blue light. In bred triangles indicate sample collection time. The dark phase is indicated by black bars above the graphs;the light phase by white, blue, or red bars, depending onthe light quality used. The statistical significance was determined using two-way ANOVA with Dunnett’s multiple comparisons tests (a) of log10 transformed mRNA data as indicated in the graphs. The statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons tests (b). Asterisks indicated the p-values compare to WT (*p< 0.05; ***p < 0.001; ****p< 0.0001; ns, not significant). In some cases, the error bars are smaller than the data point symbols. Detailed statistical analyses are presented in the Source Data File. Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 7 In Arabidopsis, phototropins (Phot1 and Phot2) play a crucial role inregulating starchdegradationinguardcellsinresponsetoblue light, a process essential for stomatal opening, whichmanages gas exchange and water balance in plants. Upon exposure to blue light, phototropin signaling triggers a pathway that leads to the rapid breakdown of starch in guard cells. This involves the coordinated action of the enzymes β-amylase 1 (BAM1) and α-amylase 3 (AMY3). The starch degradation is linked to the activation of the plasma membrane H +-ATPase, which promotes stomatal opening and contributes to overall plant growth20.Ourfindings reveal yet another fascinating differencebetweenvascularplants andChlamydomonas phototropins; in contrast to the Arabidopsis Phots, Chlamydomonas PHOT acts on the mRNA level of a key metabolic enzyme GAP1 to modulate cellular accumulation of starch. In the green microalgae Chlorella, a possible explanation for why blue light represses starch accumulation22 may lie in early studies showing that under blue light carbohydrate catabolism is enhanced37,38.WhilethespecificroleofGAP1inChlorella remains to be investigated, we recently demonstrated that under red light, Chlorella accumulated significant starch levels even after a prolonged 7-day exposure39, in accordance to earlier short-term experiments22.In Chlamydomonas, our results showed that both starch synthesisand catabolism-related genes were upregulated in the phot mutant (Supplementary Fig. 9). Although proteomics does not necessarily capture Starch content (μg/million cells) a b WT pmsk1 phot phot pmsk1 WT pmsk1 phot phot pmsk1 WT pmsk1 phot phot pmsk1 0 10 20 30 ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱ ✱✱ ✱✱ ✱ ns 10 -2 10 -1 10 0 10 1 10 2 10 3 Rel. GAP1 mRNA abundanceRel. GAP1 mRNA abundance WT pmsk1 10 -2 10 -1 10 0 10 1 10 2 10 3 phot 0 6 12 18 24 0 6 12 18 24 phot pmsk1 Time (h) Fig. 5 | PHOT regulates starch metabolism via PMSK1 and GAP1 in Chlamydomonas reinhardtii. aGAP1 transcription level and bstarch content in single and double phot and pmsk1 mutants synchronized to a 12/12 light dark cycle under different light qualities. Data are represented as mean± SD (n = 3 biologically independent samples). Line colors indicated light qualities. Black, white light; Red, red light; Blue, blue light. Red triangles indicate sample collection time. Dark phase is indicated by black bars above the graphs; the light phase by white, blue, or red bars, depending on the light quality used. Red triangles indicated samples collection time. The statistical significance was determined using two-way ANOVA with Dunnett’s multiple comparisons tests (a) of log10 transformed mRNA data as indicated in the graphs. The statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons tests (b). Asterisks indicated the p-values compare to WT (*p < 0.05; **p <0.01; ***p< 0.001; ****p < 0.0001; ns, not significant). In some cases, the error bars are smaller than the data point symbols. Detailed statistical analyses are presented in the Source Data File. Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 8 enzyme activity, it is worth mentioning that our whole-cell proteomics data (Supplementary Data 2) did not reveal any statistically significant changes in the levels of alpha-amylase, the enzyme responsible for the major hydrolytic activity involved in starch degradation in Chlamydomonas40. In summary, our findings not only demonstrate that the PHOTPMSK1 pathway regulates starch metabolism in Chlamydomonas,but they also shed light on the importance of this mechanism within the context of the organism’s diurnal rhythm. The regulatory effect of PHOT-PMSK1 plays a crucial role in modulating GAP1 mRNA levels, particularly at the onset of illumination following the dark phase. By acting through blue light, PHOT-PMSK1 fine-tunes the induction of GAP1, preventing an immediate spike in its expression. This controlled response ensures that starch synthesis does not overwhelm the cell’s energy reserves, allowing for a balanced distribution of resources between starch production and other essential cellular functions. This regulation likely prevents excessive starch accumulation, which could otherwise compromise cellular health, highlighting the critical nature of this finely tuned-mechanism. Nonetheless, the higher starch accumulation in the phot mutant under the low light intensity conditions used in our experiments, does not affect growth and photosynthesis (Supplementary Fig. 5). This finding indicates that this PHOT-mediated light-signaling pathway overcomes a key resource allocation trade-off present in Chlamydomonas, whereby carbon, fixed by photosynthesis, is allocated to energy reserves (e.g. starch) at the cost of growth14.Ourfindings pave the way for the application of precise kinase engineering of PMSK1 as a sustainable way to produce starch from green microalgae in biotechnological applications. Methods Statistics Statistical methods were not used to predetermine the sample size. The experiments were not randomized, and the investigators were not blinded to allocation during experimental procedures and data assessment. Algal Material The strains used in this study included Chlamydomonas phot (defective in PHOT; gene ID: Cre03.g199000) and phot-C1 (phot strain complemented with WT PHOT gene), as well as their background strain CC-125, which have been previously described25. Additionally, Chlamydomonas acry24 (defective in animal-type cryptochrome, aka aCRY;geneID:Cre06.g278251),pcry (defective in plant-type cryptochrome, aka pCRY; gene ID: Cre06.g295200), and acrypcry (defective in both animal-type and plant-type cryptochrome) were generated through CRISPR-CAS9 provided by following the protocol described in ref. 24.Thepmsk1 (defective in phototropin-mediated signaling kinase 1, aka PMSK1; gene ID: Cre16.g659400) and photpmsk1 (defective in both PHOT and PMSK1) mutants were generated using insertional CRISPR-Cas9 RNP method described by Kim et al.41 with a few modifications. The target sgRNA sequence of PMSK1 was Fig. 6 | Proposed model depicting the mechanisms of blue-light dependent regulation of starch metabolism in Chlamydomonas reinhardtii. Phosphorylated PMSK1 at S120 (PMSK1-P) activates GAP1, leading to increased starch biosynthesis. In the presence of blue light, PHOTOTROPIN detects the signal and initiates the dephosphorylation of PMSK1,resulting in decreasedGAP1 mRNA levels and reduced starch accumulation. PPase is a hypothetical phosphatase that could dephosphorylate PMSK1-P in a PHOTOTROPIN-dependent manner. The model also includes a PHOTOTROPIN-dependent but PMSK1-independent repression of GAP1, as GAP1 in pmsk1 mutants remains regulated by red light—a regulation lost in the phot pmsk1 double mutant (Fig. 5a). Additionally, other blue-light-responsive proteins are proposed to independentlyinhibit starchaccumulationor GAP1 outside of the PHOT and GAP1 pathways. This is based on observations that: (i) phot mutants accumulate more starch under red light (Fig. 1d), despite consistently high GAP1 mRNA levels across light conditions (Fig. 2d), and (ii) red light enhances GAP1 mRNA in the phot-kin strain (Fig. 2g). Article https://doi.org/10.1038/s41467-025-57809-3 Nature Communications | (2025) 16:2545 9