Multiomics integration unveils photoperiodic plasticity in the molecular rhythms of marine phytoplankton

The Plant Cell, 2025, 37, koaf033 https://doi.org/10.1093/plcell/koaf033 Advance access publication 11 February 2025 Large-Scale Biology Multiomics integration unveils photoperiodic plasticity in the molecular rhythms of marine phytoplankton Ana B. Romero-Losada, 1,2,† Christina Arvanitidou(Χριστίνα Αρβανιτίδου), 1,2,† M. Elena García-Gómez, 1 María Morales-Pineda, 1 M. José Castro-Pérez, 3 Yen Peng Chew, 4 Gerben van Ooijen, 4 Mercedes García-González, 1, * Francisco J. Romero-Campero 1,2, * 1 Institute for Plant Biochemistry and Photosynthesis, Universidad de Sevilla – Consejo Superior de Investigaciones Científicas, Av. Américo Vespucio 49, Seville 41092, Spain 2 Department of Computer Science and Artificial Intelligence, Universidad de Sevilla, Av. Reina Mercedes s/n, Seville 41012, Spain 3 Institute for Biomedicine in Seville, Universidad de Sevilla – Consejo Superior de Investigaciones Científicas, Av. Manuel Siurot s/n, Seville 41012, Spain 4 School of Biological Sciences, University of Edinburgh, Max Born Crescent, Edinburgh EH9 3BF, UK *Author for correspondence: [email protected] (F.J.R.-C.), [email protected] (M.G.-G.) † These authors contributed equally and both should be considered as first authors. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plcell/pages/General-Instructions) is: Francisco J. Romero-Campero ([email protected]). Abstract Earth’s tilted rotation and translation around the Sun produce pervasive rhythms on our planet, giving rise to photoperiodic changes in diel cycles. Although marine phytoplankton plays a key role in ecosystems, multiomics analysis of its responses to these periodic environmental signals remains largely unexplored. The marine picoalga Ostreococcus tauri was chosen as a model organism due to its cellular and genomic simplicity. Ostreococcus was subjected to different light regimes to investigate its responses to periodic environmental signals: long summer days, short winter days, constant light, and constant dark conditions. Although <5% of the transcriptome maintained oscillations under both constant conditions, 80% presented diel rhythmicity. A drastic reduction in diel rhythmicity was observed at the proteome level, with 39% of the detected proteins oscillating. Photoperiod-specific rhythms were identified for key physiological processes such as the cell cycle, photosynthesis, carotenoid biosynthesis, starch accumulation, and nitrate assimilation. In this study, a photoperiodic plastic global orchestration among transcriptome, proteome, and physiological dynamics was characterized to identify photoperiod-specific temporal offsets between the timing of transcripts, proteins, and physiological responses. Received June 07, 2024. Accepted February 3, 2025 © The Author(s) 2025. Published by Oxford University Press on behalf of American Society of Plant Biologists. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Introduction Marine phytoplankton plays a pivotal role in Earth’s ecosystems, acting as primary producers by contributing to ∼45% of global photosynthetic net primary production (Field et al. 1998). Consequently, it not only sustains the existence of most oceanic life but also supports life across the entire planet. Light availability has a profound impact on marine phytoplankton growth and physiology (Edwards et al. 2015). Earth’s rotation produces the most pervasive rhythmic environmental signal, giving rise to alternating cycles of days (light periods or photoperiods) and nights (dark periods or skotoperiods), collectively known as diel cycles. Earth’s tilted rotational axis and its translation around the Sun also lead to photoperiodic variations in diel cycles, resulting in long days (LDs) in summer and short days (SDs) in winter in temperate zones, the regions between the tropics and the polar circles. Seasonality plays a central regulatory role in marine phytoplankton dynamics (Bolaños et al. 2020; Chen et al. 2021; Mondal and Banerjee 2022). However, the different molecular rhythms underpinning these responses are not yet characterized. Chronobiology is a multidisciplinary field that focuses on the study of the timing and regulation of biological rhythms and their synchronization with environmental cycles (Kuhlman et al. 2018). Specifically, the autonomous oscillating molecular systems that have evolved to anticipate and respond to diel cycles are referred to as circadian clocks. These systems are entrained to external inputs, diel cycles, and produce rhythmic output of biological processes. Circadian rhythms are self-sustained, maintaining rhythmicity of about 24 h under constant conditions. To distinguish these biological rhythms from diel rhythms that respond to the environment, chronobiology experiments are typically designed as a sequence of several consecutive days under alternating light/dark cycles, followed by several consecutive days of constant light (LL) or dark (DD), called free-running conditions (Kuhlman et al. 2018). Extensive chronobiological studies on circadian and diel rhythms have been conducted at the molecular and physiological levels in many model organisms (Bläsing et al. 2005; Miller et al. 2007; Patke et al. 2020; Ma et al. 2021; Petersen et al. 2022; Häfker et al. 2023) uncovering key molecular mechanisms and regulators controlling circadian rhythms. However, only initial steps have been taken to analyze such systems in marine phytoplankton using omics technologies (Monnier et al. 2010; Annunziata et al. 2019; Kay et al. 2021), which are beginning to characterize molecular rhythms at the transcriptomic and proteomic levels and identify their circadian clock regulators. A recent kingdom-wide comparative analysis of diel gene expression profiles across Archaeplastida has highlighted evolutionary trends in rhythmic patterns across the green lineage (Ferrari et al. 2019). However, this study solely focused on the transcriptomic layer and did not explore the effect of different photoperiods and free-running conditions. Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 To characterize the responses of marine phytoplankton to photoperiodic variations in diel cycles, the model phytoplanktonic picoeukaryote Ostreococcus tauri (Ostreococcus) was selected due to its cellular simplicity (Henderson et al. 2007) and fully sequenced and annotated genome (Derelle et al. 2006; Palenik et al. 2007; Blanc-Mathieu et al. 2014). Moreover, Ostreococcus occupies a key position in the green lineage (Viridiplantae) as a representative of the Class Mamiellophyceae, which diverged early from the group that would give rise to land plants and, therefore, could provide insights into the ancestral traits of green plants and their evolutionary processes (van Baren et al. 2016; de los Reyes et al. 2017; Bachy et al. 2022). Intensive and extensive studies on the physiology of Ostreococcus have provided a solid foundation for systems biology and omics studies. Chronobiological studies (Corellou et al. 2009; Monnier et al. 2010; O’-Neill et al. 2011; van Ooijen et al. 2011; Freeney et al. 2016) have uncovered the fundamental components of the Ostreococcus circadian clock revealing key rhythmic processes. Cell-cycle analysis (Corellou et al. 2005; Robbens et al. 2005; Gil-Rodríguez et al. 2024) have provided insights into the molecular components regulating cell-cycle progression in Ostreococcus under different light regimes while photobiology approaches (Sands et al. 2023) have highlighted the role of light quality in different biological processes. Research into sexual reproduction (Benites et al. 2021), viral infections dynamics (Derelle et al. 2018; Castillo et al. 2021), and biomass composition (Ral et al. 2004; Degraeve-Guilbault et al. 2017, 2020; Guyon et al. 2018) have expanded the knowledge of Ostreococcus’ potential strategies to cope with biotic and nutritional stresses. Finally, omics analysis (Monnier et al. 2010; Le Bihan et al. 2011, 2015; Henríquez-Castillo et al. 2018; Bousquet et al. 2020; Kay et al. 2021) have provided global characterizations of the transcriptome and proteome separately. These investigations have unveiled that the Ostreococcus circadian clock is notably simpler than that of other photosynthetic organisms like Arabidopsis thaliana. It has been shown to be constituted only by a reduced central loop comprising potential orthologues of CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) and TIMING OF CAB EXPRESSION 1 (TOC1). In contrast, Arabidopsis has a more complex clock composed of interlocking loops, including a morning loop constituted among others by PSEUDO RESPONSE REGULATORS 9, 7, and 5 (PRR9, PRR7, and PRR5), the central loop consisting in CCA1 and TOC1, and an evening loop involving additional components such as ZEITLUPE and GIGANTEA (Serrano-Bueno et al. 2017). Despite these advances, Ostreococcus responses to photoperiodic variations in diel cycles, as well as free-running conditions, remain to be explored, particularly through multiomics integrative analysis. This study aimed to characterize the Ostreococcus transcriptome, proteome, and physiological rhythmicity, under long and short photoperiods. Additionally, to determine circadian rhythms, our experiments were extended to free-running conditions consisting of LL or DD. Integration of transcriptomic and proteomic data with physiological measurements unveiled a photoperiodic plastic global orchestration between transcriptome, proteome, and physiological dynamics, with photoperiodspecific temporal offsets between the phase of transcript, protein, and physiological rhythms. Results Transcript rhythmicity under diel cycles and free-running conditions In this study, photochemostats operated in a continuous regime (Supplementary Fig. S1A) were used simulating LD conditions (16 h light:8 h dark) or SD conditions (8 h light:16 h dark). The illuminating system simulated the progressive light intensity increase and decrease during solar daylight cycles with a maximum light irradiance of 1,500 μE m −2 s −1 . Temperature and pH were maintained constant at 20 °C and 8. Samples were collected for 3 consecutive days every 4 h starting at ZT0, Zeitgeber time 0, the time point corresponding to the beginning of the photoperiod simulating dawn (Supplementary Fig. S1B). Robust global rhythmicity was detected in the transcriptomes under LD and SD conditions by using hierarchical clustering (HC) and principal component analysis (PCA) (Supplementary Fig. S1, C to E). A cyclic circular organization of the transcriptomes was revealed over diel cycles (Supplementary Fig. S1, D and E). Photoperiodic variations did not affect transcriptome rhythmicity, as the sets of rhythmic genes under LD and SD entrainment were near-identical, comprising ∼80% of the genome under either condition (Fig. 1A). Arrhythmic genes were either not expressed or lowly expressed. Indeed, rhythmic genes presented significant maximum expression levels 3-fold greater than arrhythmic genes (Fig. 1B). These genes are mainly involved in stress responses such as viral infection, and might not be expressed under these laboratory conditions. In order to distinguish between circadian-regulated genes and those responding to the rhythmic environment, cultures were transferred to free-running conditions consisting of LL or DD. No samples were collected during the first day of constant conditions to allow culture acclimation, and samples were collected thereafter to generate transcriptomic data for 2 consecutive days starting at circadian time 0 (CT0), the time point corresponding to subjective dawn (Supplementary Fig. S1B). A clear reduction in rhythmicity was observed under constant conditions: ∼21% of the transcriptome was completely reliant on diel cycles to maintain rhythmicity since these genes lost their rhythms under both LL and DD (Fig. 1, C and E; Supplementary Table S1), independently from the entrainment regime. In contrast, the transcriptome proportion that maintained oscillations exclusively under LL or DD, but not both, was dependent on the previous entrainment. Whereas 15% kept cycling under LL after LD entrainment, only 6% were rhythmic after SD entrainment. The oscillating genes under LL, both after LD and SD entrainment, were found to be significantly involved in DNA replication and chromosome organization (Fig. 1D; Supplementary Fig. S2). The detrimental effect of DD over transcriptome rhythmicity was smaller than that of LL, with almost 22% maintaining oscillations after LD entrainment and 39% after SD entrainment. This observation is remarkable, as previous studies have found a complete lack of transcription under DD (O’-Neill et al. 2011). This can be attributed to the growth conditions in our experiments, which allow Ostreococcus cells to accumulate substantial amounts of starch to be used as energy source under DD to maintain oscillations (see section on starch content analysis). The biological processes RNA processing and ribosome biogenesis were found to be significantly enriched among the rhythmic genes under DD, both after LD and SD entrainment (Fig. 1F; Supplementary Fig. S3). Orthologues of the central components of the circadian clock, ostta06g02340 (CCA1) and ostta13g01820 (TOC1), exhibited rhythmicity under both LL and DD free-running conditions. CCA1 presented higher levels of expression under DD, in contrast to the halt in oscillations reported previously under different growth conditions (O’-Neill et al. 2011), while TOC1 showed higher expression levels under LL (Supplementary Fig. S4). Orthologues of the light receptors, ostta15g0100 (CRYPTOCHROME 1, CRY1), ostta01g06470 (CRYPTOCHROME 3, CRY3), and ostta03g05620 (UV RESISTANCE 2, UVR2), were rhythmic under both LL and DD 2 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 free-running conditions, with higher expression levels under LL than DD. Their rhythmic expression profiles were affected by photoperiodic variations except CRY3, which presented an identical pattern under both LD and SD but with an additional peak of expression at the beginning of the photoperiod under LD. This suggests a central position of these genes at the regulatory core of the circadian clock responding to photoperiodic variations in Ostreococcus (Supplementary Fig. S4). Genes presenting rhythmicity under both LD and SD as well as maintaining their rhythmic expression profiles under both freerunning conditions were considered in this study as those more directly regulated by the circadian clock in Ostreococcus. A clear dependence on the previous entrainment regime was observed, with 1,647 genes keeping rhythmicity under both free-running conditions after LD entrainment, versus 1,034 following SD entrainment. These 2 sets overlapped partially, identifying 350 genes Figure 1. Transcriptome rhythmicity under alternating light/dark cycles and identification of circadian genes including free-running conditions. A) Venn diagram comparing rhythmic genes under LD (blue circle) and SD (light red circle). B) Boxplot representing the maximum expression level of rhythmic and arrhythmic genes. Medians are represented by central horizontal lines, upper and lower quartiles by boxes, minimum and maximum values by whisker ends. Gene expression levels are measured as FPKM. Significance was computed according to Mann–Whitney–Wilcoxon nonparametric test. C) Bar plot representing with blue colors different rhythmic gene sets under LD conditions. D) Gene expression profiles during 3 consecutive days under LD, 2 consecutive days under LL, and 2 consecutive days under DD of DNA polymerase alpha subunit B (ostta11g01400, POLA2). White rectangles represent photoperiods (light periods or days), blue-filled rectangles correspond to skotoperiods (dark periods or nights) under LD, light blue rectangles mark subjective nights or days under LL and DD, respectively after LD entrainment. ZTN, Zeitgeber time N, marks the time point N hours after dawn (lights on, ZT0). CTN, circadian time N denotes the time point N hours after subjective dawn. A discontinuity is shown on the time axis to indicate that samples were collected after 24 h acclimation to the corresponding free-running conditions. E) Bar plot representing with red colors different rhythmic gene sets under SD conditions. F) Gene expression profiles during 3 consecutive days under SD and 2 consecutive days under LL and DD of Ribosomal protein L7Ae (ostta02g01210, RPL7AE). White rectangles represent photoperiods, red-filled rectangles correspond to skotoperiods under SD, light red rectangles mark subjective nights or days under LL and DD, respectively, after SD entrainment. G) Venn diagram comparing circadian genes identified after LD entrainment (blue circle) and after SD entrainment (red circle). H) Treemap summarizing the biological processes significantly enriched over rhythmic genes under LD, SD, LL, and DD. Rectangle sizes represent significance levels. I) Gene expression profiles during 3 consecutive days under LD and SD; and 2 consecutive days under LL and DD of RuBisCO Activase (ostta04g02510) exemplifying rhythmic genes under LD, SD, LL, and DD involved in photosynthesis related processes. Photoperiodic plasticity in marine phytoplankton rhythms | 3 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 corresponding to 4.6% of the Ostreococcus transcriptome (Fig. 1G; Supplementary Table S1). Functional enrichment analysis revealed that these genes are significantly involved in photosynthesis, chloroplast organization, and pigment metabolic processes (Fig. 1, H and I). The specific circadian genes detected after LD entrainment are significantly involved in ribosome biogenesis whereas those found after SD entrainment are primarily associated with cell cycle. This indicates a strong influence of the circadian clock over these processes. However, genes identified as circadian after LD entrainment lose rhythmicity under LL after SD entrainment, and the reverse for the specific circadian genes after SD entrainment that lose rhythmicity under DD after LD entrainment. This suggests a strong detrimental impact on rhythmicity of LL after SD entrainment with a short photoperiod, as well as constant darkness after LD entrainment with a short dark period (Supplementary Fig. S5). In LD-entrained cultures, significant reductions in amplitude were detected under both LL and DD conditions for the LD/LL and LD/DD rhythmic genes, respectively (Fig. 2A). In SD-entrained cultures, amplitude reduction was only noted when transferred to LL for SD/LL rhythmic genes but no substantial change was found when transferred to DD for SD/DD rhythmic genes (Fig. 2D). Free-running conditions also affected the phase, or time point of maximum expression level, in rhythmicity patterns. Independently from the previous entrainment regime, phases were delayed when cultures were transferred to LL (Fig. 2, B and C, E and F, left). Whereas advanced phases (i.e. phases occurring earlier) were found when cultures were transferred to DD (Fig. 2, B and C, E and F, right), with this effect more pronounced for LD-entrained cultures and only slight for SD-entrained cultures. The phase delays in LL were more drastic for SD than LD-entrained cultures although the advances in DD were more evident for LD than SD-entrained cultures. Effects of photoperiodic variations over gene expression profiles Differences were identified in phase and amplitude of rhythmic gene expression patterns between LD and SD (Fig. 3A) with only 31 genes exhibiting the same expression profile under both photoperiods. A significant reduction in overall amplitude was observed in SD compared to LD-entrained cultures (Fig. 3B). Under both LD and SD, most genes peaked at night (Fig. 3C), indicating a “nocturnal transcriptome profile. Advanced phases were noted under SD with respect to LD conditions. Under LD entrainment, peak phases occur uniformly from the end of the day (ZT12) to the end of the night (ZT20), while under SD-entrainment peak phases predominantly occur during the first half of the night (ZT8 to ZT16; Fig. 3C). Gene clusters were defined based on peak phases (Supplementary Table S2). Distinct diel timings of maximum gene expression levels for specific biological processes were identified by performing functional enrichment analysis over each gene cluster under LD (Fig. 3D; Supplementary Fig. S6) or SD conditions (Fig. 3E; Supplementary Fig. S7). Changes in the phase of biological processes under LD and SD conditions were detected. These differences in diel timings were produced by the previously mentioned advanced gene phases under SD with respect to LD. For example, genes involved in photosynthesis peak at ZT8 under LD and at ZT4 under SD, midday coinciding with the moment of maximal light irradiance in both conditions. Similarly, genes involved in DNA replication reached maximum expression level at late day (ZT12) under LD, and at early night (ZT8) under SD. Rearrangements in the order of peak phases for biological processes were also found. For instance, under LD conditions, genes involved in amino acid biosynthesis, ribosome biogenesis, and translation reached their highest level at ZT20, ZT0, and ZT4, respectively. However, in SD conditions, ribosome biogenesis genes were upregulated at ZT12, while both amino acid biosynthesis and translation-related genes reached maximum expression levels at ZT16. To identify potential regulators of these distinct diel timings, transcription factor-binding sites (TFBS) enrichment analysis was performed over the promoters of genes in each time point cluster (Fig. 4). DNA motifs recognized by the families of plant transcription factors myeloblastosis (MYB), DNA-binding with one finger (DOF), basic leucine zipper (bZIP), cysteine-rich polycomb-like protein (CPP), and Homeobox were identified, suggesting a key role in regulating diel rhythmic gene expression patterns. Gene clusters corresponding to different time points under LD and SD conditions exhibited enrichment for the same DNA motifs, as these clusters share many genes in common due to the advanced gene phases under SD compared to LD. For instance, gene clusters corresponding to ZT12 under LD and ZT8 under SD showed enrichment for the DNA sequence recognized by the CPP transcription factor family. This family has been implicated in cell proliferation in Arabidopsis (Wang et al. 2018). Consistent with these findings the genes ostta20g00800 and ostta09g04220, encoding CPP transcription factors, peaked at ZT12 under LD and ZT8 under SD conditions. Notably, the evening element (EE) motif, AAATATCT, was significantly identified in the promoters of the genes peaking at ZT12 under LD conditions and ZT0 and ZT4 under SD conditions. In Arabidopsis, the EE is associated with the binding site of CCA1 (Harmer et al. 2000). This transcription factor binds to its targets in the morning to repress their expression, resulting in maximum expression later in the evening (Kamioka et al. 2016). A similar gene expression pattern is observed in our transcriptomic data from Ostreococcus, suggesting a strong conservation at the central core of the circadian clock between Arabidopsis and Ostreococcus. Based on the presence of the EE in the gene promoters discussed above, CCA1 Ostreococcus orthologue, ostta06g02340, would be expected to bind in the morning to genes peaking at ZT12 under LD or late in the night to genes peaking at ZT0 or ZT4 under SD. To capture the previously described changes in the phase and amplitude of rhythmic gene expression profiles resulting from photoperiodic variations, a co-sinusoidal dynamical model was developed. This model predicts the phase and amplitude of each rhythmic gene for a specific day of the year using linear interpolation between the phase and amplitude identified under LD or SD entrainment (Video 1). Previously published microarray data, generated under neutral day conditions (ND) (12 h light:12 h dark) (Monnier et al. 2010) were used to evaluate its predictive power. The phases of approximately two-thirds (63%) of the rhythmic genes were successfully predicted with ±4 h error. The limited predictive power of our model suggests that while two-thirds of the rhythmic genes respond to changes in photoperiod by gradually adjusting their phases, the remaining one-third exhibit a more complex response to photoperiod changes. Another difference observed between LD and SD conditions was related to the occurrence of gene expression profiles with 2 daily peaks (Fig. 3F). Although gene profiles with 2 peaks were detected in both LD and SD cultures (Supplementary Table S3), a drastic increase in the number of genes with bimodal rhythmicity was observed under SD condition (1,855 versus 376 genes; Fig. 3G). Bimodal rhythmicity was not maintained under free-running conditions in which distinct single-peak profiles were apparent under 4 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 LL and DD (Fig. 3H). Nonlinear squares (Baty et al. 2015) were applied to decompose the observed bimodal gene profile in SD-entrained cultures into 2 different single-peak profiles, one peaking during the photoperiod and the other during the skotoperiod. Under LL free-running condition, only the photoperiod-peaking profile was maintained, whereas under DD condition only the skotoperiod-peaking profile was present. For genes with bimodal rhythmicity, a dynamical model was developed combining 2 distinct co-sinusoidal profiles to capture changes in the phase and amplitude, resulting from changes in photoperiod length. Under LD conditions, these profiles overlap in time producing a singlepeak profile, whereas under SD condition they become out of phase, resulting in a bimodal profile (Video 2). To test this model, predictions were run to assess rhythmicity profiles under ND Figure 2. Free-running conditions effects over gene expression profiles. A) Boxplot representing rhythmic genes amplitude reached under LD conditions, when cultures were transferred to free-running conditions consisting of LL (light blue) and DD (dark blue) after LD entrainment. Medians are represented by central horizontal lines, upper and lower quartiles by boxes, minimum and maximum values by whisker ends. Gene expression levels are measured as FPKM. Significance was computed according to Mann–Whitney–Wilcoxon nonparametric test. B) Histograms showing the distribution of the number of genes exhibiting advanced and delayed phases when cultures are transferred from LD to LL (left) and DD (right). Vertical dashed lines mark no shift. C) Gene expression profiles during 3 consecutive days under LD and 2 consecutive days under LL and DD of sedoheptulose-bisphosphatase (ostta03g05500, SBPase). Vertical black arrows mark LD phases, vertical gray arrows mark LL and DD phases and horizontal black arrows represent delayed or advanced phases. White rectangles represent photoperiods (light periods or days), blue-filled rectangles correspond to skotoperiods (dark periods or nights) under LD, light blue rectangles mark subjective nights or days under LL and DD, respectively after LD entrainment. ZTN, Zeitgeber time N, marks the time point N hours after dawn (lights on, ZT0). CTN, circadian time N, denotes the time point N hours after subjective dawn. A discontinuity is shown on the time axis to indicate that samples were collected after 24 h acclimation to the corresponding free-running conditions. D) Boxplot representing rhythmic genes amplitude or maximum expression level reached under SD conditions, when cultures were transferred to LL and DD free-running conditions after SD entrainment. Significance was computed according to Mann–Whitney–Wilcoxon nonparametric test. E) Histograms showing the distribution of the number of genes exhibiting advanced and delayed phases when cultures are transferred from SD to free-running conditions consisting in LL (left) and DD (right). Vertical dashed lines mark no shift. F) Gene expression profiles during 3 consecutive days under LD and 2 consecutive days under LL and DD of SBPase. Vertical black arrows mark SD phases, vertical gray arrows mark LL and DD phases, and horizontal black arrows represent delayed phases. Photoperiodic plasticity in marine phytoplankton rhythms | 5 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Figure 3. Photoperiodic effects over gene expression profiles. A) Gene expression profiles during 3 consecutive days under LD (blue) and SD (red) conditions of CYCB (ostta01g06150, top) and delta-9 acyl-lipid desaturase 1 (ostta01g00790, ADS1, bottom). Blue and red vertical dotted arrows mark LD and SD phases. Horizontal black arrow represents advanced phase under SD when compared with LD. Blue and red horizontal dashed lines mark LD and SD amplitudes. Vertical black arrows represent the reductions in amplitude under SD with respect to LD. Light regimes are represented as described in Fig. 1. B) Boxplot representing rhythmic genes amplitude under LD in blue and SD in red. Medians are represented by central horizontal lines, upper and lower quartiles by boxes, minimum and maximum values by whisker ends. Gene expression levels are measured as FPKM. Significance was computed according to Mann–Whitney–Wilcoxon nonparametric test. C) Histograms showing the distribution of the number of genes with phase at specific time points during the day under LD (blue, top) and SD conditions (red, bottom). ZTN, Zeitgeber time N, marks the time point N hours after dawn (lights on). D) Circular heatmap representing the temporal organization of gene expression profiles under LD conditions. Dark blue stands for low expression, whereas yellow represents high expression. Genes are clustered depending on their phases. Genes with phase at ZT0 are located in the outer circle, while genes with phase at subsequent zeitgeber (ZT) are placed sequentially into inner circles. Biological processes enriched in the gene set with phase at each specific time point are depicted capturing the transcriptional program over diel cycles under LD condition. E) Similarly, circular heatmap representing the temporal organization of the transcriptional program under SD condition. F) Gene expression profiles during 3 consecutive days under LD (blue line) and SD (red line) of Ferredoxin-NADP + reductase (ostta18g01250, FNR). G) Bar plots representing in different green colors from top to bottom are the number of arrhythmic, single-peak rhythmic, and 2 peaks rhythmic genes under LD and SD conditions. H) Gene expression profiles during 3 consecutive days under SD and 2 consecutive days under free-running conditions consisting of LL or DD of FNR (ostta18g01250). This gene exemplifies how 2 peaks expression patterns under SD conditions could emerge as the combination of 2 distinct rhythmic profiles. One depending on the photoperiod (dotted line) with phase marked with a gray vertical arrow maintaining its rhythmicity only under LL (top). Another expression profile is apparent depending on the skotoperiod (dashed line) with phase marked with a black vertical arrow maintaining its rhythmicity only under DD (bottom). 6 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 conditions. The model accurately predicted the emergence of bimodal rhythmic profiles, as identified in microarray data from ND conditions, validating our model (Supplementary Fig. S8). Proteome rhythmicity under photoperiodic variations and integration with transcriptomic rhythmic patterns As proteins are the primary actors of biological processes, proteomic analyses provide a direct assessment of the functional consequences of environmental rhythms (Kay et al. 2021). In this study, a total of 3,672 proteins were successfully quantified under LD or SD conditions (Supplementary Fig. S9A), accounting for ∼48% of the predicted Ostreococcus proteome (Supplementary Table S4). The predicted subcellular location of these proteins indicates coverage of all subcellular locations (Supplementary Fig. S9B). The estimated number of rhythmic proteins was notably much lower than that of rhythmic genes. Specifically, 928 and 1,442 rhythmic proteins were identified under LD and SD conditions, respectively (Fig. 5A; Supplementary Fig. S9, C and Dand Table S5). Globally, temporal phase offsets were detected between rhythmic protein abundance profiles and the corresponding transcript profiles with gene expression preceding protein abundance several hours (Fig. 5B). A direct relationship between transcript and protein abundance becomes clear upon calculating correlation following phase alignment of transcript and protein profiles Figure 4. TFBS enriched in each time point gene cluster under LD and SD conditions. Photoperiodic plasticity in marine phytoplankton rhythms | 7 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 (Fig. 5, C and D). This suggests that protein and transcript abundance relate linearly but with a clear temporal separation. In LD conditions, the distribution of transcript/protein phase temporal offsets centered around 5 to 6 h, consistent with protein abundance following transcript abundance. However, under SD, offsets followed a more uniform distribution and were significantly longer than under LD (Fig. 5E), indicating that many proteins do not follow on predictably from transcripts. Under both conditions, a “diurnal character was observed for the Ostreococcus proteome with most proteins reaching maximum abundance during the day (Fig. 5F). Notably, this is in contrast with the “nocturnal character observed for the transcriptome (Fig. 3B). Transcript/protein phase offsets result in different diel timings of biological processes based on the time point when the corresponding proteins reach their maximum abundances (Fig. 5G) compared to the diel timings of maximum gene expression levels (Fig. 3, D and E). Transcript/protein phase offsets did not correlate to any biochemical properties computed from protein sequences, such as amino acid composition, charge, or hydrophobicity (Supplementary Fig. S10). Similarly, transcript/protein phase offset did not correlate with transcript phase under LD condition. However, under SD condition transcript/protein offsets were significantly longer for transcripts peaking during the skotoperiod, ZT8, ZT12, ZT16, and ZT20, when compared to those genes with transcript phases during the photoperiod, ZT0 and ZT4 (Fig. 6A). This may be due to translation occurring preferentially during the light period as suggested in our analysis (Fig. 5G) and in previous studies for Arabidopsis (Seaton et al. 2018). Notably, distinct short or long transcript/protein phase offsets were observed between genes involved in different biological processes, as identified by gene ontology (GO) analyses (Fig. 6B). Specifically, biological processes associated with short offsets were DNA replication and photosynthesis. Translation on the other hand was one of the most representative biological processes exhibiting long transcript/protein offsets (Fig. 6, C and D). Proteome and transcriptome rhythmicity combine to orchestrate physiological rhythms Rhythmic patterns have been described in microalgae, including Ostreococcus, for physiological processes such as cell division, photosynthesis, and metabolism (Mittag 2001; Moulager et al. 2007; Sorokina et al. 2011; Shi et al. 2022). In this study, transcriptomic and proteomic data have been integrated with measurements of cell-cycle progression and metabolism to elucidate the temporal orchestration that underlies dynamic physiology. Cell-cycle progression Flow cytometry was used to assess the cell cycle based on total DNA content in Ostreococcus observing rhythmicity in all cell-cycle phases under both LD and SD conditions (Fig. 7A; Supplementary Table S6). A reduction of ∼24% was observed in the number of cells entering S phase under SD compared to LD. In SD conditions, cells remained in either G1 or G2 phase. Furthermore, significant delays of ∼4 h were observed in the cell-cycle phases under LD compared to SD (Fig. 7A). These observations are consistent with the shifts in the time points of maximum transcript and protein abundance found under short versus long photoperiods. For key genes associated with different cell-cycle phases (Robbens et al. 2005), the corresponding phase of transcript and protein abundance under LD and SD conditions were assessed (Supplementary Table S6). Transcript and protein abundance profiles were compared to the cell-cycle phases using violin plots. Clear temporal offsets were observed with gene expression preceding protein abundance levels and shorter temporal offsets between protein abundance and the cell percentage in each cell-cycle phase (Fig. 7B). These gene expression and protein abundance profiles refer to the genes associated with different cell-cycle phases in Supplementary Table S6. Cyclins and cyclin-dependent kinases (CDKs) are essential components of the molecular machinery governing cell-cycle progression (Corellou et al. 2005). Their gene expression and protein abundance profiles were examined and integrated with the temporal cell-cycle phase progression under LD and SD (Fig. 7, C to E). CyclinD (CYCD, ostta18g01570) expression increased from the Video 1. Changes in the phase and amplitude of rhythmic gene expression profiles as a consequence of seasonal photoperiod lengthening/shortening. Animation illustrating how seasonal photoperiod lengthening may result in a gradually delayed phase coupled to a gradual increase in amplitude. Accordingly, seasonal photoperiod shortening may result in a gradually advanced phase coupled to a gradual decrease in amplitude. White rectangles represent photoperiods (light periods or days) and color-filled rectangles correspond to skotoperiods (dark periods or nights). Transitions from red to blue colors and vice versa represent seasonal changes in photoperiods and skotoperiods. Video 2. Emergence of 2 peaks gene expression patterns as a consequence of seasonal photoperiod or day shortening (skotoperiod or night lengthening). Animation illustrating how seasonal day or photoperiod shortening (night or skotoperiod lengthening) may result in the emergence of 2 peaks gene expression patterns as a consequence of the 2 constituent gene expression profiles becoming out of phase. The observed gene expression pattern represented by a continuous thick line can result from the combination of 2 distinct expression profiles represented by thin dotted and dashed lines. One of these expression profiles (dotted line) may not respond to seasonal changes in photoperiod/skotoperiod, whereas the other one (dashed line) may experience phase shifts and amplitude changes. Under LD conditions (alternating 16 h light/8 h dark) the phases or maximum expression-level time point of both expression profiles may coincide resulting in a single-peak expression pattern. Whereas under SD conditions (alternating 8 h light/16 h dark) the phases may be reached at different time points producing a 2 peaks expression pattern. Transitions from red to blue colors and vice versa represent seasonal changes in photoperiods and skotoperiods resulting in gradual transitions from 2 peaks to a single-peak expression pattern. 8 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Figure 5. Proteome rhythmicity under alternating light/dark cycles and temporal offsets with respect to the corresponding transcriptome. A) Bar plots representing the number of identified proteins under LD (top) and under SD (bottom) conditions. The number of rhythmic proteins under LD condition is represented in blue and under SD condition in red. Arrhythmic proteins are represented in white. B) Scaled transcript (light line) and protein (dark line) abundance profiles during 3 consecutive days under LD (top, blue) and SD (bottom, red) conditions for Minichromosome Maintenance 2 (ostta11g00910, MCM2). White rectangles represent photoperiods (light periods or days), blueand red-filled rectangles correspond to skotoperiods (dark periods or nights) under LD and SD, respectively. ZTN, Zeitgeber time N, marks the time point N hours after dawn (lights on, ZT0). C) Boxplots representing the global distribution of the correlations between transcript and protein abundance profiles (white box) and shifted aligned profiles with coincident phases (gray). Medians are represented by central horizontal lines, upper and lower quartiles by boxes, minimum and maximum values by whisker ends. D) Top, protein abundance (continuous blue line) and gene expression (dashed light blue line) profiles during 3 consecutive days under LD condition for MA3 domain-containing translation regulatory factor (ostta05g02330). Bottom, phase-aligned protein abundance (dotted blue line) and gene expression (dashed light blue line) profiles. E) Histograms showing the distribution of the number of proteins with specific offsets between transcript and protein abundance phases (time points of maximum transcript/protein abundance) under LD condition (left, blue) and SD condition (right, red). F) Histograms showing the distribution of the number of proteins with phase or maximum abundance at specific time points under LD condition (left, blue) and SD condition (right, red). G) Temporal organization of biological processes based on the time point when proteins involved in the corresponding process reach its maximum abundance under LD condition, left in blue, and SD condition, right in red. Photoperiodic plasticity in marine phytoplankton rhythms | 9 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 decreasing as light intensity increased, concomitant with an accumulation of both lutein and dihydrolutein (Fig. 10B). The enzymes involved in the interconversion of these carotenoids remain to be identified, and the comparison of their gene expression and protein accumulation was not feasible. Nitrate assimilation In this section, nitrate assimilation in Ostreococcus is examined as an example of a physiological process that exhibited long temporal offsets between transcripts and proteins. Nitrate assimilation is a fundamental process in the metabolism of photosynthetic organisms as it converts a fully oxidized nitrogen form into organic reduced nitrogen, apt for the synthesis of essential biomolecules to allow cell growth and biomass production (Sanz-Luque et al. 2015). To analyze the adaptive response of nitrate assimilation to photoperiodic variations in diel cycles and its implications for optimizing nutrient uptake and metabolism, the transcript and protein abundance profiles corresponding to this pathway in Ostreococcus were analyzed (Fig. 11). Genes in this pathway only maintained rhythmicity under LL being repressed under DD after both LD and SD entrainment. This suggests that, although the circadian clock controls partially these genes, there exists a strong regulation exerted by light over them. Nitrate assimilation starts with the uptake of nitrate mediated by Nitrate Transporters 2 and 3 (NRT2, ostta10g00950 and NRT3, Figure 11. Integration of enzymatic activity profiles with transcript and protein abundances for transporters and enzymes involved in nitrate assimilation under LD and SD conditions. A) Graphical representation of the nitrate assimilation pathway in Ostreococcus. For each transporter and enzyme, its corresponding protein and transcript abundance under LD (top in blue) and SD (bottom in red) conditions are represented using heatmaps. Black represents low, blue medium, and yellow high abundances. White rectangles represent photoperiods (light periods or days), blueand red-filled rectangles correspond to skotoperiods (dark periods or nights) under LD and SD, respectively. B) Mean scaled transcript (dashed line) and protein (continuous line) abundance profiles of 3 consecutive days for NR (left) and GS (right) under LD (top, blue) and SD (bottom, red) conditions. Vertical lines represent SE. Activity profiles for each enzyme under LD and SD conditions are represented using heatmaps, where black corresponds to low activity and yellow to high. 16 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 ostta10g00940), followed by its reduction to nitrite by nitrate reductase (NR, ostta10g00920). Nitrite is further reduced to ammonium by nitrite reductase (NIR, ostta10g00930). The central part of ammonium assimilation is played by the glutamine synthetase (GS, ostta01g05020) and glutamine oxoglutarate aminotransferase (GOGAT, ostta14g01900) cycle, which incorporates inorganic ammonium nitrogen first into glutamine and hence into glutamate, the central precursor for the biosynthesis of all other nitrogen-containing biomolecules (Fig. 11A; Supplementary Fig. S11). Under LD conditions, NRT2/3, NR, and NIR gene expression reached their maximum at dawn (ZT0), while their protein abundances peaked 8 h later at midday (ZT8), coinciding with the time point of maximum irradiance. GS and GOGAT gene expression and protein abundance peaked at the beginning of the day (ZT4) without noticeable temporal offsets. However, a slight increase in protein abundance was detected at the end of the day (ZT12) for both GS and GOGAT (Fig. 11A; Supplementary Fig. S11). In contrast, under SD conditions, all genes encoding the transporters and enzymes involved in nitrate assimilation showed their maximum expression at midnight (ZT12–ZT16), preceding their protein abundance peaks, reached at dawn (ZT0) or midday (ZT4), by 8 h or more. Notably, GOGAT gene expression displayed a bimodal pattern under SD conditions, maintaining the peak observed under LD at ZT4 besides a new peak at ZT16 (Fig. 11A; Supplementary Fig. S11). To validate these temporal offsets between transcripts and proteins, NR and GS activities were measured throughout complete diel cycles under both LD and SD conditions. Enzyme activity profiles were significantly rhythmic (Supplementary Table S9). NR maximal catalytic activity reached its peak at ZT8 under both LD and SD conditions, coinciding with the protein abundance peak under LD and exhibiting a short forward temporal offset under SD. Similarly, GS activity peaked at ZT0 under both LD and SD, concomitant with the protein abundance peaks (Fig. 11B). Discussion Seasonality plays a key role in the natural geographical growth dynamics of marine phytoplankton, including the model picoalga Ostreococcus (Bolaños et al. 2020). Nevertheless, the molecular rhythms underpinning its responses to photoperiodic variations remained to be characterized. In this study, a multiomics approach has been adopted to unravel the plasticity in the orchestration between transcriptome and proteome rhythms governing cyclic physiological responses to changes in photoperiod length. Transcriptomic analysis revealed that photoperiodic variations had no effect over the identity of rhythmic genes, which comprised 80% of the transcriptome. Arrhythmic genes were associated with stress responses and exhibited almost complete repression. These genes could become rhythmic once highly expressed under the corresponding stress condition resulting in a full rhythmic transcriptome. Our results are in agreement with previous studies (Monnier et al. 2010; Zones et al. 2015) indicating that transcriptome rhythmicity in chlorophyte phytoplankton is much higher than in other organisms such as A. thaliana 30% to 50% (Bläsing et al. 2005), Solanum tuberosum 18% to 45% (Hoopes et al. 2022), Drosophila melanogaster 24% (Ma et al. 2021), or Mus musculus 3% to 10% (Miller et al. 2007). Nevertheless, only a small fraction of the transcriptome was rhythmic under LL and DD free-running conditions. These genes were significantly involved in photosynthesis and chloroplast organization, indicating a predominant control by the autonomous circadian clock. Gene rhythmicity is strongly influenced by alternating light/dark cycles in this picoalga. For instance, DNA replication genes maintained rhythmicity under LL but were repressed under DD, while ribosome biogenesis genes exhibited rhythmicity under DD but lost it under LL. Generally, LL had a more detrimental effect on transcriptome rhythmicity than DD, in agreement with studies in plants such as Medicago truncatula (Wang et al. 2021), S. tuberosum (Hoopes et al. 2022), and Hordeum vulgare (Müller et al. 2020). Notably, rhythmic patterns under freerunning conditions differed significantly from those under diel cycles. LL conditions reduced rhythmic amplitude, likely due to decreased cellular synchrony (Supplementary Fig. S13), as observed in Arabidopsis leaves (Yakir et al. 2011; Wenden et al. 2012), consistent with the cell-autonomous character of circadian clocks. Typical responses of “nocturnal character were found, such as advanced and delayed phases under DD and LL, respectively (Hundahl et al. 2012; Imai et al. 2020). Indeed, most transcripts peaked during the night, consistent with the “Escape from light theory (Pittendrigh 1993). During transcription, DNA is unwound leaving it in an exposed state more susceptible to light-induced damage. To mitigate this risk, Ostreococcus might concentrate transcription predominantly during the night. Although rhythmic genes were almost identical under LD and SD conditions, significant effects on their profiles were observed. Specifically, advanced phases and reductions in amplitude under SD were found, as demonstrated at the level of individual genes in the model chlorophyte microalgae Chlamydomonas reinhardtii (Serrano et al. 2009). Simulations showed that two-thirds of the rhythmic genes with a single peak respond to changes in photoperiod by gradually adjusting their phases while the remaining third adjust their phases according to a more complex mechanism. Another response to photoperiod shortening was the emergence of rhythmic bimodal gene expression profiles. This type of rhythmicity has been suggested to be induced by the cyclic occurrence of tides in coastal environments resulting in circatidal rhythms in benthic diatoms (Bilcke et al. 2021). However, our experimental design did not simulate tidal conditions and Ostreococcus is described as a planktonic picoalga. Moreover, the exact period of circatidal rhythms is 12.4 h, which would result in a daily phase shift of ∼1 h, which was not observed in our data. If these expression patterns were truly circatidal, bimodality should have been maintained under free-running conditions. However, they disappeared resulting in the maintenance of only one of the peaks under LL and the other one under DD. This suggests the combination of 2 distinct profiles, 1 dependent on light and the other 1 on dark. Model simulations showed how, under LD conditions, the 2 distinct profiles overlap in time, producing a single peak, while as the photoperiod shortens they become out of phase producing a bimodal profile under SD conditions. Bimodal rhythmic gene expression profiles have also been identified in plants (Filichkin et al. 2011) and animals (Weger et al. 2021). To further explore the seasonal responses in Ostreococcus at the molecular level, proteomic data were generated and integrated with the transcriptomic data. Our 48% proteome coverage represented an improvement compared to previous studies, such as 12% in A. thaliana (Seaton et al. 2018), 30% in D. melanogaster (Wang et al. 2020), and 9% in M. musculus (Chiang et al. 2014), and lies between published studies in Ostreococcus with 27% in Le Bihan et al. (2011) and 85% in Kay et al. (2021). In contrast to the high transcriptome rhythmicity, a drastic reduction in proteome rhythmicity was found consistent with (Kay et al. 2021). In general, no coincidences between rhythmic protein and transcript abundance profiles were observed with temporal offsets of several hours between them. This observation pointed to the decoupling of transcription and translation and to the existence of a Photoperiodic plasticity in marine phytoplankton rhythms | 17 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 significant regulation over translation initiation (Freeney et al. 2016; Kay et al. 2021). Temporal offset lengths were not uniform suggesting the existence of a differential posttranscriptional regulation for each specific biological process. Similar temporal shifts between transcripts and proteins have been reported in other organisms, such as M. musculus (Robles et al. 2014). Notably, a photoperiodic effect on transcript/protein temporal offsets was detected, with longer offsets under SD than LD. Moreover, in SD-entrained cultures, significantly longer temporal offsets for transcripts peaking during the skotoperiod were found. Physiological measurements were integrated with transcriptome and proteome dynamics, aiming to elucidate the full temporal orchestration at different molecular levels underlying the responses to changes in photoperiod length. Cell-cycle progression, photosynthesis, starch accumulation, carotenoid biosynthesis, and nitrate assimilation presented advanced rhythmic patterns under SD compared to LD conditions, in agreement with transcriptome and proteome responses to photoperiod shortening. In line with the described Ostreococcus algal blooms in spring and summer (Bolaños et al. 2020; Chen et al. 2021; Mondal and Banerjee 2022), in LD-entrained cultures, increased number of cells entering the S phase, enhanced photosynthetic activity, greater starch accumulation, and a more active xanthophyll cycle were patent. While temporal offsets of several hours have been observed between transcripts and proteins, almost coincident protein abundance profiles and rhythmic physiological measurements occur. These temporal offsets shed light on the physiological significance of the observed transcriptional programs over complete diel cycles. For instance, considering the temporal offset of 8 h under LD between the transcript and protein abundances of NR, an enzyme that requires light for its activation, peaking of its transcript at dawn (ZT0) suffices to make the protein peak coincide with maximum light irradiance (ZT8). However, under SD, with a photoperiod of only 8 h, to ensure that the protein peaks at ZT4 (the point of maximum light irradiance), the corresponding transcript must reach its maximum expression level around midnight (ZT12–ZT16). This suggests the existence of molecular mechanisms that allow Ostreococcus to adjust its transcriptome timing depending on the photoperiod length, considering the temporal offsets between transcripts and proteins to ensure that proteins are available at the appropriate moment of the day. Such photoperiodic plastic orchestration between transcriptome, proteome, and physiological timing might play a crucial role in the ability of Ostreococcus to thrive and optimize its physiological processes accordingly under photoperiodic variations in diel cycles. A network model identified specific transcription factor (TF) families as potential key regulators of the rhythmic physiology detected in Ostreococcus such as the families MYB, DOF, bZIP, and CPP. This was further supported by the TFBS enrichment analysis, which identified significant DNA motifs associated to these TF families in specific gene clusters (Fig. 4). Notably, the CCA1 orthologue, ostta06g2340, was found to have a significantly high predictive power over starch content, cell-cycle phases, and carotenoid content suggesting a regulatory role over the corresponding biological processes (Fig. 12A). Furthermore, the EE motif, recognized by this transcription factor in Arabidopsis, was found significantly enriched in genes peaking at the end of the day under LD conditions, some of which encode key proteins in starch metabolism, cell-cycle progression, and carotenoid biosynthesis (Fig. 12B). This indicates a conserved role of this transcription factor in regulating the core of the circadian clocks in Ostreococcus and Arabidopsis. Materials and methods Culture conditions O. tauri (Ostreococcus) sequenced strain RCC4221 was used for all experiments. The growth medium was prepared using artificial sea water (24.55 g NaCl, 0.75 g KCl, 4.07 g MgCl 2 ·6H 2 O, 1.47 g CaCl 2 ·2H 2 O, 6.04 g MgSO 4 ·7H 2 O, and 0.21 g NaHCO 3 per 1 L distilled water) supplemented with 1 mL of Solution I (100 g NaNO 3 in 250 mL distilled water), 1 mL of Solution II (700 mg Na 2 HPO 4 and 2.5 g K 2 HPO 4 in 250 mL distilled water), 1 mL of Solution III with trace metals (2.68 g NH 4 Cl, 5.2 g Fe-EDTA, 37.2 g Na 2 -EDTA, 23 mg ZnSO 4 , 14 mg CoSO 4 , 7.89 mg Na 2 MoO 4 ·2H 2 O, 2.5 mg CuSO 4 , 1.7 mg H 2 SeO 3 , and 180 mg MnCl 2 ·4H 2 O in 500 mL distilled water), and 1 mL of f/2 vitamin solution. Experiments were performed in photochemostats consisting of water jacketed bubble columns with 2 L capacity (7 cm diameter and 50 cm height) containing 1.8 L of cell suspension continuously sparged with air to ensure culture homogenization. The flow of water through the jacket, from an external cooling device, maintained temperature at 20 °C. A pH probe was submerged into the culture and connected to a pH meter serving as input to a LabJack that controlled an electrovalve, allowing on demand injection of CO 2 into the air stream entering the culture to maintain the pH at 8.0. Each photochemostat was illuminated during the corresponding light period using 6 Phillips PL-32 W/840/4p whitelight fluorescent lamps. The illuminating system simulated the progressive light intensity increase and decrease during solar daylight cycles with a maximum light irradiance of 1,500 μE m −2 s −1 . Under LL conditions light intensity was maintained to this maximum level. Each photochemostat was kept within a wooden case and covered by a completely opaque fabric, to ensure that the illumination is only provided by the fluorescent lamps. The illumination regime was 16 h of light and 8 h of darkness for LD condition and 8 h of light and 16 h of darkness for SD condition. Photochemostats were operated in continuous mode adding fresh medium continuously during the light period at a flow rate of 45 mL h −1 with a peristaltic pump. Since the photoperiod under LD is twice as long as under SD, the dilution rates were 0.4 d −1 under LD and 0.2 d −1 under SD. Excess culture was removed at the same rate by the overflow in order to keep a constant volume. For the comparative analysis of starch content, Ostreococcus cells were grown following the methodology outlined in O’-Neill et al. (2011). Sample collection, RNA extraction, and purification Samples were collected for 3 consecutive days every 4 h at ZT0, ZT4, ZT8, ZT12, ZT16, and ZT20 where Zeitgeber time N (ZTN) marks the time point N hours after the beginning of the photoperiod simulating dawn. Specifically, ZT0 was collected immediately after lights were on; and ZT8 (under SD conditions) and ZT16 (under LD conditions) were collected immediately after light were off. Subsequently, cultures were transferred to free-running conditions consisting of LL and DD. No samples were collected during the first day to allow cultures acclimation, and then samples were collected for 2 consecutive days every 4 h at CT0, CT4, CT8, CT12, CT16, and CT20 where Circadian time N (CTN) denotes the time point N hours after the subjective dawn. For each time point, 50 mL of culture were collected for RNA extraction. Cells were washed with phosphate-buffered saline (PBS) solution using centrifugation for 1 min at 13,000 ×g and 4 °C. After supernatant removal, cells were immediately flash frozen in liquid nitrogen and stored at −80 °C. 18 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Frozen pellets were resuspended in 400 μL of disruption buffer (García-Domínguez and Florencio 1997) and added to a 1.5 mL Eppendorf tubes (RNAse free and phenol-proof) containing 400 μL of phenol:chloroform 1:1 and 100 μL of acid washed glass beads (0.25 to 0.3 mm diameter; Braun, Melsungen, Germany). Mechanical disruption was performed for 30 min by alternating cycles of vortexing for 60 s and ice incubation for 60 s. Extracts were centrifuged at 4 °C for 15 min at 13,000 ×g, the upper aqueous phase was collected and mixed with 400 μL of phenol:chloroform 1:1 and centrifuged in the same conditions for 5 min. This process was repeated a total of 3 times. In the last wash, only chloroform was used to remove phenol from the samples. The supernatant was incubated overnight at −20 °C in a solution of 80 μL 10 M LiCl and 550 μL 100% EtOH for RNA precipitation. Subsequently, samples were centrifuged for 10 min at 13,000×g at 4 °C. Pellets were dried to avoid EtOH contamination. Further RNA purification was performed using the Isolate II RNA Plant Kit (Bioline) following the manufacturer instructions. RNA concentration and integrity were measured using a Bioanalyzer 2100 (Agilent RNA 6000 Nano Kit). RNA-seq data generation and analysis Library preparation was carried out following the manufacturer’s instructions and sequencing was performed on an Illumina NextSeq500 sequencer for 3 replicates for each time point under LD and SD conditions. Approximately 10 million 75 nt long single end reads were generated for each sample. RNA-seq data were analyzed using our pipeline MicroAlgae RNA-seq and Chip-seq AnalysiS (MARACAS) (Romero-Losada et al. 2022). Specifically, the high quality of the sequencing data was assessed using the software package FASTQC. The O. tauri genome sequence and annotation v3.0 (https://phycocosm.jgi.doe.gov/Ostta4221_ 3/Ostta4221_3.home.html) were used as reference genome. Reads were mapped to the reference genome with HISAT2 (Kim et al. 2019). Transcript assembly and gene expression estimation measured as fragments per kilobase of exon and millions of mapped reads (FPKM) were performed using StringTie2 (Kovaka et al. 2019) and the bioconductor R package Ballgown (Frazee et al. 2015). PCA and HC were performed using the R package FactoMineR (Lê et al. 2008). The R code for this analysis is available from the GitHub repository SANDAL (https://github.com/franromero-campero/SANDAL). RT-qPCR for CCA1 was performed using the forward primer (CTAGTACGTCGTCGAGC) and the reverse primer (CCACGAACG GACTCAT), as internal reference we used EF1 alpha with forward primer (GACGCGACGGTGGATCAA) and reverse primer (CGACT GCCATCGTTTTACC) following the methodology described previously in Corellou et al. (2009). Sample collection and protein extraction Sample collection was performed as described for RNA analysis for 3 consecutive days under LD and SD conditions. For cell disruption, 1 mL of TRIsure, 100 μL of acid washed glass beads (0.25 to 0.3 mm diameter) and 40 μL of protein inhibitor cocktail (25×) were added onto frozen pellets, followed by 3 disruption cycles (60 s agitation—60 s incubation on ice) using a Mini-Beadbeater (BioSpec Products). Proteins were extracted using TRIsure Reagent (Sigma-Aldrich), according to the manufacturer’s instructions. The resulting proteins pellets were resuspended with 2 mL of 0.3 M guanidine solution in 95% EtOH using 10 sonication cycles (30 s sonication—30 s of incubating at 4 °C) and then centrifuged at 4 °C during 5 min at 8,000×g. This washing process was repeated twice, followed by 2 additional washes using 90% EtOH. The final pellets were resuspended in NH 4 HCO 3 50 mM/ 0.2% Rapidgest (Waters) and total proteins were quantified using a Qubit device. For each sample, 50 μg of proteins were incubated with dithiothreitol (final concentration 4.5 mM) for 30 min at 60 °C. Iodoacetamide was added to a final concentration of 10 mM and incubated for 30 min, under total darkness at room temperature. An overnight trypsin treatment was done at 37 °C in a 1:40 trypsin: Figure 12. Integration of transcription factor transcriptomic data and physiological measurements. A) Network representation of a sPLS model integrating transcription factor gene expression as predictor variables with physiological measurements as response variables. Circular nodes represent genes encoding transcription factors from different families, indicated by distinct colors. Rounded square nodes represent physiological measurements: S and G1 cell cycle phases (in green), Starch content (in white) and carotenoid content (in pink) prasinoxanthin (Prasi), violaxanthin (Vio), antheraxanthin (Anthe), zeaxanthin (Zea), neoxanthin (Neo), micromonal (Micro), and uriolide (Uri). An edge between a transcription factor node and a physiological measurement node is drawn when the expression profile of the corresponding transcription factor has a significantly high predictive power over the corresponding physiological measurement, suggesting a potential regulatory role. The CCA1 orthologue (ostta06g02340) is identified as a potential central regulator. B) Identification of the EE motif, recognized by the transcription factor CCA1, in the promoters of genes encoding key proteins involved in starch metabolism, cell cycle progression, and carotenoid biosynthesis. Photoperiodic plasticity in marine phytoplankton rhythms | 19 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 protein. Subsequently, formic acid was added and incubated at 37 °C for 1 h. Finally, 2% acetonitrile (v/v) were added to reach a concentration of the digested sample of ∼0.5 μg of protein/μL. Sequential windowed acquisition of all theoretical mass spectra proteomics data generation and analysis We utilized a label free quantification platform that employs independent data acquisition called Sequential Windowed Acquisition of all THeoretical Mass Spectra (SWATH-MS) (Ludwig et al. 2018) using a time-of-flight (TOF) triple quadrupole hybrid mass spectrometer MS (5600 plus, Sciex) equipped with a nanoelectrospray source coupled to a nano-HPLC Eksigent model 425. The Sciex software Analyst TF 1.7 was used for the equipment control and data acquisition. Peptides were first loaded onto a trap column (Acclaim PepMap 100 C18, 5 µm, 100 Å, 100 µm id× 20 mm, Thermo Fisher Scientific) under isocratical order in 0.1% formic acid/2% acetonitrile (v/v) at a flow rate of 3 μL/min for 10 min. Subsequently, they were eluted on a reversed-phase analytical column, Acclaim PepMap 100 C18, 3 µm, 100 Å, 75 µm id× 250 mm, Thermo Fisher Scientific, coupled to a PicoTip emitter (F360-20-10-N-20_C12 from New Objective). Formic acid 0.1% (v/v) was used as Solvent A and 2% acetonitrile with formic acid 0.1% (v/v) as Solvent B. Peptides were eluted with a linear gradient of 5% to 35% (v/v) of Solvent B in 120 min at a flow rate of 300 nL/min. The source voltage was selected at 2,600 V and the temperature was maintained at 100 °C. Gas 1 was selected at 20 PSI, Gas 2 at 0, and curtain gas at 25 PSI. For proteins identification, a TOF-MS with a scan window of 400 to 1,250 m/z (accumulation time of 250 ms) was used followed by 50 MS/MS with a scan window of 230 to 1,500 m/z (accumulation time of 65 ms) and with a cycle time of 2,574 s. Spectral libraries for LD and SD conditions were constructed by making one run with a mixture of the biological replicates corresponding to each time point. ProteinPilot v5.0.1 software (Sciex) was used to identify the proteins in the library. A pooled search of all runs was performed. The parameters of the Paragon method were: trypsin as enzyme and iodoacetamide as cysteine alkylating agent. The O. tauri annotated proteome v3.0 file from: https:// phycocosm.jgi.doe.gov/Ostta4221_3/Ostta4221_3.home.html linked to a Sciex Contaminants database were used in library construction. A false positive analysis (FDR) was performed and those with FDR<0.01 were considered. For each sample, the equivalent of 1 µg of digested protein was injected into each run. Previously, the equipment was selfcalibrated using a standard, MS synthetic peptide calibration kit from Sciex, to control sensitivity and chromatographic conditions. Protein identification and quantification were performed using SWATH runs with 60 ms of accumulation time and 3.7 s of cycle time. Three technical replicates for each 1 of the 3 biological replicates were analyzed resulting in 9 replicates per time point. The generated libraries (1% FDR) were analyzed using the Sciex software PeaKView 2.2 with the microapp SWATH 2.0, together with the data obtained from the SWATH runs. Using this software, the chromatographic traces of the ions were extracted and dumped into the Marker view 1.2.1.1 program where the list of identified proteins with their corresponding areas was generated. The parameters for extraction of ions and areas were: 10 peptides per protein, 7 transitions of each peptide, threshold of confidence of the peptides set at 90, and FDR 1%. The R package NormalyzerDE (Willforss et al. 2019) was used to perform Quantile normalization. Data were imputed with mean imputation method between the 9 technical and biological replicates. Cell cycle data acquisition and analysis A volume of 1.5 mL of cell suspension was harvested for each time point under LD and SD conditions and diluted 1:10 in PBS. Two milliliter of these dilutions were centrifuged and cells in the pellets were fixed with 10 mL of 100% EtOH before storage at −20 °C for, at least, 24 h. After fixation, cell suspensions were centrifuged for 5 min at 3,500×g (room temperature) and resuspended in 1 mL of PBS, washed once with PBS and sonicated for 3 min in an Ultrasonic Cleaner (JSP, US21, ultrasonic power 50 W), in order to eliminate cell clumps and aggregates before staining. In the staining process, 2 μL of the Vibrant Dye Cycle Green (V35004, Thermo Fisher) (10 μM final stain concentration) were added to each sample and incubated 30 min (37 °C) for selective DNA labeling. After incubation, cells were washed and transferred to flow cytometry tubes for cell cycle analysis. Flow cytometry acquisition was performed with a BD FACS Canto II (BD Biosciences) where stained DNA was excited by a 488 nm laser and emission was collected in a 530/30 nm photomultiplier tube (PMT). Flow rate was low and linear amplification was established for the acquisition. Data were analyzed using FlowJo v.10.6.1 (Becton Dickinson & Company BD). Analysis was performed using the Watson pragmatic algorithm provided by FlowJo (Watson et al. 1987) to adjust the data to the model. Confocal microscopy Confocal microscopy images of Ostreococcus cells were acquired using a spectral Laser Scanning Confocal Microscope (Olympus FLOUVIEW FV3000). Excitation was performed with a 488 nm laser. Emission signals were detected within the green channel (500 to 540 nm, gain 504) and red channel (650 to 750 nm, gain 432). Photosynthetic activity: sample collection and data acquisition Fresh cultures were harvested for each time point under LD and SD conditions. Samples were diluted 1:1 with growth medium and incubated at 20 °C in total darkness during 10 min. In order to analyze photosynthetic parameters, pulse-amplitude-modulation (PAM) fluorometry measurements were performed using a Waltz DUAL-PAM-100. After darkness incubation, non-actinic modulated light (450 nm, 2.8 μE m −2 s −1 ) was turned on to measure the fluorescence basal level, F 0 . A saturating red light pulse (655 nm and 5,000 μE m −2 s −1 ) was applied for 400 ms to determine the maximum fluorescence level, F m . Fv/Fm, the maximum potential quantum efficiency of PSII when all reaction centers are open, was calculated as Fv/Fm=(F m −F 0 )/F m . Starch content determination At each time point, 50 mL of fresh culture were harvested, centrifuged at 7,000×g for 10 min. The pellets were washed with 1% ammonium formiate (p/v) to eliminate salts, and lyophilized. Approximately 2 to 3 mg of lyophilized biomass were added to hermetic tubes containing 1 mL of glass beads (0.25 to 0.3 mm diameter) and 2 mL of chloroform:methanol (2:1). Three disrupting cycles (60 s agitation−60 s incubation on ice) were applied using Mini-Beadbeater (BioSpec Products). Then, cell extracts were separated from the beads and saved in new tubes. Cell extracts were centrifuged for 4 min at 13,000×g and the supernatant was discarded. Pellets were washed with chloroform:methanol (2:1) until 20 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 they became white ensuring pigments and lipids removal. Pellets were dried and the spectrophotometric protocol described in García-Cubero et al. (2018) was used to determine starch content. Carotenoid content determination Four milligrams of lyophilized biomass were added to a hermetic tube containing 1 mL of glass beads (0.25 to 0.3 mm diameter) and 1 mL of pure acetone. Three disrupting cycles (60 s agitation−60 s incubation on ice) were applied using Mini-Beadbeater (BioSpec Products). Carotenoids extraction was achieved following the protocol developed by Del Campo et al. (2004). A Hitachi HPLC (Elite LaChrom), equipped with a photodiode-array detector (Hitachi L-2455) was used. Separation was performed on a Waters NovaPak C-18 (3.9 ×150 mm, 4 µm particle size, 60 Å pore size) column. The eluents used to create a gradient through the mobile phase were: Eluent A (0.1 M ammonium acetate and 15:85 v/v H 2 O–methanol) and Eluent B (44:43:13 v/v methanol–acetonitrile–acetone). Temperature was maintained constant (20 °C) during the whole process and eluents flowed at 800 μL min −1 . Different carotenoids were identified according to retention times and absorption profiles of known carotenoids. Quantification was determined as a percentage of the total peak area corresponding to carotenoids. Identification of genes, proteins and physiological measurements exhibiting rhythmic patterns and their statistical comparative analysis The bioconductor R package rhythmicity analysis incorporating nonparametric methods (RAIN) (Thaben and Westermark 2014) was used to identify genes, proteins, and physiological measurements exhibiting rhythmic patterns. A Benjamini–Hochberg corrected P-value threshold equal to 0.05 was used in all the cases under study. Rhythmic patterns with a single peak per day were identified by setting the period parameter to 24 h. Rhythmic patterns exhibiting 2 peaks per day were identified by setting the period parameter to 12 h. For statistical comparison, rhythmic patterns were fitted to a co-sinusoidal curve characterized by 3 parameters: mesor, amplitude, and phase. The statistical significance of the differences in amplitude and phase between different groups was performed using the R package CircaCompare (Parsons et al. 2020). A Benjamini–Hochberg corrected P-value threshold of 0.05 was used to determine statistically significant differences. The nonparametric method implemented in RAIN was used first since it searches for any type of rhythmic patterns. The methods in CircaCompare were also employed identifying approximately the same rhythmicity. Once it was confirmed that our data could be further analyzed using the parametric methods in CircaCompare, significance analysis of differences in phases and amplitudes were performed using CircaCompare. The significance of the global differences in these parameters was assessed using the Mann–Whitney–Wilcoxon nonparametric test implemented in the R function wilcox.test. Phase differences between transcript and protein expression profiles were computed assuming that transcript peaks precede those of proteins. Significant differences between specific features not related to rhythmicity were also assessed using the Mann–Whitney– Wilcoxon nonparametric test implemented in the R function wilcox.test. Functional annotation of gene sets Functional enrichment analysis over different gene sets was performed using our online tool AlgaeFUN, microALGAE FUNctional enrichment tool (Romero-Losada et al. 2022) that in turn is based on the bioconductor packages clusterProfiler, enrichplot, and pathview (Luo and Brouwer 2013; Wu et al. 2021) and the functional annotation package developed by our group org.Otauri.eg.db for O. tauri (https://github.com/fran-romero-campero/AlgaeFUN/ tree/master/packages/annotation_packages). TFBS enrichment analysis The identification of plant specific TFBS significantly enriched in the promoters of the genes peaking at specific time points was performed using the software for motif discovery and next generation sequencing analysis Hypergeometric Optimization of Motif EnRichment (Heinz et al. 2010). The function findMotifsGenome.pl was applied with the parameter for plant specific TFBS-mset plants. The length of the gene promoter was fixed to 500 nt upstream of the start codon and the motif lengths to 6 and 8 nt. Integrative model based on sparse partial least squares The R package mixomics (Rohart et al. 2017) was used to develop a network model for the integration of physiological and transcription factor gene expression data based on the multivariate projection method sparse partial least squares (sPLS) (Lê Cao et al. 2009). Physiological data corresponding to cell cycle phases, starch and carotenoid contents were considered as response variables and gene expression data for transcription factors as predictors. The model was constructed with the function spls in regression mode and keeping 3 components in the projection. The function network with a cutoff of 0.45 was applied to generate the network representation of the model. The software tool Cytoscape (Shannon et al. 2003) was used for the graphical representation of this network using the yFiles Organic Layout. Enzyme activities At each time point, 1 mL of fresh culture was used to measure enzyme activities. Specifically, NR and GS activities were measured as previously described by Herrero et al. (1981) and García-Domínguez and Florencio (1997), respectively. Accession numbers The O. tauri RCC4221 genome sequence used in this article can be found in the GenBank data libraries under accession number GCA_000214015.2. Acknowledgments We would like to acknowledge Eloisa Andújar and Mónica Pérez from the CABIMER Genomics Unit for their assistance with highthroughput sequencing, Rocío Rodríguez from the IBVF Proteomics Units for guidance with SWATH-MS proteomics, Carlos Parejo from the IBVF Chromatography and Mass Spectrometry Unit for his contribution in carotenoid content determination, Alicia Orea from the IBVF Microscopy Service for her help with microscopy imaging, José Enrique Frías from the Cic-Cartuja Cell Culture Unit for his assistance with Ostreococcus tauri cultivation, and José Moreno Fernández for setting up Ostreococcus tauri cultivation in photochemostats for this project. We would like to thank Miguel G. Guerrero and Inmaculada Couso for critical reading of this manuscript. Photoperiodic plasticity in marine phytoplankton rhythms | 21 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Author contributions A.B.R.-L., M.E.G.-G., M.M.-P., and M.G.-G. performed wet lab experiments. A.B.R.-L., C.A., and F.J.R.-C. generated and analyzed RNA-seq transcriptomic and SWATH-MS proteomic data. A.B.R.-L., M.J.C.-P., and F.J.R.-C. generated and analyzed cell-cycle data. A.B.R.-L. and F.J.R.-C. generated and analyzed photosynthetic efficiency measurements. A.B.R.-L., M.E.G.-G., M.M.-P., and C.A. measured and analyzed starch accumulation, carotenoid content, and enzyme activities. C.A. performed RT-qPCR analysis. G.O. and Y.P.C. performed experiments in batch mode with low nitrate medium. F.J.R.-C., M.G.-G., A.B.R.-L., and G.O. interpreted the results and wrote the manuscript. All authors read and approved the final manuscript. Supplementary data The following materials are available in the online version of this article. Supplementary Figure S1. Experimental design and transcriptomic RNA-seq data reliability. Supplementary Figure S2. Biological processes significantly enriched in the genes exhibiting rhythmicity under alternating light/ dark cycles and constant light. Supplementary Figure S3. Biological processes significantly enriched in the genes exhibiting rhythmicity under alternating light/ dark cycles and constant dark. Supplementary Figure S4. Gene expression profiles of clock components and light receptors. Supplementary Figure S5. Functional enrichment over circadian genes after LD and SD entrainment. Supplementary Figure S6. Transcriptional temporal program of the distribution of biological processes over diel cycles under LD conditions (16 h light/8 h dark). Supplementary Figure S7. Transcriptional temporal program of the distribution of biological processes over diel cycles under short day conditions (8 h light/16 h dark). Supplementary Figure S8. Validation of predicted emergence of bimodal rhythmic profiles under neutral day conditions. Supplementary Figure S9. Proteomic data normalization and reliability. Supplementary Figure S10. Transcript/protein phase offsets did not correlate to any biochemical properties computed from protein sequences. Supplementary Figure S11. Heatmaps integrating transcript and protein abundance under LD and SD conditions for different biological processes. Supplementary Figure S12. Gene expression and starch content under constant dark. Supplementary Figure S13. Reductions in amplitude under free-running conditions can be explained by a decline in culture synchrony. Supplementary Table S1. Rhythmic genes under different light regimes. Supplementary Table S2. Set of rhythmic genes peaking at each time point under longand short-day conditions. Supplementary Table S3. Set of rhythmic genes with 2 peaks. Supplementary Table S4. Protein abundances under longand short-day conditions. Supplementary Table S5. Rhythmic proteins under longand short-day conditions. Supplementary Table S6. Cell cycle data under longand shortday conditions. Supplementary Table S7. Starch content data under longand short-day conditions. Supplementary Table S8. Carotenoid content data under longand short-day conditions. Supplementary Table S9. Nitrate reductase and glutamine synthetase activities under longand short-day conditions. Funding This work was supported by grants BIO2017-84066-R (MINOTAUR) and PID2021-123984OB-I00 (ELECTRA) from the Spanish Ministry of Science and Innovation. C.A. was supported by Conserjería de Conocimiento, Investigación y Universidad, Junta de Andalucia (Grant PREDOC_00999). G.O. and Y.P.C. were supported by a Wellcome Trust Award (225212/Z/22/Z) and a BBSRC Award (APP3975). Conflict of interest statement. The authors declare no competing interests. Data availability RNA-seq data generated in this study are freely available from the Gene Expression Omnibus database under the accession number GSE155535. The SWATH-MS proteomics data generated in this study have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al. 2022) partner repository with the dataset identifier PXD046992. The data analysis code developed using the statistical programming language R is freely available from the following GitHub repository SANDAL: https:// github.com/fran-romero-campero/SANDAL. The results presented in this paper can be further explored using the online tool MINOTAUR: https://greennetwork.us.es/MINOTAUR/. References Annunziata R, Ritter A, Fortunato AE, Manzotti A, CheminantNavarro S, Agier N, Huysman MJJ, Winge P, Bones AM, Bouget FY, et al. bHLH-PAS protein RITMO1 regulates diel biological rhythms in the marine diatom Phaeodactylum tricornutum. Proc Natl Acad Sci U S A. 2019:116(26):13137–13142. https://doi.org/10. 1073/pnas.1819660116 Bachy C, Wittmers F, Muschiol J, Hamilton M, Henrissat B, Worden AZ. The land-sea connection: insights into the plant lineage from a green algal perspective. Annu Rev Plant Biol. 2022:73(1): 585–616. https://doi.org/10.1146/annurev-arplant-071921-100530 Baty F, Ritz C, Charles S, Brutsche M, Flandrois JP, Delignette-Muller ML. A toolbox for nonlinear regression in R: the package nlstools. J Stat Softw. 2015:66(5):1–21. https://doi.org/10.18637/jss.v066.i05 Benites LF, Bucchini F, Sanchez-Brosseau S, Grimsley N, Vandepoele K, Piganeau G. Evolutionary genomics of sex-related chromosomes at the base of the green lineage. Genome Biol Evol. 2021:13(10):evab216. https://doi.org/10.1093/gbe/evab216 Bilcke G, Osuna-Cruz CM, Santana-Silva M, Poulsen N, D’hondt S, Bulankova P, Vyverman W, De Veylder L, Vandepoele K. Diurnal transcript profiling of the diatom Seminavis robusta reveals adaptations to a benthic lifestyle. Plant J. 2021:107(1):315–336. https:// doi.org/10.1111/tpj.15291 Blanc-Mathieu R, Verhelst B, Derelle E, Rombauts S, Bouget FY, Carré I, Château A, Eyre-Walker A, Grimsley N, Moreau H, et al. An improved genome of the model marine alga Ostreococcus tauri unfolds by assessing Illumina de novo assemblies. BMC Genomics. 2014:15(1):1–12. https://doi.org/10.1186/1471-2164-15-1103 22 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Bläsing OE, Gibon Y, Günther M, Höhne M, Morcuende R, Osuna D, Thimm O, Usadel B, Scheible WR, Stitt M. Sugars and circadian regulation make major contributions to the global regulation of diurnal gene expression in Arabidopsis. Plant Cell. 2005:17(12): 3257–3281. https://doi.org/10.1105/tpc.105.035261 Bolaños LM, Karp-Boss L, Choi CJ, Worden AZ, Graff JR, Haëntjens N, Chase AP, Della Penna A, Gaube P, Morison F, et al. Small phytoplankton dominate western North Atlantic biomass. ISME J. 2020:14(7):1663. https://doi.org/10.1038/s41396-020-0636-0 Bousquet L, Hemon C, Malburet P, Bucchini F, Vandepoele K, Grimsley N, Moreau H, Echeverria M. The medium-size noncoding RNA transcriptome of Ostreococcus tauri, the smallest living eukaryote, reveals a large family of small nucleolar RNAs displaying multiple genomic expression strategies. NAR Genomics Bioinforma. 2020:2(4):lqaa080. https://doi.org/10.1093/ nargab/lqaa080 Castillo YM, Forn I, Yau S, Morán XAG, Alonso-Sáez L, Arandia-Gorostidi N, Vaqué D, Sebastián M. Seasonal dynamics of natural Ostreococcus viral infection at the single cell level using VirusFISH. Environ Microbiol. 2021:23(6):3009–3019. https://doi. org/10.1111/1462-2920.15504 Chen TY, Lai CC, Tai JH, Ko CY, Shiah FK. Diel to seasonal variation of picoplankton in the tropical South China Sea. Front Mar Sci. 2021:8:732017. https://doi.org/10.3389/fmars.2021.732017 Chiang CK, Mehta N, Patel A, Zhang P, Ning Z, Mayne J, Sun WYL, Cheng HYM, Figeys D. The proteomic landscape of the suprachiasmatic nucleus clock reveals large-scale coordination of key biological processes. PLOS Genet. 2014:10(10):e1004695. https://doi.org/10.1371/journal.pgen.1004695 Corellou F, Camasses A, Ligat L, Peaucellier G, Bouget FY. Atypical regulation of a green lineage-specific B-type cyclin-dependent kinase. Plant Physiol. 2005:138(3):1627–1636. https://doi.org/10. 1104/pp.105.059626 Corellou F, Schwartz C, Motta JP, Djouani-Tahri EB, Sanchez F, Bougeta FY. Clocks in the green lineage: comparative functional analysis of the circadian architecture of the picoeukaryote ostreococcus. Plant Cell. 2009:21(11):3436–3449. https://doi.org/10. 1105/tpc.109.068825 Degraeve-Guilbault C, Bréhélin C, Haslam R, Sayanova O, Marie-Luce G, Jouhet J, Corellou F. Glycerolipid characterization and nutrient deprivation-associated changes in the green picoalga Ostreococcus tauri. Plant Physiol. 2017:173(4):2060. https://doi.org/10.1104/pp. 16.01467 Degraeve-Guilbault C, Gomez RE, Lemoigne C, Pankansem N, Morin S, Tuphile K, Joubès J, Jouhet J, Gronnier J, Suzuki I, et al. Plastidic Δ6 fatty-acid desaturases with distinctive substrate specificity regulate the pool of C18-PUFAs in the ancestral picoalga Ostreococcus tauri. Plant Physiol. 2020:184(1):82–96. https://doi.org/ 10.1104/pp.20.00281 Del Campo JA, Rodríguez H, Moreno J, Vargas M, Rivas J, Guerrero MG. Accumulation of astaxanthin and lutein in Chlorella zofingiensis (Chlorophyta). Appl Microbiol Biotechnol. 2004:64(6): 848–854. https://doi.org/10.1007/s00253-003-1510-5 de los Reyes P, Romero-Campero FJ, Teresa Ruiz M, Romero JM, Valverde F. Evolution of daily gene co-expression patterns from algae to plants. Front Plant Sci. 2017:8:1217. https://doi.org/10. 3389/fpls.2017.01217 Derelle E, Ferraz C, Rombauts S, Rouzé P, Worden AZ, Robbens S, Partensky F, Degroeve S, Echeynié S, Cooke R, et al. Genome analysis of the smallest free-living eukaryote Ostreococcus tauri unveils many unique features. Proc Natl Acad Sci U S A. 2006:103(31): 11647–11652. https://doi.org/10.1073/pnas.0604795103 Derelle E, Yau S, Moreau H, Grimsley NH. Prasinovirus attack of Ostreococcus is furtive by day but savage by night. J Virol. 2018: 92(4):e01703-17. https://doi.org/10.1128/JVI.01703-17 Edwards KF, Thomas MK, Klausmeier CA, Litchman E. Light and growth in marine phytoplankton: allometric, taxonomic, and environmental variation. Limnol Oceanogr. 2015:60(2):540–552. https://doi.org/10.1002/lno.10033 Ferrari C, Proost S, Janowski M, Becker J, Nikoloski Z, Bhattacharya D, Price D, Tohge T, Bar-Even A, Fernie A, et al. Kingdom-wide comparison reveals the evolution of diurnal gene expression in Archaeplastida. Nat Commun. 2019:10(1):737. https://doi.org/10. 1038/s41467-019-08703-2 Fernández-Marín B, Roach T, Verhoeven A, García-Plazaola JI. Shedding light on the dark side of xanthophyll cycles. New Phytol. 2021:230(4):1336–1344. https://doi.org/10.1111/nph.17191 Field CB, Behrenfeld MJ, Randerson JT, Falkowski P. Primary production of the biosphere: integrating terrestrial and oceanic components. Science. 1998:281(5374):237–240. https://doi.org/10.1126/ science.281.5374.237 Filichkin SA, Breton G, Priest HD, Dharmawardhana P, Jaiswal P, Fox SE, Michael TP, Chory J, Kay SA, Mockler TC. Global profiling of rice and poplar transcriptomes highlights key conserved circadiancontrolled pathways and cis-regulatory modules. PLoS One. 2011:6(6):e16907. https://doi.org/10.1371/journal.pone.0016907 Frazee AC, Pertea G, Jaffe AE, Langmead B, Salzberg SL, Leek JT. Ballgown bridges the gap between transcriptome assembly and expression analysis. Nat Biotechnol. 2015:33(3):243–246. https:// doi.org/10.1038/nbt.3172 Freeney KA, Hansen LL, Putker M, Olivares-Yañez C, Day J, Eades LJ, Larrondo LF, Hoyle NP, O’Neill JS, van Ooijen G. Daily magnesium fluxes regulate cellular timekeeping and energy balance. Nature. 2016:532(7599):375–379. https://doi.org/10.1038/nature17407 García-Cubero R, Moreno-Fernández J, Acién-Fernández FG, García-González M. How to combine CO 2 abatement and starch production in Chlorella vulgaris. Algal Res. 2018:32:270–279. https://doi.org/10.1016/j.algal.2018.04.006 García-Domínguez M, Florencio FJ. Nitrogen availability and electron transport control the expression of glnB gene (encoding PII protein) in the cyanobacterium Synechocystis sp. PCC 6803. Plant Mol Biol. 1997:35(6):723–734. https://doi.org/10.1023/A:10058466 26187 Gil-Rodríguez S, Crosby P, Hansen LL, Grünewald E, Beale AD, Spangler RK, Rabbitts BM, Partch CL, Stangherlin A, O’Neill JS, et al. Potassium rhythms couple the circadian clock to the cell cycle. BioRxiv 587153. https://doi.org/10.1101/2024.04.02.587153, 3 April 2024, preprint: not peer reviewed. Graf A, Schlereth A, Stitt M, Smith AM. Circadian control of carbohydrate availability for growth in Arabidopsis plants at night. Proc Natl Acad Sci U S A. 2010:107(20):9458–9463. https://doi.org/10. 1073/pnas.0914299107 Guyon JB, Vergé V, Schatt P, Lozano JC, Liennard M, Bouget FY. Comparative analysis of culture conditions for the optimization of carotenoid production in several strains of the picoeukaryote Ostreococcus. Mar Drugs. 2018:16(3):76. https://doi.org/10.3390/ md16030076 Häfker NS, Andreatta G, Manzotti A, Falciatore A, Raible F, Tessmar-Raible K. Rhythms and clocks in marine organisms. Ann Rev Mar Sci. 2023:15(1):509–538. https://doi.org/10.1146/annurevmarine-030422-113038 Harmer SL, Hogenesch JB, Straume M, Chang H, Han B, Zhu T, Wang X, Kreps JA, Kay S. Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science. 2000:290(5499): 2110–2113. https://doi.org/10.1126/science.290.5499.2110 Photoperiodic plasticity in marine phytoplankton rhythms | 23 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Hashimoto H, Uragami C, Cogdell RJ. Carotenoids and photosynthesis. Subcell Biochem. 2016:79:111–139. https://doi.org/10.1007/9783-319-39126-7_4 Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, Cheng JX, Murre C, Singh H, Glass CK. Simple combinations of lineagedetermining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010:38(4): 576–589. https://doi.org/10.1016/j.molcel.2010.05.004 Henderson GP, Gan L, Jensen GJ. 3-D ultrastructure of O. tauri: electron cryotomography of an entire eukaryotic cell. PLoS One. 2007:2(8):e749. https://doi.org/10.1371/journal.pone.0000749 Henríquez-Castillo C, Botebol H, Mouton A, Ramírez-Flandes S, Lozano JC, Lelandais G, Andrade S, Trefault N, De la Iglesia R, Bouget FY. Ostreococcus tauri luminescent reporter lines as biosensors for detecting pollution from copper-mine tailing effluents in coastal environments. Front Environ Sci. 2018:6:337387. https:// doi.org/10.3389/fenvs.2018.00022 Herrero A, Flores E, Guerrero MG. Regulation of nitrate reductase levels in the cyanobacteria Anacystis nidulans, Anabaena sp. Strain 7119, and Nostoc sp. strain 6719. J Bacteriol. 1981:145(1):175–180. https://doi.org/10.1128/jb.145.1.175-180.1981 Hoopes GM, Zarka D, Feke A, Acheson K, Hamilton JP, Douches D, Buell CR, Farré EM. Keeping time in the dark: potato diel and circadian rhythmic gene expression reveals tissue-specific circadian clocks. Plant Direct. 2022:6(7):e425. https://doi.org/10.1002/pld3.425 Hundahl CA, Fahrenkrug J, Hay-Schmidt A, Georg B, Faltoft B, Hannibal J. Circadian behaviour in neuroglobin deficient mice. PLoS One. 2012:7(4):e34462. https://doi.org/10.1371/journal.pone. 0034462 Imai R, Makino H, Katoh T, Kimura T, Kurita T, Hokamura K, Umemura K, Nakajima Y. Desflurane anesthesia shifts the circadian rhythm phase depending on the time of day of anesthesia. Sci. Reports. 2020:10:1–9. Kamioka M, Takao S, Suzuki T, Taki K, Higashiyama T, Kinoshita T, Nakamichi N. Direct repression of evening genes by CIRCADIAN CLOCK-ASSOCIATED1 in the Arabidopsis circadian clock. Plant Cell. 2016:28(3):696–711. https://doi.org/10.1105/tpc.15.00737 Kay H, Grünewald E, Feord HK, Gil S, Peak-Chew SY, Stangherlin A, O’Neill JS, van Ooijen G. Deep-coverage spatiotemporal proteome of the picoeukaryote Ostreococcus tauri reveals differential effects of environmental and endogenous 24-hour rhythms. Commun Biol. 2021:4(1):1147. https://doi.org/10.1038/s42003-021-02680-3 Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019:37(8):907–915. https://doi.org/10.1038/s41587019-0201-4 Kovaka S, Zimin AV, Pertea GM, Razaghi R, Salzberg SL, Pertea M. Transcriptome assembly from long-read RNA-seq alignments with StringTie2. Genome Biol. 2019:20(1):278. https://doi.org/10. 1186/s13059-019-1910-1 Kuhlman SJ, Craig LM, Duffy JF. Introduction to chronobiology. Cold Spring Harb Perspect Biol. 2018:10(9):a033613. https://doi.org/10. 1101/cshperspect.a033613 Lê S, Josse J, Husson F. FactoMineR: an R package for multivariate analysis. J Stat Softw. 2008:25(1):1–18. https://doi.org/10.18637/ jss.v025.i01 Le Bihan T, Hindle M, Martin SF, Barrios-Llerena ME, Krahmer J, Kis K, Millar AJ, van Ooijen G. Label-free quantitative analysis of the casein kinase 2-responsive phosphoproteome of the marine minimal model species Ostreococcus tauri. Proteomics. 2015:15(23–24): 4135–4144. https://doi.org/10.1002/pmic.201500086 Le Bihan T, Martin SF, Chirnside ES, van Ooijen G, Barrios-LLerena ME, O’Neill JS, Shliaha PV, Kerr LE, Millar AJ. Shotgun proteomic analysis of the unicellular alga Ostreococcus tauri. J Proteomics. 2011:74(10):2060–2070. https://doi.org/10.1016/j.jprot.2011.05. 028 Lê Cao K, Martin PGP, Robert-Granié C, Besse P. Sparse canonical methods for biological data integration: application to a crossplatform study. BMC Bioinformatics. 2009:10(1):34. https://doi.org/ 10.1186/1471-2105-10-34 Ludwig C, Gillet L, Rosenberger G, Amon S, Collins BC, Aebersold R. Data-independent acquisition-based SWATH-MS for quantitative proteomics: a tutorial. Mol Syst Biol. 2018:14(8):e8126. https://doi.org/10.15252/msb.20178126 Luo W, Brouwer C. Pathview: an R/Bioconductor package for pathwaybased data integration and visualization. Bioinformatics. 2013:29(14): 1830–1831. https://doi.org/10.1093/bioinformatics/btt285 Ma D, Przybylski D, Abruzzi KC, Schlichting M, Li Q, Long X, Rosbash M. A transcriptomic taxonomy of drosophila circadian neurons around the clock. Elife. 2021:10:1–19. https://doi.org/10.7554/eLife. 63056 Masojídek J, Ranglová K, Lakatos GE, Benavides AMS, Torzillo G. Variables governing photosynthesis and growth in microalgae mass cultures. Process. 2021:9(5):820. https://doi.org/10.3390/ pr9050820 Miller BH, McDearmon EL, Panda S, Hayes KR, Zhang J, Andrews JL, Antoch MP, Walker JR, Esser KA, Hogenesch JB, et al. Circadian and CLOCK-controlled regulation of the mouse transcriptome and cell proliferation. Proc Natl Acad Sci U S A. 2007:104(9): 3342–3347. https://doi.org/10.1073/pnas.0611724104 Mittag M. Circadian rhythms in microalgae. Int Rev Cytol. 2001:206: 213–247. https://doi.org/10.1016/S0074-7696(01)06023-5 Mondal A, Banerjee S. Effect of productivity and seasonal variation on phytoplankton intermittency in a microscale ecological study using closure approach. Sci Rep. 2022:12(1):5939. https://doi.org/ 10.1038/s41598-022-09420-5 Monnier A, Liverani S, Bouvet R, Jesson B, Smith JQ, Mosser J, Corellou F, Bouget FY. Orchestrated transcription of biological processes in the marine picoeukaryote Ostreococcus exposed to light/dark cycles. BMC Genomics. 2010:11(1):192. https://doi.org/10.1186/ 1471-2164-11-192 Moulager M, Monnier A, Jesson B, Bouvet R, Mosser J, Schwartz C, Garnier L, Corellou F, Bouget FY. Light-dependent regulation of cell division in Ostreococcus: evidence for a major transcriptional input. Plant Physiol. 2007:144(3):1360. https://doi.org/10.1104/pp. 107.096149 Müller LM, Mombaerts L, Pankin A, Davis SJ, Webb AAR, Goncalves J, Von Korff M. Differential effects of day/night cues and the circadian clock on the barley transcriptome. Plant Physiol. 2020:183(2): 765–779. https://doi.org/10.1104/pp.19.01411 O-Neill JS, Van Ooijen G, Dixon LE, Troein C, Corellou F, Bouget FY, Reddy AB, Millar AJ. Circadian rhythms persist without transcription in a eukaryote. Nature. 2011:469(7331):554–558. https://doi. org/10.1038/nature09654 Palenik B, Grimwood J, Aerts A, Rouzé P, Salamov A, Putnam N, Dupont C, Jorgensen R, Derelle E, Rombauts S, et al. The tiny eukaryote Ostreococcus provides genomic insights into the paradox of plankton speciation. Proc Natl Acad Sci U S A. 2007:104(18): 7705–7710. https://doi.org/10.1073/pnas.0611046104 Parsons R, Parsons R, Garner N, Oster H, Rawashdeh O. CircaCompare: a method to estimate and statistically support differences in mesor, amplitude and phase, between circadian rhythms. Bioinformatics. 2020:36(4):1208–1212. https://doi.org/10. 1093/bioinformatics/btz730 24 | The Plant Cell, 2025, Vol. 37, No. 2 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025 Patke A, Young MW, Axelrod S. Molecular mechanisms and physiological importance of circadian rhythms. Nat Rev Mol Cell Biol. 2020:21(2):67–84. https://doi.org/10.1038/s41580-019-0179-2 Perez-Riverol Y, Bai J, Bandla C, García-Seisdedos D, Hewapathirana S, Kamatchinathan S, Kundu DJ, Prakash A, Frericks-Zipper A, Eisenacher M, et al. The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Res. 2022:50(D1):D543–D552. https://doi.org/10.1093/nar/ gkab1038 Petersen J, Rredhi A, Szyttenholm J, Mittag M. Evolution of circadian clocks along the green lineage. Plant Physiol. 2022:190(2):924. https://doi.org/10.1093/plphys/kiac141 Pittendrigh CS. Temporal organization: reflections of a Darwinian clockwatcher. Annu Rev Physiol. 1993:55(1):16–54. https://doi.org/ 10.1146/annurev.ph.55.030193.000313 Ral JP, Derelle E, Ferraz C, Wattebled F, Farinas B, Corellou F, Buléon A, Slomianny MC, Delvalle D, d’Hulst C, et al. Starch division and partitioning. A mechanism for granule propagation and maintenance in the picophytoplanktonic green alga Ostreococcus tauri. Plant Physiol. 2004:136(2):3333–3340. https://doi.org/10.1104/pp. 104.044131 Robbens S, Khadaroo B, Camasses A, Derelle E, Ferraz C, Inzé D, Van De Peer Y, Moreau H. Genome-wide analysis of core cell cycle genes in the unicellular green alga Ostreococcus tauri. Mol Biol Evol. 2005:22(3):589–597. https://doi.org/10.1093/molbev/msi044 Robles MS, Cox J, Mann M. In-vivo quantitative proteomics reveals a key contribution of post-transcriptional mechanisms to the circadian regulation of liver metabolism. PLoS Genet. 2014:10(1): e1004047. https://doi.org/10.1371/journal.pgen.1004047 Rohart F, Gautier B, Singh A, Lê Cao K. mixOmics: an R package for omics feature selection and multiple data integration. PLoS Comput Biol. 2017:13(11):e1005752. https://doi.org/10.1371/journal. pcbi.1005752 Romero-Losada AB, Arvanitidou C, de los Reyes P, García-González M, Romero-Campero FJ. ALGAEFUN with MARACAS, micro ALGAE FUNctional enrichment tool for MicroAlgae RnA-seq and Chip-seq AnalysiS. BMC Bioinformatics. 2022:23(1):113. https:// doi.org/10.1186/s12859-022-04639-5 Sands E, Davies S, Puxty RJ, Vergé V, Bouget F-Y, Scanlan DJ, Carré IA. Genetic and physiological responses to light quality in a deep ocean ecotype of Ostreococcus, an ecologically important photosynthetic picoeukaryote. J Exp Bot. 2023:74(21):6773–6789. https:// doi.org/10.1093/jxb/erad347 Sanz-Luque E, Chamizo-Ampudia A, Llamas A, Galvan A, Fernandez E. Understanding nitrate assimilation and its regulation in microalgae. Front Plant Sci. 2015:6:899. https://doi.org/10.3389/fpls.2015. 00899 Seaton DD, Graf A, Baerenfaller K, Stitt M, Millar AJ, Gruissem W. Photoperiodic control of the Arabidopsis proteome reveals a translational coincidence mechanism. Mol Syst Biol. 2018:14(3): e7962. https://doi.org/10.15252/msb.20177962 Serrano G, Herrera-Palau R, Romero JM, Serrano A, Coupland G, Valverde F. Chlamydomonas CONSTANS and the evolution of plant photoperiodic signaling. Curr Biol. 2009:19(5):359–368. https://doi.org/10.1016/j.cub.2009.01.044 Serrano-Bueno G, Romero-Campero FJ, Lucas-Reina E, Romero JM, Valverde F. Evolution of photoperiod sensing in plants and algae. Curr Opin Biol. 2017:37:10–17. https://doi.org/10.1016/j.pbi.2017. 03.007 Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003:13(11):2498–2504. https://doi.org/10.1101/gr. 1239303 Shi Q, Chen C, He T, Fan J. Circadian rhythm promotes the biomass and amylose hyperaccumulation by mixotrophic cultivation of marine microalga Platymonas helgolandica. Biotechnol Biofuels Bioprod. 2022:15(1):75. https://doi.org/10.1186/s13068-022-02174-2 Sorokina O, Corellou F, Dauvillée D, Sorokin A, Goryanin I, Ball S, Bouget FY, Millar AJ. Microarray data can predict diurnal changes of starch content in the picoalga Ostreococcus. BMC Syst Biol. 2011:5(1):36. https://doi.org/10.1186/1752-0509-5-36 Thaben PF, Westermark PO. Detecting rhythms in time series with rain. J Biol Rhythms. 2014:29(6):391–400. https://doi.org/10.1177/ 0748730414553029 van Baren MJ, Bachy C, Reistetter EN, Purvine SO, Grimwood J, Sudek S, Yu H, Poirier C, Deerinck TJ, Kuo A, et al. Evidence-based green algal genomics reveals marine diversity and ancestral characteristics of land plants. BMC Genomics. 2016:17(1):267. https://doi.org/ 10.1186/s12864-016-2585-6 van Ooijen G, Dixon LE, Troein C, Millar AJ. Proteasome function is required for biological timing throughout the twenty-four hour cycle. Curr Biol. 2011:21(10):869–875. https://doi.org/10.1016/j. cub.2011.03.060 Wang C, Shui K, Ma S, Lin S, Zhang Y, Wen B, Deng W, Xu H, Hu H, Guo A, et al. Integrated omics in Drosophila uncover a circadian kinome. Nat Commun. 2020:11(1):1–15. https://doi.org/10.1038/ s41467-019-13993-7 Wang L, Zhou A, Li J, Yang M, Bu F, Ge L, Chen L, Huang W. Circadian rhythms driving a fast-paced root clock implicate species-specific regulation in Medicago truncatula. J Integr Plant Biol. 2021:63(8): 1537–1554. https://doi.org/10.1111/jipb.13138 Wang W, Sijacic P, Xu P, Lian H, Liu Z. Arabidopsis TSO1 and MYB3R1 form a regulatory module to coordinate cell proliferation with differentiation in shoot and root. Proc Natl Acad Sci U S A. 2018:115:3045–3054. https://doi.org/10.1073/pnas.1715903115 Watson JV, Chambers SH, Smith PJ. A pragmatic approach to the analysis of DNA histograms with a definable G1 peak. Cytometry. 1987:8(1):1–8. https://doi.org/10.1002/cyto.990080101 Weger BD, Gobet C, David FPA, Atger F, Martin E, Phillips NE, Charpagne A, Weger M, Naef F, Gachon F. Systematic analysis of differential rhythmic liver gene expression mediated by the circadian clock and feeding rhythms. Proc Natl Acad Sci U S A. 2021: 118(3):e2015803118. https://doi.org/10.1073/pnas.2015803118 Wenden B, Toner DLK, Hodge SK, Grima R, Millar AJ. Spontaneous spatiotemporal waves of gene expression from biological clocks in the leaf. Proc Natl Acad Sci U S A. 2012:109(17):6757–6762. https://doi.org/10.1073/pnas.1118814109 Willforss J, Chawade A, Levander F. NormalyzerDE: online tool for improved normalization of omics expression data and highsensitivity differential expression analysis. J Proteome Res. 2019: 18(2):732–740. https://doi.org/10.1021/acs.jproteome.8b00523 Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, et al. clusterProfiler 4.0: a universal enrichment tool for interpreting omics data. Innovation. 2021:2(3):100141. https://doi.org/ 10.1016/j.xinn.2021.100141 Yakir E, Hassidim M, Melamed-Book N, Hilman D, Kron I, Green RM. Cell autonomous and cell-type specific circadian rhythms in Arabidopsis. Plant J. 2011:68(3):520–531. https://doi.org/10.1111/j. 1365-313X.2011.04707.x Zones JM, Blaby IK, Merchant SS, Umen JG. High-resolution profiling of a synchronized diurnal transcriptome from Chlamydomonas reinhardtii reveals continuous cell and metabolic differentiation. Plant Cell. 2015:27(10):2743–2769. https://doi.org/10.1105/tpc.15. 00498 Photoperiodic plasticity in marine phytoplankton rhythms | 25 Downloaded from https://academic.oup.com/plcell/article/37/2/koaf033/8009039 by guest on 27 March 2025