Heat Stress Response in the Closest Algal Relatives of Land Plants Reveals

NRC Publications Archive Archives des publications du CNRC

Heat stress response in the closest algal relatives of land plants reveals conserved stress signaling circuits de Vries, Jan; de Vries, Sophie; Curtis, Bruce A.; Zhou, Hong; Penny, Susanne; Feussner, Kirstin; Pinto, Devanand M.; Steinert, Michael; Cohen, Alejandro M.; Schwartzenberg, Klaus; Archibald, John M.

This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur. For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

Publisher’s version / Version de l'éditeur: https://doi.org/10.1111/tpj.14782 The Plant Journal, pp. 1-24, 2020-04-24

NRC Publications Archive Record / Notice des Archives des publications du CNRC : https://nrc-publications.canada.ca/eng/view/object/?id=e5464b82-883a-46f0-8b15-9261477a4416 https://publications-cnrc.canada.ca/fra/voir/objet/?id=e5464b82-883a-46f0-8b15-9261477a4416

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at https://nrc-publications.canada.ca/eng/copyright READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site https://publications-cnrc.canada.ca/fra/droits LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

Questions? Contact the NRC Publications Archive team at [email protected]. If you wish to email the authors directly, please see the first page of the publication for their contact information.

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à [email protected]. The Plant Journal (2020) doi: 10.1111/tpj.14782 Heat stress response in the closest algal relatives of land plants reveals conserved stress signaling circuits

Jan de Vries1,2,3,4,*,† , Sophie de Vries1,5,† , Bruce A. Curtis1, Hong Zhou6, Susanne Penny7, Kirstin Feussner8,9 , Devanand M. Pinto7,10, Michael Steinert2, Alejandro M. Cohen11 , Klaus von Schwartzenberg6 and John M. Archibald1,12,* 1Department of Biochemistry and Molecular Biology, Dalhousie University, Sir Charles Tupper Medical Building, 5850 College Street, Halifax, NS, B3H 4R2, Canada, 2Institute of Microbiology, Technische Universitat€ Braunschweig, Spielmannstr. 7, 38106, Braunschweig, Germany, 3Department of Applied Bioinformatics, Institute for Microbiology and Genetics, University of Goettingen, Goldschmidtstr. 1, 37077, Goettingen, Germany, 4Goettingen Center for Molecular Biosciences (GZMB), University of Goettingen, 37077, Goettingen, Germany, 5Institute of Population Genetics, Heinrich-Heine University Duesseldorf, Universitatsstr.€ 1, 40225, Duesseldorf, Germany, 6Microalgae and Zygnematophyceae Collection Hamburg (MZCH) and Aquatic Ecophysiology and Phycology, Institute of Plant Science and Microbiology, Universitat€ Hamburg, 22609, Hamburg, Germany, 7National Research Council, Human Health Therapeutics, 1411 Oxford Street, Halifax, NS, B3H 3Z1, Canada, 8Department of Plant Biochemistry, Albrecht-von-Haller-Institute for Plant Sciences, University of Goettingen, Justus-von-Lie- big-Weg 11, 37077, Goettingen, Germany, 9Service Unit for Metabolomics and Lipidomics, Goettingen Center for Molecular Biosciences (GZMB), 37077, Goettingen, Germany, 10Department of Chemistry, Dalhousie University, 6274 Coburg Rd, Halifax, NS, B3H 4R2, Canada, 11Biological Spectrometry Core Facility, Life Sciences Research Institute, Dalhousie University, Halifax, NS, B3H 4R2, Canada, and 12Canadian Institute for Advanced Research, 661 University Ave, Suite 505, Toronto, ON, M5G 1M1, Canada

Received 20 December 2019; revised 28 March 2020; accepted 8 April 2020. *For correspondence (e-mails: [email protected]; [email protected]). †These authors contributed equally to this work.

SUMMARY All land plants (embryophytes) share a common ancestor that likely evolved from a filamentous freshwater alga. Elucidating the transition from algae to embryophytes – and the eventual conquering of Earth’s surface – is one of the most fundamental questions in plant evolutionary biology. Here, we investigated one of the organismal properties that might have enabled this transition: resistance to drastic temperature shifts. We explored the effect of heat stress in Mougeotia and Spirogyra, two representatives of Zygnemato- phyceae – the closest known algal sister lineage to land plants. Heat stress induced pronounced phenotypic alterations in their plastids, and high-performance liquid chromatography-tandem mass spectroscopy-based profiling of 565 transitions for the analysis of main central metabolites revealed significant shifts in 43 compounds. We also analyzed the global differential gene expression responses triggered by heat, generating 92.8 Gbp of sequence data and assembling a combined set of 8905 well-expressed genes. Each organism had its own distinct gene expression profile; less than one-half of their shared genes showed concordant gene expression trends. We nevertheless detected common signature responses to heat such as elevated transcript levels for molecular chaperones, thylakoid components, and – corroborating our metabolomic data – amino acid metabolism. We also uncovered the heat-stress responsiveness of genes for phosphorelay-based signal transduction that links environmental cues, calcium signatures and plastid biology. Our data allow us to infer the molecular heat stress response that the earliest land plants might have used when facing the rapidly shifting temperature conditions of the terrestrial habitat.

Keywords: early plant evolution, stress physiology, streptophyte algae, plant terrestrialization, signal transduction, charophytes, heat stress, RNA-seq, metabolomics.

© 2020 The Authors. 1 The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. 2 Jan de Vries et al.

INTRODUCTION substrate and establish the terrestrial flora as we know it (Becker and Marin, 2009; Delwiche and Cooper, 2015; de Land plants evolved from streptophyte algae. Comparing Vries and Archibald, 2018; Rensing, 2018; Furst-Jansen€ data gleaned from streptophyte algae and land plants et al., 2020). One of these challenges was rapid but drastic allows for the inference of properties of the progenitors of shifts in temperature (de Vries and Archibald, 2018). Land embryophytes – from which organisms that eventually plants face these challenges on a frequent basis and gave rise to the entire terrestrial macroflora evolved (Del- evolved an elaborate temperature stress response (Chin- wiche and Cooper, 2015; de Vries and Archibald, 2018). nusamy et al., 2007; Ohama et al., 2017). They process Yet, there are several classes of streptophyte algae that live these environmental inputs using the plant perceptron, in freshwater and terrestrial habitats (Lewis and McCourt, which integrates external cues into a complex array of sig- 2004), the habitats in which the land plant progenitors naling molecules – such as phytohormones – and regula- likely dwelled (Becker and Marin, 2009; Harholt et al., tory switches (Scheres and van der Putten, 2017). One of 2016). Furthermore, trait evolution across the streptophyte these environmental cues is heat. algae paints a complicated picture (Delwiche and Cooper, Heat stress has several biophysical consequences that 2015; Delwiche, 2016; de Vries and Archibald, 2018), and affect the biology of all cells. These include increased which specific streptophyte algal lineages have the poten- membrane fluidity and the accumulation of misfolded pro- tial to be the most informative is hence a key question. teins (Murata and Los, 1997; Kotak et al., 2007). Cells Major phylogenomic efforts have thus been devoted to respond to these heat-induced maladies using the appro- clarifying the relationships among streptophytes. Phyloge- priate counter-action. Membrane fluidity is counteracted netic analyses based on chloroplast (Turmel et al., 2006; using fatty acid desaturases (Falcone et al., 2004) and mis- Ruhfel et al., 2014) and nuclear genomic data (Wodniok folded proteins are dealt with by molecular chaperones, et al., 2011; Wickett et al., 2014; Puttick et al., 2018; Lee- many of which are also known as heat shock proteins bens-Mack et al., 2019) have consistently recovered the (HSPs) (Nakamoto and Vıgh, 2007; Kotak et al., 2007). Zygnematophyceae as the closest algal relatives to land Mounting the molecular response to temperature stress is plants. This has important implications for inferring the based on intricate signaling pathways. These include the nature of the last common ancestor of land plants and transcription factor (TF) network of heat shock factors algae – and hence the earliest land plants themselves. (Ohama et al., 2017); such generic responses are hard- The streptophyte algae can be split into two grades: the wired into the molecular biology of all cellular life (Feder Mesostigmatophyceae, Chlorokybophyceae and Kleb- and Hofmann, 1999). In land plant and algal cells, the plas- sormidiophyceae (KCM) (de Vries et al., 2016) and the tid (chloroplast) must also be considered. Indeed, a recent Zygnematophyceae, Coleochaetophyceae and Charo- study highlighted that the cyclopentenones dinor-12-oxo- phyceae (ZCC) (de Vries et al., 2016). The ZCC grade and 10,15(Z)-phytodienoic acid (dn-OPDA) and OPDA, which land plants form the monophyletic Phragmoplastophyta are synthesized in the plastid, provide tolerance to heat (Lecointre and Le Guyader, 2006). Among all phragmoplas- stress in the land plants Arabidopsis thaliana and tophytic streptophyte algae, the Zygnematophyceae exhi- Marchantia polymorpha, as well as the streptophyte alga bit the least morphological complexity (see also Klebsormidium nitens (Monte et al., 2020). discussions in Wodniok et al., 2011 and Wickett et al., Plastids are a major site of abiotic stress response 2014). This illustrates some of the challenges in inferring (Fernandez and Strand, 2008; Chan et al., 2016). Because the early evolution of plants (recently reviewed in Cooper most of the proteins found in green plastids are encoded and Delwiche [2015] as well as de Vries and Archibald, in the nucleus (Timmis et al., 2004; Ferro et al., 2010; Tera- 2018). Yet, as previously highlighted (de Vries et al., 2018), shima et al., 2011), the plastid communicates with the Zygnematophyceae possess specific physiological proper- nucleus to adequately respond to external cues (Chan ties that appear to tie them to land plants. These include, et al., 2016). The molecular biological chassis underpinning for example, a complete set of genes that is homologous plastid retrograde signaling has been the subject of much to a signaling cascade required for transducing the stress recent debate and conceptual reordering (Page et al., 2017; phytohormone ABA of land plants (de Vries et al., 2018). Song et al., 2018; Wu et al., 2018; Zhao et al., 2018; Kacpr- Zygnematophyceaen physiology is further relevant in the zak et al., 2019; Shimizu et al., 2019). These discussions context of a successful establishment on land because they nevertheless revolve around a (relatively) robust core set exhibit, among other noteworthy properties, astounding of proteins including those that are tetrapyrrole biosynthe- levels of desiccation tolerance (Holzinger and Karsten, sis-associated (Chan et al., 2016; Hernandez-Verdeja and 2013; Pichrtova et al., 2016; Rippin et al., 2017; Herburger Strand, 2018). Several carotenoid-derived metabolites et al., 2019). (Ramel et al., 2012; Xiao et al., 2012; Avendano-V~ azquez The first land plants must have successfully overcome a et al., 2014; Kleine and Leister, 2016) and the phosphonu- barrage of terrestrial stressors in order to rise above their cleotide 30-phosphoadenosine 50-phosphate (Estavillo

© 2020 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2020), doi: 10.1111/tpj.14782 Heat stress in land plants’ closest relatives 3 et al., 2011) have also been added. Recent genomic and and Marion-Poll, 2005). The presence of ABA in the cytosol transcriptomic data suggest that (i) most of the genes is perceived by the PYRABACTIN RESISTANCE(-LIKE)/REG- required for canonical land plant-like retrograde signaling ULATORY COMPONENT OF ABA RECEPTOR (PYR/PYL/ were already present in ZCC-grade streptophyte algae RCAR receptor) (Ma et al., 2009; Park et al., 2009). Once (Nishiyama et al., 2018; Zhao et al., 2019) and (ii) these ABA is bound by this receptor, it inhibits the activity of the genes respond to high light and cold stress (de Vries et al., phosphatases of the PROTEIN PHOSPHATASE 2C (PP2C) 2018). family, including ABA INSENSITIVE 1 (ABI1) and HYPER- Heat stress also impairs plastid physiology. Studies in SENSITIVE TO ABA1 (HAB1) (Rubio et al., 2009; Nishimura land plants and the green alga Chlamydomonas reinhardtii et al., 2010). The inhibition of the PP2Cs then results in the show that heat-induced perturbations of plastid physiology release of the SUCROSE NONFERMENTING 1-RELATED include damage to reactions at the thylakoid membrane PROTEIN KINASEs (SnRKs), including the well-known pro- (Berry and Bjorkman,€ 1980; Sharkey, 2005), as well as tein OPEN STOMATA 1 (OST1, i.e. SnRK2.6) (Vlad et al., reduced abundance (Hemme et al., 2014) and inhibition 2009; Umezawa et al., 2009). The SnRK kinases are then (Tewari and Tripathy, 1998) of chlorophyll biosynthesis free to phosphorylate various downstream targets, includ- enzymes. It is hence not surprising that heat stress is con- ing TFs such as ABSCISIC ACID RESPONSIVE ELEMENT- nected to adjustments of plastid physiology on multiple BINDING FACTORs (ABFs) and ion channels such as SLOW levels (Lin et al., 2014; Wang et al., 2014). Interestingly, ANION CHANNEL 1 (SLAC1) (Furihata et al., 2006; Soon heat induces calcium signatures in the plastid (Stael et al., et al., 2012). 2011; Lenzoni and Knight, 2018); indeed, among the prime The core set of genes that constitutes the ABA-triggered signaling components triggered by heat are Ca2+-mediated signaling loop appears conserved across land plants (Ume- processes. zawa et al., 2010; Hauser et al., 2011; Eklund et al., 2018; Experiments with angiosperms and moss show that heat Sussmilch et al., 2019). Recently, we have noticed that the stress induces an influx of calcium (Ca2+) into the cyto- zygnematophyceaen alga Zygnema circumcarinatum has plasm that generates a characteristic Ca2+ profile (Klein all the genes to potentially use this signaling cascade – and Ferguson, 1987; Gong et al., 1998; Saidi et al., 2009; including a gene homologous to the PYR/PYL/RCAR genes Finka et al., 2012); experimental tampering with these Ca2+ of land plants (de Vries et al., 2018). The presence of PYR/ signatures by inhibiting the action of calcium channels PYL/RCAR genes in Zygnematophyceae was confirmed decreases the thermotolerance of plants (Larkindale and through sequencing of the nuclear genome of Mesotae- Knight, 2002). Plant cells interpret Ca2+ signatures using a nium endlicherianum (Cheng et al., 2019). Zygnemato- core set of well-defined gene families (Edel and Kudla, phyceae are hence the first algal group shown to have the 2015; Edel et al., 2017). Among these is the CALCIUM- complete gene set for ABA-mediated signaling. Our most DEPENDENT PROTEIN KINASE (CPK/CDPK) family (Hrabak recent functional data on Z. circumcarinatum, however, et al., 2003; Reddy et al., 2012), whose members are show that, although zygnematophyceaen PYL does regu- involved in responses to a variety of biotic and abiotic late the downstream phosphatases, it does so in an stressors (Schulz et al., 2013). CDPKs consist of a Ca2+ ABA-independent manner (Sun et al., 2019). Hence, the sensing domain with a kinase output domain (Harper cascade likely acts in the modulation of stress physiology, et al., 1994; Harmon et al., 2000). Upon Ca2+-dependent although the regulating agent is unknown. Nonetheless, activation, they phosphorylate downstream targets such as the presence and partially shared functionality makes the TFs that control stress response and intersect with hor- capacities of the molecular stress physiology of Zygne- mone-mediated signaling, such as by ABA (Kaplan et al., matophyceae of particular relevance for the evolution of 2006; Lynch et al., 2012; Edel and Kudla, 2016). Land plant the physiology of land plants (see also discussions in genomes usually encode numerous CDPKs (e.g. 10 in the Furst-Jansen€ et al., 2020). lycophyte Selaginella moellendorffii;34inA. thaliana) Here, we investigated the molecular stress response of (Edel and Kudla, 2015); streptophyte algae also harbor mul- Zygnematophyceae by analyzing two representative spe- tiple CDPK-encoding genes (Edel and Kudla, 2015; cies that likely diverged at an early time point, thus span- Nishiyama et al., 2018). ning close to 500 million years of zygnematophyceaen Calcium-mediated signaling is not only important upon evolution (dating based on Morris et al., 2018). We uncov- heat stress, it is also one of the classical broad-spectrum ered ubiquitous signature responses to heat, including responses towards a variety of stressors (Reddy et al., metabolomic changes in amino acid levels and the induc- 2012). Another such broad-spectrum response are signal- tion of global gene expression patterns associated with ing processes mediated by the stress phytohormone ABA biomolecule homeostasis. Importantly, we found the (Zhu, 2016). In land plants, ABA is synthesized from carote- responsiveness of putative components in signal transduc- noids; most of these key steps occur in the plastid, with tion to be conserved between land plants and their closest the finishing touches occurring in the cytosol (Nambara algal relatives.

RESULTS AND DISUSSION 228.6 Æ 82.9 µm2 (P = 3.9 9 10À15), whereas the change in area of the cell was less pronounced, from 813.1 Æ 180.6 Heat stress induces changes in plastid morphology in two µm2 to 763.1 Æ 196.4 µm2 (P = 0.063). In relative terms, this Zygnematophyceae means a change in plastid area from 63.9 Æ 5.8% to The Zygnematophyceae represents the closest algal sister 30.8 Æ 10.6% (P < 2.2 9 10À16). Furthermore, Mougeotia’s group of land plants (Wickett et al., 2014). We used two plastids were often visibly contorted, giving them a fringed representative species of filamentous Zygnematophyceae, and jagged appearance. Mougeotia sp. (MZCH240) and Spirogyra pratensis In S. pratensis, heat stress induced a variety of drastic (MZCH10213); these two species span some of the deepest changes of the overall morphology of any given plastid. branches of the phylogeny of Zygnematophyceae More than 95% (n = 67) of the plastids of heat-treated Spir- (Gontcharov and Melkonian, 2010; Wickett et al., 2014; ogyra filaments had lost their characteristic spiral arrange- Leebens-Mack et al., 2019). Mougeotia is characterized by a ment (Figure 1a). Among these 95%, the spiral plastid was plate-like plastid that takes up most of the cell; S. pratensis sometimes vaguely apparent, although it was less tightly instead has a twisted, helical plastid. Both species grow as spiraled, yet often the plastid had an entirely irregular predominantly unbranched filaments – despite the fact that shape. Concomitantly, the pigmentation of the plastid branching can occur – and are ubiquitous in most freshwa- often changed from grass green towards orange. Further- ter habitats. more, the entire cellular content of heat-treated Spirogyra We cultivated both species at 22°C and cells appeared more granular. 120 µmol quanta mÀ2 secÀ1 from a light-emitting diode Because both species showed alterations in plastid mor- (LED) light source (photosynthetically active radiance), phology, we made use of the autofluorescence of the plas- which is referred to as the control condition. Drastic shifts tid and used confocal microscopy to more finely pinpoint in temperature are a known terrestrial stressor (de Vries the structural changes of the plastid associated with heat and Archibald, 2018). To investigate the effect of tempera- stress treatment. For Mougeotia sp. the confocal micro- ture stress, we used heat stress. We shifted the algae from graphs confirmed the huddled and jagged appearance of the control condition to 37°C (but same light regime, i.e. the plastids. In the case of S. pratensis, this revealed that 120 µmol quanta mÀ2 secÀ1) for 24 h and used microscopy the pronounced changes upon heat stress in gross mor- to assess their phenotypic responses (Figure 1). phology were indeed a result of plastid ‘unwinding’. This Mougeotia sp. and S. pratensis both showed phenotypic looser corkscrew shape (i.e. loss of some twists and turns) alterations upon heat stress treatment (Figure 1a). In was even more apparent under fluorescence. Moreover, Mougeotia sp., the plastids appeared more clumped S. pratensis plastids often had (granular) specks of high together and changed significantly (P = 2.2 9 10À16)in fluorescence and a more jagged surface upon heat treat- length from an average of 48.1 Æ 9.2 µm to 24.7 Æ 7.5 µm ment. Overall, our observations reveal that heat stress (Figure 1b); the length of the cells themselves did not impacted both species and morphological alterations were change significantly from an average of 64.4 Æ 12.8 µmto most obvious in their plastids. 60.6 Æ 14.3 µm(P = 0.063). Because we obtained tractable Mougeotia and Spirogyra show reduced differential (and thus comparable) measures for each given cell and transcript abundance of house-keeping genes upon heat plastid exposed to control (n = 40) and heat (n = 56) condi- stress tions, we calculated the relative length the plastid occupied in relation to the total length of the cell. The relative length To determine how these two Zygnematales respond to of the plastids decreased significantly from an average of heat on a molecular level, we performed comparative tran- 74.8 Æ 4.8% to 41.9 Æ 13.0% (P < 2.2 9 10À16). The width scriptome sequencing using the NovaSeq 6000 platform of the plastids showed pronounced heat-induced fluctua- (Illumina, San Diego, CA, USA; operated by Genome tions. On average, however, plastid width merely, but sig- Quebec, Montreal, Canada). RNA was extracted for each nificantly, decreased from 10.7 Æ 1.2 µm to 9.3 Æ 1.8 µm species and condition from three (and in the case of (P = 7.2 9 10À6), whereas cell width stayed exactly the Mougeotia sp. control samples six) biological replicates. In same (12.6 Æ 1.2 µm and 12.6 Æ 1.2 µm in control and total, we obtained 232 911 820 and 226 275 398 paired-end heat-treated cells, respectively; P = 0.99). For the relative reads (47.05 and 45.71 Gbp) with a length of 100 bp for width of the plastid (in relation to the width of the cell), this Mougeotia sp. and S. pratensis, respectively (Data S1). meant a decrease from 85.5 Æ 7.0% to 74.0 Æ 11.7% After (i) filtering for sequence quality, (ii) de novo assem- (P = 4.8 9 10À7). Having obtained length and width mea- bly, and (iii) filtering for high expression level as well as sures, we calculated the area of the plastid, assuming a taxonomic affiliation (for details, see Material and Meth- non-square rectangular geometry. As expected from the ods), we obtained 4955 contigs for Mougeotia sp. and 3950 width and length measurements, the area of the plastid contigs for S. pratensis. For each gene (determined by the decreased significantly from 518.5 Æ 120.6 µm2 to TRINITY clustering algorithm) (Haas et al., 2013), we

Figure 1. Heat stress induces alterations in the (a) Light microscopy Confocal microscopy morphology of plastids in Mougeotia sp. and Spir- Control Heat stress Control Heat stress ogyra pratensis. (a) Light (Nomarski) and confocal micrographs of samples heat-treated for 24 h at 37°C, as well as control samples maintained at 22°C. In the confocal micrographs, the cells of the filaments were visual- sp. MZCH 240 ized with 1% calcofluor white staining (teal false colored); the plastids were visible as a result of Mougeotia chlorophyll a autofluorescence (false-colored in a 200 μm 100 μm 200 μm 50 μm 50 μm red-to-orange gradient). (b) Quantification of plastid morphological alterations in Mougeotia sp. elicited by heat treatment MZCH 1013 for 24 h at 37°C. Data were visualized as box plots that show their interquartile range (IQR; 50 Æ 25%). The horizontal lines in each of the boxes indicate the median (50%); the dotted whiskers span to the 200 μm 100 μm 200 μm 100 μm furthest data points within the 1.5 9 IQR range. Spirogyra pratensis 20 μm Outliers are indicated by circles. Statistics were per- Mougeotia sp. MZCH 240 formed using either Mann–Whitney U-ort-tests (b) Cell Plastid Relative size of the plastid (plastid/cell) 2 2 (two sample or Welch’s, depending on the vari- Length [μm] Width [μm] Area [μm ] Length [μm] Width [μm] Area [μm ] Length Width Area ns ns ns ance). ***P < 0.001. *** *** *** *** *** *** Heat (n = 56) 0.4 0.5 0.6 0.7 0.8 Control (n = 40) 0.3 6 8 10 12 0.5 0.6 0.7 0.8 0.9 1 9101112131415 10 20 30 40 50 60 0.2 400 600 800 1000 1200 100 200 300 400 500 600 700 40 50 60 70 80 90 0.1 0.2 0.3 0.4 0.5 0.6 0.7 defined the transcript with the highest gene expression as pathways that exhibited differential regulation, 32 and 43 the representative isoform. Additionally, we applied an had 10 or more differentially regulated orthologs (up or expression-based cut-off by retaining only those genes that down regulation) in Mougeotia sp. and S. pratensis, had an average transcripts per million (TPM) over all the respectively (Figure 2b). The KEGG pathway with the most libraries that was higher than the number of libraries differentially regulated orthologs included information pro- sequenced. The representative isoforms were in silico cessing such as ‘Ribosome’ (three and two up-regulated, translated using a custom perl script that extracted the best 47 and 29 down-regulated in Mougeotia sp. and S. praten- protein alignments obtained by BLASTX (Altschul et al., sis, respectively), ‘Spliceosome’ (nine and zero up, 15 and 1990) screening against protein data from selected chloro- 41 down) and ‘RNA transport’ (six and zero up, 16 and 35 phyte and streptophyte genomes; protein data were used down). We interpret these as signatures of the necessity to to link annotation with the differential gene expression val- implement major shifts in molecular/cell biology and phys- ues that were calculated via EDGER (Robinson et al., 2010) iology in response to heat. (Data S2). Heat stress perturbs protein homeostasis. A common To obtain a global perspective on the influence of heat theme across all three domains of life, thus, is an up-regu- on the two Zygnematales, we used the pathways in the lation of molecular chaperones: the HSPs (Lindquist and Kyoto Encyclopedia of Genes and Genomes (KEGG) to Craig, 1988). In KEGG, many HSPs fall into the umbrella contextualize our gene expression values. For this, we first category of ‘Protein processing in the ER’, which hence identified KEGG orthologs among our de novo assembled showed the second-most and most up-regulated orthologs protein-coding transcripts using BLASTKOALA (Kanehisa et al., in Mougeotia sp. (seven up-regulated and six down-regu- 2016); if multiple transcripts were annotated as the same lated) and S. pratensis (six up-regulated and 26 down-reg- KEGG ortholog, we added up their gene expression values. ulated), respectively (Figure 2c). The finding of ‘Protein In total, we identified 1178 and 1000 unique KEGG ortho- processing in the ER’ as a differentially responding KEGG logs in Mougeotia sp. and S. pratensis that mapped onto category also makes sense in a broader context. As in plant KEGG pathways. In Mougeotia sp., heat induced an other eukaryotes, in plants and algae there is a connection up-regulation (log2[fc] ≥ 1) of 116 and down-regulation between heat stress, ER response and an accumulation of

(log2[fc] ≤ –1) of 439 KEGG orthologs; in S. pratensis,a unfolded proteins (Deng et al., 2011; Mittler et al., 2012; mere 22 KEGG orthologs showed up-regulation, whereas Perez-Mart ın et al., 2014; Schroda et al., 2015). In the con- 625 displayed down-regulation (Figure 2a). Of the KEGG text of misfolded protein accumulation, it is thus not

(a) number of KEGG orthologs up down regulated: 5101520253032 ≥86420 024681012141618≥20 116 623 439 Mougeotia up sp. down Spirogyra up unchanged pratensis down up down 22 353 625 (b) 50

number of KEGG orthologs regulated in: 40 Mougeotia sp. Spirogyra pratensis up up 30 down down

10 number of KEGG orthologs regulated 0 Ribosome Peroxisome Phagosome Endocytosis Spliceosome RNA transport Photosynthesis RNA degradation Pyruvate metabol. Purine metabolism mRNA surveillance Cys & Met metabol. Pyrimidine metabol. Glutathione metabol. Glycerolipid metabol. Ribosome biogenesis N-Glycan biosynthesis Terpenoid biosynthesis Gly, Ser & Thr metabol. Ub-mediated proteolysis Citrate cycle (TCA cycle) Porphyrin & chl. metabol. Oxidative phosphorylation Gly/dicarboxylate metabol. Protein processing in ER Pentose phosphate pathway Non-homologous end-joining Glycolysis / Gluconeogenesis Amino & nucl. sugar metabol. Glycerophospholipid metabol. Fructose & mannose metabol.

(c) average differential expression per KEGG ortholog [log2(fold change)] ≥21.51 0.5 0 -0.5 -1 -1.5 ≤-2 not detected Mougeotia Spirogyra OS9 VCP STT3 SEL1 PLAA SKP1 UFD1 RNF5 OST2 OST3 OST1 SAR1 SSR2 RBX1 SHP1 OTU1 ERN1 CALR PREB MAN1 CANX HUGT WBP1 PDIA6 PDIA1 ATXN3 06-Mar SEC63 SEC31 HSP20 SEC24 SEC13 SEC62 SEC23 HSPA5 STUB1 RAD23 UBE4B ERO1L SYVN1 UBXN6 UBE2D EDEM1 UBQLN EDEM2 GANAB UBE2J1 UBE2J2 HSPA1s DNAJA2 SEC61B HSP90B HSP90A SEC61A DNAJC3 UBE2G1 HSPBP1 TXNDC5 MBTPS2 MBTPS1 NPLOC4 EIF2AK4 PRKCSH E3.5.1.52 DNAJB11 DERL2_3

Figure 2. Responsiveness of metabolic pathway to heat stress in Mougeotia sp. and Spirogyra pratensis. Per species, de novo assembled contigs from RNA-seq data were mapped against the Kyoto Encyclopedia of Genes and Genomes (KEGG). For each obtained KEGG ortholog, all gene expression data (TPMTMM-normalized) were summed up and set relative to control (heat/control). A KEGG ortholog was considered up- and down-regulated if it had a ≥ 2-fold change in gene expression levels. TMM, trimmed mean of M values; TPM, transcripts per million. (a) A heat map visualizing the number of up- (green to blue gradient colored) and down-regulated (blue to purple gradient) KEGG orthologs in each of the 118 detected KEGG plant pathways; each color tick represents one pathway. All data were sorted by the responsiveness (sum of the number of occurrences of ≥2- fold up- and down-regulation) of the KEGG pathway in Mougeotia sp. The stacked bar plot to the right shows the total occurrence of up- and down-regulation, as well as no change, for each of the 1178 and 1000 unique KEGG orthologs in Mougeotia sp. (top) and S. pratensis (bottom). (b) A bar diagram highlighting the 32 KEGG pathways (labeled on the x-axis) that had at least 10 responsive KEGG orthologs in Mougeotia sp. (c) A heat map of the gene expression response of KEGG orthologs of the KEGG pathway ‘Protein processing in endoplasmic reticulum (ER)’. Data are calculated as a log2 (fold changeheat/control) and visualized by a gradient color from green (up-regulation) to blue (no change) to purple (down-regulation); white color means that a KEGG ortholog was not detected (likely below the expression cut-off) in that respective species.

surprising to find the pathway ’Ubiquitin-mediated proteol- S. pratensis, respectively) and ‘Glycerophospholipid meta- ysis’ to be responsive; indeed, ubiquitin ligases have even bolism’ (one and one up-regulated, 12 and 13 down-regu- been found to directly contribute to plants’ resistance to lated) were heat responsive. heat stress (Liu et al., 2014). The effects of heat stress on molecular biology have Next to protein homeostasis, other biomolecule mainte- been studied across the tree of life (Feder and Hofmann, nance processes were highlighted in our KEGG-based anal- 1999), and our observed gene expression patterns speak to ysis. RNA stability is known to be influenced by heat (Su the occurrence of some of the classical responses to heat et al., 2018), which is consistent with finding ‘mRNA in the Zygnematales investigated in the present study. surveillance’ and ‘RNA degradation’ among the top 30 These include, among others, the up-regulation of chaper- most responsive KEGG pathways. Alterations in lipid com- ones, response to ER-stress, and changes in lipid metabo- position are a typical response to heat stress, including in lism. This confirms that our differential RNA-sequencing plants (Narayanan et al., 2016), and we found that the (RNA-seq) analysis approach captures signature responses KEGG pathway ‘Glycerolipid metabolism’ (two and two up- that occur upon heat. Various amino acid metabolism regulated, 11 and 14 down-regulated in Mougeotia sp. and pathways were also found to be highly responsive. KEGG

© 2020 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2020), doi: 10.1111/tpj.14782 Heat stress in land plants’ closest relatives 7 pathways associated with the metabolism of 14 different general stress response pattern among Zygnematophyceae proteinogenic amino acids had more than ten KEGG ortho- – and potentially Zygnematophyceae and land plants. Both logs responsive (up- and down-regulation) in either of the Mougeotia and Spirogyra also showed elevated levels of two species. This suggests a pronounced impact of heat Ser in response to heat (adjusted P < 0.01), as in heat- stress on the general metabolism of both organisms. We stressed C. reinhardtii (Hemme et al., 2014), whereas therefore performed a targeted metabolomic approach to changes in this amino acid was associated with a cold explore this possibility further. stress-specific response in Arabidopsis (cf. Kaplan et al., 2004). The aromatic amino acids Phe and Tyr are key com- Targeted metabolomic profiling of heat-stressed pounds for the production of secondary metabolites, Mougeotia and Spirogyra highlights differential changes including phenylpropanoid-derived ones (Vogt, 2010) – for in amino acid abundance which the genetic basis is present in many streptophyte The shaping of metabolite profiles by heat stress has been algae (de Vries et al., 2017; see also Figure S1). Further- investigated in land plants such as Arabidopsis (Kaplan more, flavonoids, which branch off the same route, have et al., 2004) as well as chlorophytic green algae such as been detected in the zygnematophyceaen alga Penium C. reinhardtii (Hemme et al., 2014). Heat stress has been margaritaceum (Jiao et al., 2019). It is tempting to specu- shown to affect the level of various polar metabolites late that these aromatic amino acids are funneled into across the breadth of green organisms, from Chlamy- (phenylpropanoid-derived) secondary metabolite produc- domonas to Arabidopsis. The KEGG analysis outlined in tion under heat stress, which is known to occur in land the previous section suggests that levels of similar plants, too (Guy et al., 2008), also recognizing that the ele- metabolites were being modified upon heat stress in the vation of Phe was not significant. Such claims await further two species of Zygnematophyceae. We thus used a tar- investigations. geted tandem mass spectrometry approach to identify The non-proteinogenic amino acid c-aminobutyric acid changes in the metabolism of Mougeotia sp. and (GABA) was found to increase in different land plants upon S. pratensis in response to heat stress. The analysis com- a variety of stressors – including temperature stress (Kin- prised 565 parent ion to daughter ion transitions for the nersley and Turano, 2000). In Mougeotia sp. and S. praten- analysis of main central metabolites such as amino acids, sis, GABA was concordantly and significantly (adjusted sugars, lipids, fatty acids, small dicarboxylic acids and P < 0.01) induced upon heat stress (Figure 3). In the meta- nucleobases. Stable isotope internal standards for all 20 bolome of Arabidopsis, GABA is induced upon cold and proteinogenic amino acids were used. Relative metabolite heat (Kaplan et al., 2004). The exact role of GABA as a sig- intensities were corrected for drift by linear regression naling molecule in the stress response of land plants is still based on data from quality control samples that were unclear. GABA can result from the catabolism of glutamate injected after every tenth sample. Using the METABOANALYST R (Hildebrandt et al., 2015); hence, it could simply be the package (Chong et al., 2019), intensity values were log- result of protein degradation and reallocation of resources transformed and analysis of variance (ANOVA) was used to under heat stress. That said, GABA has been proposed to determine significant changes in the metabolome. act as a signaling molecule via the regulation of alu- The heat metabolome of Mougeotia sp. and S. pratensis minum-activated malate transporter membrane channels featured elevated levels of various amino acids (Data S3), (Ramesh et al., 2015). Furthermore, application of exoge- many of which increased significantly (Figure 3). These nous GABA has been shown to increase tolerance; for increases in amino acid levels align well with metabolomic example, towards heat stress in the grass Agrostis stoloni- studies of temperature-stressed Arabidopsis (Kaplan et al., fera (Li et al., 2016). GABA is hence a putative signaling 2004) and heat-stressed C. reinhardtii (Hemme et al., 2014). molecule that might have acted in the molecular physiol- The amino acids Ala, Asn, Pro and Tyr showed concordant ogy of the earliest land plants. and significant enrichment (adjusted P < 0.01) in heat- The metabolome of S. pratensis is noteworthy in light of stressed Mougeotia sp. and S. pratensis; these amino the significant elevation of a few signature metabolites of acids also increase in the metabolome of temperature- heat stress. These include a few additional amino acids stressed (heat, cold) A. thaliana (Kaplan et al., 2004). (Leu, Ile, His and Lys) and sucrose, a common stress Furthermore, Rippin et al. (2019) recently performed a responsive metabolite in plants (Kaplan et al., 2004; Rizh- metabolomic investigation on samples collected from algal sky et al., 2004). Importantly, sucrose can also act as a sig- mats formed by the filamentous Zygnematophyceae Zyg- naling molecule in land plants (e.g. Rolland et al., 2006). nema in the harsh environment of the High Arctic. It was Furthermore, the amino acids Leu and Ile both increased found that the upper layers of these mats, which are more significantly in S. pratensis in response to heat stress, exposed to the elements as well as high irradiances, accu- which aligns with data on Arabidopsis (Kaplan et al., 2004) mulated higher levels of the amino acids Ala and Pro; thus, and C. reinhardtii Hemme et al. (2014), where steady heat- accumulation of these amino acids might be a shared, associated increases of Leu and Ile were found.

Mougeotia sp. Spirogyra change Figure 3. Metabolomic profiling of heat-stressed control heat control heat Mougeotia sp. and Spirogyra pratensis. Metabolites were extracted from both Mougeotia sp. and Spir- 2 tentative Deoxyadenosine ogyra pratensis as control samples maintained at Proline 22°C and as samples heat treated at 37°C in biologi- Sucrose cal triplicates (I–III). Metabolomics was performed Leucine using selected reaction monitoring-based liquid Isoleucine chromatography-tandem mass spectroscopy with Succinic acid internal standards for the amino acids. The analysis Xylose comprised 565 transitions for the analysis of main tentative Decanoylcarnitine central metabolites. Forty-three intensity profiles m/z 385 → m/z 367 01 with different levels are shown in either the species PC(32:0) or treatments (P > 0.01, determined by analysis of Histidine variance). Confidently identified metabolites are Lysine indicated in black, tentatively identified metabolites m/z 133 → m/z 115 are indicated in grey and, for ambiguous cases, the Beta−Alanine transition for the parent ion to daughter ion is m/z 157 → m/z 113 given. The data were scaled and hierarchically clus- Cytosine tered. Each heatmap shows the relative abundance m/z 112 → m/z 95 in a color gradient: red indicates higher relative −2 −1 m/z 166 → m/z 149 abundance and blue indicates lower relative abun- Tyramine dance. Note that the data are relative; comparisons 2−Ketobutyric acid cannot be performed between metabolites, just tentative Betaine aldehyde between samples. Heat maps were generated using Myoinositol METABOANALYST (Chong et al., 2019). The light orange m/z 124 → m/z 80 boxes label those metabolites that show a concor- Tyrosine dant and significant relative elevation upon heat m/z 136 → m/z 119 stress in the two species; light blue labels indicate Alanine concordant depletion. Gamma−Aminobutyric acid Asparagine Serine tentative Glycylproline PC(36:2) PC(36:1) PG(32:0) tentative Acetylcarnitine m/z 401 → m/z 91 tentative N−Acetylglutamine tentative Citramalic acid tentative Hypoxanthine tentative Ubiquinol−10 tentative 4,5−Dihydroorotic acid m/z 73 → m/z 45 Malic acid Fumaric acid

Overall, there were few shared responses between Zygnematophyceae, allowing for inferences with regard to Mougeotia sp. and S. pratensis. Of the 43 signiﬁcantly the heat stress signaling network of the earliest land changed features (adjusted P < 0.01, determined by an plants. To pinpoint these responses, we analyzed those ANOVA), 34 could be attributed to species-speciﬁc differ- protein-coding transcripts that are likely orthologous to ences alone. Altogether, we observed metabolomic one another. For this, we performed a reciprocal best blast changes that were quite distinct between the two strepto- hit (RBBH) approach in which we queried the protein data- phyte algae, many of which are nevertheless in line with sets of the two Zygnematales against one another. In total, what has been observed in heat metabolome studies of we detected 2134 orthologs shared between the two spe- other ‘green’ organisms. Next, we turned again to our cies. Of those 2134 orthologs, Mougeotia sp. up-regulated

RNA-seq data and investigated which transcriptional 305 (log2[fold change] ≥ 1) and down-regulated 567 (log2 responses were shared between the two organisms. [fold change] ≤ À1), whereas S. pratensis up-regulated 197 and down-regulated 789; 1262 and 1148 orthologs showed More than one-half of the genes shared by Mougeotia and no regulation in Mougeotia sp. and S. pratensis, respec- Spirogyra respond differently to heat stress tively (Figure 4a). Heat stress responses that (i) are shared by the two species How do the expression patterns of these 2134 orthologs and (ii) can be linked to relevant biology in land plants are in Mougeotia sp. and S. pratensis compare? To evaluate likely to have played a role in stress response in the last this, we quantiﬁed how frequently both orthologs showed common ancestor shared by land plants and up-regulation (log2[fold change] ≥ 1), non-regulation

(a) 305 1262 Mougeotia sp. S. pratensis differential expression 197 789 567

[log2(fold change)] ≥43-2 1 0 -1 -2 3≤-41148 up down unchanged

(b) concordant discordant up down unchanged one up one down opposite

40 207 676 269 789 153

115 154 471 318 S.p. M. S.p. M. (c) HSP20-like chaperone poly (ADP ribose) polymerase (PARP) FKBP peptidyl-prolyl cis-trans isomerase FKBP peptidyl-prolyl cis-trans isomerase chaperonin 10 heat shock protein 21 caseionolytic proteinase alanine-tRNA ligase, putative (DUF760) Aha1 domain-containing protein auxin transport protein (BIG) heat shock protein Fes1A Cox19-like CHCH GNS1/SUR4 fatty acid biosynthesis PQ-loop repeat / transmembrane Cellulose-synthase-like C5 Malectin/receptor-like protein kinase MUTS-homologue 5 heat shock organizing protein (HOP) Hyp O-arabinosyltransferase-like heat shock protein 60-2 DegP protease 1 2-oxoglutarate dehydrogenase zinc induced facilitator-like 1 Protein-tyrosine phosphatase-like, PTPLA

heat shock protein 81.4 Mougeotia sp. restriction endonuclease, type II-like Spirogyra pratensis NADH dehydrogenase ubiquinone 1 β-subunit structural maintenance of chromosome 3 SOUL heme-binding cinnamyl alcohol dehydrogenase 9 glutathione S-transferase PHI 9 general transcription factor II H2 membrane bound O-acyl transferase chloroplast RNA TPR outer membrane trp-rich sensory protein-like NAD(P)-binding Rossmann-fold superfamily chloroplast stromal processing peptidase Ribosomal protein S11 0246

differential expression [log2(fold change)]

Figure 4. Differential gene expression patterns of putative orthologs shared between Mougeotia sp. and Spirogyra pratensis. Using a reciprocal best BLAST hit (RBBH) approach, we determined orthologs shared between Mougeotia sp. and Spirogyra pratensis and analyzed their differential gene expression changes

(log2[fold changeheat/control], calculated using EDGER). (a) A heat map depicting the differential gene expression changes of all detected orthologs shared between Mougeotia sp. and Spirogyra pratensis, sorted by their average expression change. Each colored tick represents the log2(fold changeheat/control), gradient colored from green (up-regulation) to blue (unchanged) to purple (down-regulation). The stacked bar plot to the right shows the total occurrence of up-regulation (log2[fc] ≥ 1), down-regulation (log2[fc] ≤ À1) and unchanged gene expression (|log2(fc)| ≯ 1) among the 2134 shared orthologs for each of the two species. (b) A quantiﬁcation of the concordant and discordant differential gene expression patterns among the 2134 shared orthologs. We deﬁned concordant trends as both orthologs being up-regulated (both had a log2[fc] ≥ 1), down-regulated (both had a log2[fc] ≤ À1) and unchanged (both had a |log2(fc)| ≯ 1). Discordant trends means that one ortholog was up-regulated in species A at the same time as unchanged in species B (‘one up’), down-regulated in species A at the same time as unchanged in species B (‘one down’), and up-regulated in species A at the same time as down-regulated in species B (‘opposite’). Fanned out below the 269 ‘one up’ and 789 ‘one down’ is an illustration of how often this was caused by orthologs being up-/down-regulated in Spirogyra pratensis (S.p.) and Mougeotia sp. (M), and unchanged in the other respective species.

(c) The precise differential gene expression (as log2[fold changeheat/control]) of the 40 concordantly up-regulated, shared orthologs; the data in the diagram are sorted by the between species average log2(fc) values for each given ortholog.

(|log2[fold change]| ≯ 1), and down-regulation (log2[fold two species that had PARPs that match the domain compo- change] ≤ À1). 923 orthologs showed the same (concor- sition of Moug13104_c0_g1: K. nitens (KFL_000210040; dant) gene expression trends and 1211 showed discordant genome published by Hori et al., 2014) and Chara braunii trends (Figure 4b). Discordance was most frequently a (CBR_g19059 and CBR_g16983; genome published by result of orthologs showing non-regulation in one species Nishiyama et al., 2018). Our previous RNA-seq analyses and down-regulation in the other; we observed this in 789 (de Vries et al., 2018) uncovered a transcript coding for a cases, with 318 being caused by down-regulation in similar protein in Coloechaete scutata (Cscu- Mougeotia sp. and 471 in S. pratensis. We counted 269 tata_DN35021_c1_g1). These ANK::WWE::PARP might rep- cases of discordance as a result of orthologs with non-reg- resent streptophyte algae-specific DNA repair proteins. ulation in one species and up-regulation in the other; such The third and fourth highest (average) concordant up- cases roughly occurred in an equal distribution across the regulated orthologs were two FK506-binding proteins two species (154 for Mougeotia sp. and 115 for S. praten- (FKBP). FKBPs are known for their involvement in stress sis). Only 153 showed the opposite regulation (i.e. up in tolerance, including tolerance to temperature stress as a one species and down in the other). The concordant differ- result of the assurance of correct HSP accumulation (Meiri ential gene expression trends of orthologs was dominated and Breiman, 2009). Chloroplastidial FKBPs are important by non-regulation: 676 cases of ‘unchanged’ out of all 923 for the correct assembly of the photosystem II components concordantly regulated orthologs (Figure 4b). There were at the thylakoid membrane (Lima et al., 2006) and overex- 207 orthologs that exhibited concordant down-regulation pression of FKBP16 in Arabidopsis resulted in an increased (Figure S2) and only 40 displayed concordant up-regula- tolerance to abiotic stressors such as drought and high tion. light because of elevated photosystem stability (Seok The 40 concordantly up-regulated orthologs likely et al., 2014). The targeting prediction tool GTP (Fuss et al., include key candidates for the zygnematophyceaen 2013) suggests that three of them might be targeted to the response to heat. To explore their molecular role, we plastid. Hence, our data on the concordant up-regulation of assigned functional annotations to the orthologs by manu- FKBPs highlight the importance of photosystem stability in ally comparing their respective best hits to Arabidopsis. As Zygnematophyceae under heat stress. expected, the concordantly up-regulated orthologs Mougeotia sp. and S. pratensis concordantly up-regu- included HSPs (e.g. a HSP20-like gene that was 100.1- and lated a gene that likely codes for a protein with homology 54.8-fold up-regulated in Mougeotia sp. and S. pratensis, to an Arabidopsis protein justly called BIG. The Arabidop- respectively); in total, seven of the 40 concordantly up-reg- sis BIG protein is 5098 amino acids in length and links ulated orthologs could be associated with HSP function. auxin transport and light signaling (Gil et al., 2001). High This further included an ortholog of the heat-inducible concentrations of auxin were shown to elicit changes in HSP70-HSP90 ORGANIZING PROTEINS (HOP). Proteins of the growth of the streptophyte alga K. nitens (Ohtaka the HOP family are involved in various aspects of plant et al., 2017). Furthermore, how auxin links (via intracellular heat tolerance, including protein quality control (Fernan- auxin transport) temperature response and growth was dez-Bautista et al., 2018). recently elaborated for Arabidopsis (Feraru et al., 2019). The ortholog with the second highest (average) concor- Yet, although streptophyte algae do harbor many of the dant up-regulation was a poly (ADP ribose) polymerase components of auxin signaling, their action in a bona fide (PARP) (8.7-fold in Mougeotia sp. and 39.6-fold in auxin response is still unclear (Nishiyama et al., 2018; S. pratensis). These PARP orthologs recovered as RBBHs Mutte et al., 2018; Bowman et al., 2019). Instead of (or in the RADICAL-INDUCED CELL DEATH1 (RCD1) proteins, addition to) a role in auxin-mediate processes, it is possi- which are land plant specific proteins that contain an RST ble that the BIG protein(s) of Mougeotia sp. and/or (RCD-SRO-TAF4) domain and sometimes a WWE domain S. pratensis are woven into Ca2+/calmodulin-mediated reg- (Jaspers et al. 2010). Plants whose PARPs are knocked out ulation, as is the case for the animal homolog calossin (Xu exhibit altered stress tolerance; both increased as well as et al., 1998). decreased, depending on the stressors (Ahlfors et al., 2004; Cinnamyl-alcohol dehydrogenase (CAD) was among the Fujibe et al., 2004). The presence of RCD homologs in concomitantly up-regulated genes. We recently showed streptophyte algae might speak to an earlier origin of these that many streptophyte algae have an almost complete plant-specific SOR proteins than previously thought. Yet, gene set corresponding to the land plant-like phenyl- the domain composition of the RCD found in the two spe- propanoid biosynthesis pathway (de Vries et al., 2017). cies could not easily be compared because the S. pratensis CAD is a key enzyme in the biosynthesis of lignin and lig- sequence was fragmented. The Mougeotia sp. ortholog nin-like compounds. To analyze this further, we conducted (Moug13104_c0_g1) had an additional ankyrin repeat a phylogenetic analysis of putative CAD sequences from (ANK), having a domain composition in the order ANK:: major lineages of land plants and streptophyte algae. Most WWE::PARP. Indeed, pHMMER searches recovered only streptophyte algae-derived sequences cluster with other

© 2020 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2020), doi: 10.1111/tpj.14782 Heat stress in land plants’ closest relatives 11 dehydrogenases, such as zinc-binding or glutathione-de- aquaporins in the membrane is known to be dynamically pendent formaldehyde dehydrogenases (annotation based adjusted under stress conditions (Maurel et al., 2015). In on BLASTP against TAIR10) (Berardini et al., 2015) (Fig- addition, an iron transporter ortholog to IRON REGULATED ure S1). The monophyletic CAD-like and CAD cluster 1 was 41.6-fold up-regulated (adjusted P = 7.6 9 10À26). included twelve streptophyte algal sequences. We recov- Hence, Mougeotia sp. appeared to undergo pronounced ered CAD groups I and V with high support (Figure S1). adjustment of its small molecule transporter profile. CAD I likely emerged in the ancestor of tracheophytes The fifth most up-regulated putative Arabidopsis ortho- because it includes a CAD I sequence from S. moellendorf- log in S. pratensis (37.4-fold, adjusted P = 2.0 9 10À15) fii, yet none from bryophytes or streptophyte algae. CAD V coded for a RING/U-box-containing protein (with likely E3 includes sequences from bryophytes (M. polymorpha and ubiquitin ligase activity) that returned as RBBH the protein Physcomitrella patens) and tracheophytes, indicating that ABA INSENSITIVE RING PROTEIN 3 (AIRP3). Although it may have been present in the ancestor of land plants. AIRP3 has obviously been tied to ABA and stress response We were unable to recover the CAD group ‘II&III’. It is split in land plants (Kim and Kim, 2013), its role (and targets) in into several sub-clades and interspersed by CAD groups I S. pratensis needs to be clarified. It is noteworthy that and V (Figure S1). Most of the 12 streptophyte algal CAD- another transcript that coded for a RING/U-box-containing like sequences cluster with sequences that fell in CAD protein was also in S. pratensis’ top 100 (up-regulated group II&III in previous analyses (de Vries et al., 2017). The 12.1-fold, adjusted P = 3.8 9 10À7). Furthermore, a putative current position suggests that the sequences are CAD-like ortholog of KOBITO1 [KOB1; aka ABA INSENSITIVE 8 sequences that do not belong to CAD group II&III, but (ABI8)] was 6.7-fold up-regulated (adjusted rather form their own clades, within the cluster of canoni- P = 8.6 9 10À13)inS. pratensis. KOB1 effects plant growth cal CAD sequences. Based on our metabolome data, it is and is a mediator of ABA signaling, possibly integrating tempting to speculate that the changes in the pool of aro- stress signals (Pagant et al., 2002; Brocard-Gifford et al., matic amino acids might indicate the activity of a pathway 2004). The putative KOB1 ortholog in Spirogyra is hence a that features CAD. promising candidate that might link stress input to the cell and the plastid morphological changes that we have Conserved and functionally diverse orthologs highlight observed. signature responses to heat In addition to KOB1, we identified other genes that We next addressed the question of which puta- might be associated with the phenotypic alterations tive orthologs that are shared by one of the two Zygne- observed in heat-stressed S. pratensis (Figure 1). The sec- matales and land plants showed the highest gene ond-highest induced transcript (59.8-fold elevated by heat, expression changes. For this, we again made use of an adjusted P = 3.1 9 10À23) coded for a protein likely RBBH approach, this time against the best-annotated land involved in microtubule dynamics (containing a predicted plant genome, that of A. thaliana. We focused on the top LisH motif); it returned as RBBH TONNEAU 1B (TON1B), 100 strongest up-regulated orthologs (Figure 5). The data which is critical for microtubule organization and hence featured many proteins that have been pinpointed by the cell division and morphogenesis (Azimzadeh et al., 2008). independent analyses that we described above, including Reorganization of the cytoskeleton might underpin the orthologs likely involved in information processing, ortho- elaborate changes in intracellular organization of Spir- logs associated with plastid biology, and chaperones/ ogyra’s cells under heat stress. Using a unidirectional BLAST HSPs. Indeed, in both species, the highest induced ortho- against the well-annotated genome of Arabidopsis thali- log was an HSP, being significantly up-regulated by 685.6- ana, we thus screened for other up-regulated homologs of fold in Mougeotia sp. and 100.1-fold in S. pratensis (Ben- proteins known to be relevant for (re-)organization of the jamini–Hochberg-adjusted P < 0.001). In terms of organel- cytoskeleton. In both species, we detected a few up-regu- lar biology, Spirogyra, for example, up-regulated two lated candidates for motor and additional microtubule-as- genes coding for Rossman-fold dehydrogenases that were sociated proteins (Figure S3); these homologs are predicted to be targeted to the chloroplast [significantly interesting candidates for underpinning the changes in (adjusted P < 0.001) 24.1- and 13.7-fold up]; one additional plastid morphology observed in both species. protein that was significantly induced (5.6-fold; adjusted Oleosin was the 19th highest up-regulated putative P < 0.001) is likely mitochondria-localized. ortholog of Arabidopsis proteins in Spirogyra. Oleosins are The third-highest up-regulated putative Arabidopsis oil body-associated proteins (Frandsen et al., 2001). From ortholog (89.4-fold, adjusted P = 5.2 9 10À40)inMougeotia other Zygnematophyceae, such as many species of Zyg- sp. coded for an aquaporin (returning NOD26-LIKE INTRIN- nema, we know that lipid body accumulation is induced SIC PROTEIN as RBBH). Aquaporins act as channels that upon stressors, including during lipid accumulation that transport water as well as small molecules, including reac- happens upon the formation of the more stress-resilient tive oxygen species (Maurel et al., 2015); the repertoire of pre-akinetes (Pichrtova et al., 2016). This aligns with our

Mougeotia sp. Figure 5. The top 100 heat-induced putative Ara- bidopsis orthologs in Mougeotia sp. and Spirogyra pratensis. Via a reciprocal best BLAST hit (RBBH) approach, we determined (pairwise) putative orthologs between Mougeotia sp. and Arabidopsis thaliana, as well as Spirogyra pratensis and A. thaliana. Based on the differential gene expression changes

(log2[fold changeheat/control], calculated using EDGER), we extracted the top 100 strongest up-regulated putative Arabidopsis orthologs. We plotted them as word clouds. The words represent the names and/ or description of the Arabidopsis (RBBH) orthologs; the size of the words represents the differential gene expression change; note the scale in the bottom right corner of each species’ word cloud.

log2(fc): pratensis

log2(fc(fc):): morphological observations in which S. pratensis cells Considering that Zygnematophyceae stand out by harboring accumulated granular contents upon heat exposure (Fig- the red- and blue-light receptor fusion proteins called neo- ure 1). chromes (Li et al., 2015), our data further highlight the pho- A few TFs were among the top 100 strongest up-regu- toreceptors in streptophyte algae. lated genes. This is noteworthy because TFs are situated Quite a few carotenoid-derived metabolites have upstream in genetic regulatory hierarchies and hence influ- recently been brought into focus by linking the plastid and ence the expression of various downstream genes. In cell physiology of plants (Ramel et al., 2012; Xiao et al., Mougeotia sp., a putative ortholog of KNOTTED1-LIKE 2012; Avendano-V~ azquez et al., 2014; Kleine and Leister, HOMEOBOX GENE 3 (KNAT3) TF was 8.0-fold up-regu- 2016; Hou et al., 2016). Mougeotia sp. up-regulated a tran- lated. In Arabidopsis, KNAT3 modulates response to envi- script 17.5-fold that returned CAROTENOID CLEAVAGE ronmental factors and ABA signaling (Kim et al., 2013). In DIOXYGENASE 1 (CCD1) as its RBBH. CCDs, and especially Spirogyra, a PLATZ TF was 20.0-fold up-regulated; this CCD1 (Vogel et al., 2008), produce a variety of apoc- PLATZ TF returned as RBBH AT1G76590, for which a arotenoids. Apocarotenoids have been proposed to play a homolog was found to be responsive to drought in the role in stress response of land plants (Hou et al., 2016). For halophyte Thellungiella (Wong et al., 2006). These TFs are example, CCD4 of saffron crocus is induced by heat stress thus key candidates for future work on how the molecular (Rubio et al., 2008) and potato RNAi lines of CCD4 display response to stress is regulated in Zygnematophyceae. a heat-stressed phenotype (Campbell et al., 2010). The phy- The blue-light signaling component cryptochrome (re- tohormones ABA and strigolactones are both carotenoid- trieving as RBBH CRY3) was 11.1-fold up-regulated in derived (Nambara and Marion-Poll, 2005; Al-Babili and Mougeotia sp.; we had already identified cryptochrome as Bouwmeester, 2015). Our observation of the regulation of being very stress responsive in Coleochaete scutata (de Vries zygnematophycean enzymes likely involved in the produc- et al., 2018). Indeed, the blue-light photoreceptor CRY1 was tion of apocarotenoids further indicates that such proteins shown to intersect with red light signaling under heat stress play a significant role in the physiology of the closest algal and Mougeotia sp. was found to 3.3-fold down-regulate (ad- relatives of land plants. justed P < 0.001) a putative ortholog of PHYTOCHROME Among the top 100 most highly up-regulated proteins INTERACTING FACTOR 3-LIKE 5 (Mousp17599_c0_g1). summarized in the present study are key factors of the

© 2020 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2020), doi: 10.1111/tpj.14782 Heat stress in land plants’ closest relatives 13 molecular stress physiology that these specific Zygnemato- regulated upon cold stress. Because Rippin et al. (2017) phyceae employ during heat stress and acclimation. We found strong up- and down-regulation of CDPKs in a differ- next aimed to integrate this information into a comparative ent strain of Z. circumcarinatum (SAG 2419), CDPKs are plant physiological perspective. hence potential general regulators of the response to stressors in Zygnematophyceae. Although we are unaware of The closest algal relatives of land plants respond to heat any functional studies in Zygnematophyceae, McCurdy via calcium signaling, histidine kinases and chloroplast– and Harmon (1992) demonstrated the Ca2+-dependent pro- nucleus communication tein-kinase activity of a CDPK homolog in the Charo- Our recent work in Z. circumcarinatum (SAG 698-1a) has phyceae Chara corralina. It is hence conceivable that the highlighted the existence of homologous signaling circuits closest algal relatives of land plants use CPDKs when deal- that in land plants unite environmental inputs, ABA signaling with a variety of stressors. We can thus further infer ing and plastid physiology (de Vries et al., 2018). Addition- that the earliest land plants likely used CDPK-mediated sig- ally, in a recent transcriptomic study, Rippin et al. (2017) naling when they faced the barrage of terrestrial stressors. found that CPKs/CDPKs were among the most responsive Future research will need to clarify the details of Ca2+-sig- signaling components in desiccation stressed Z. circumcar- natures under abiotic stress in the Zygnematophyceae and inatum (SAG 2419). Indeed, in land plants, there is a large where in the cell these changes in Ca2+-signatures occur. overlap of ABA-mediated and Ca2+-mediated signaling that The thylakoid, the site of photosynthetic light reactions, occurs under stress (Edel and Kudla, 2016). The plastid is is at the frontline of the cellular abiotic stress response among the most heat stress-responsive sub-cellular com- (Foyer et al., 1994). Therefore, components directly woven partment in plant cells (Schroda et al., 2015; Sun and Guo, into the photosynthetic light reactions are known to be reg- 2016). In particular, the components of the light reaction at ulated under stress. With a stress exposure time of 24 h, the thylakoid membrane have long been considered a pri- the two Zygnematophyceae examined in the present study mary target (Berry and Bjorkman,€ 1980). Furthermore, plas- will likely have started long-term acclimation processes tid signaling under heat is assumed to involve calcium that involve reorganization of the antenna proteins and, signaling (Sun and Guo, 2016; Lenzoni and Knight, 2018), hence, transcriptional changes (Eberhard et al., 2008). which we highlighted above as being heat responsive in Indeed, the light-harvesting chlorophyll-a/b proteins of our zygenmatalean algae. photosystem II (LHCB) showed especially pronounced tran- Ca2+-mediated signaling is a key aspect of the stress scriptional changes. Mougeotia sp. regulated four of the physiology of land plants. CDPKs unite Ca2+ sensing and eight detected LHCB2.1 homologs significantly (corrected response in a single protein (Kudla et al., 2018). Further- P < 0.001) up (98.4-, 7.4-, 3.6- and 3.5-fold); no LHCB2 more, CDPKs have been shown to be key for orchestrating homolog experienced significant down-regulation. In the signal transduction upon heat-shock in land plants S. pratensis, none of the three LHCB2.1 homologs showed (Davletova et al., 2001; Liu et al., 2008; Dubrovina et al., differential regulation; one of the three LHCB2.3 homologs 2017). Based on a query against the A. thaliana protein was significantly (corrected P < 0.05) 2.1-fold up-regulated database, we detected 32 putative CDPK homologs in (none of Mougeotia’s three LHCB2.3 homologs exhibited Mougeotia sp. and 21 in S. pratensis. A few of these CDPKs significant regulation). showed up-regulation [six in Mougeotia sp. (three signifi- Heat stress induced the expression of genes for proteins cant, Benjamini–Hochberg corrected P < 0.05) and three in that are involved in photoprotection. These photoprotec- S. pratensis (two with P < 0.05)], whereas many were tion components primarily shield proteins integrated in the down-regulated (14 in Mougeotia sp. (all with P < 0.05) light reaction chain and are hence generally stress-associ- and 15 in S. pratensis (10 with P < 0.05)] (Figure 6). Land ated. This included the significant (corrected P < 0.001) 4.7-, plant CDPKs have a well-defined domain organization. 4.1- and 3.6-fold up-regulation of three homologs of They consist of an N-terminal Ser/Thr kinase domain and a PHOTOSYSTEM II SUBUNIT S (PSBS)inMougeotia sp.; series of C-terminal calcium-binding EF-hands (typically PSBS is a major player in non-photochemical quenching four) (Cheng et al., 2002). Using INTERPROSCAN (Quevillon (NPQ) including of Zygnematophyceae (Gerotto and Moro- et al., 2005), we screened the putative zygnematalean sinotto, 2013). Furthermore, the NPQ-relevant violaxanthin CDPK sequences for this architecture; seven Mougeotia sp. de-epoxidase (NPQ1) participating in the xantophyll cycle and eight S. pratensis CDPK homologs showed this (Jahns and Holzwarth, 2012) was significantly (corrected arrangement. Four and seven were down-regulated (four P < 0.001) up-regulated in Mougeotia sp. by 4.2-fold. and five with P > 0.05); one homolog in each species EARLY LIGHT-INDUCABLE PROTEIN (ELIP) is a photopro- showed significant up-regulation. Our previous transcrip- tective chlorophyll a/b-binding protein that accumulates tome of Z. circumcarinatum (strain SAG698-1a) (de Vries under stress – especially cold and high light (Montane et al., 2018) identified four CDPK sequences with a canoni- et al., 1997; Hutin et al., 2003). Indeed, in previous work, cal S/T-kinase:EF-hand structure, one of which was down- we observed that cold treatment strongly induced the

© 2020 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2020), doi: 10.1111/tpj.14782 14 Jan de Vries et al. expression of ELIP-encoding genes in various streptophyte significantly (adjusted P < 0.05) up-regulated by 2.3-fold algae, including Z. circumcarinatum (de Vries et al., 2018); Mougeotia sp. and significantly (adjust P < 0.05) down-reg- in a differential transcriptomic analysis, Rippin et al. (2019) ulated in S. pratensis by 2.8-fold; in each species, the found that the exposed upper layer of mats of Zygnema homologs were likely plastid/mitochondrion dual localized, sp. growing in the High Arctic exhibit elevated ELIP gene with ATP2 scores of 0.664 and 0.653. The poly(A)-specific expression levels. Of the three ELIP homologs detected in ribonuclease ABA-HYPERSENSITIVE GERMINATION 2 both Mougeotia sp. and S. pratensis, only one Mougeotia affects mRNA stability under stress (Nishimura et al., ELIP was significantly up-regulated (corrected P < 0.001; 2005). A likely dual-localized homolog of Mougeotia sp. 8.4-fold). The nuclear-encoded photosystem II component AHG2 was significantly 5.3-fold up-regulated upon heat PSBW was significantly up-regulated (corrected P < 0.05) (ATP2 score of 0.533; adjusted P < 0.001); in cold-treated in both Mougeotia sp. (2.3-fold) and S. pratensis (6.9-fold); Z. circumcarinatum (de Vries et al., 2018), the putative this is noteworthy because PSBW has been linked to pho- AHG2 ortholog Zcircumcarinatum_DN45059_c2_g1 was tosystem II turnover (Hagman et al., 1997) and stability significantly (adjusted P < 0.05) 3.0-fold up-regulated and (Garcıa-Cerdan et al., 2011). Altogether, these data reveal also likely plastid/mitochondrion dual-localized (ATP2 that heat stress induces processes tied to photoprotection score of 0.577). Altogether, these data highlight the inte- and thylakoid reorganization in our two Zygnemato- gration of organellar gene expression and response in the phyceae. zygnematophyceaen reaction to heat. Next to these classical components of the light reaction RELA-SPOT HOMOLOG (RSH) proteins synthesize gua- in the thylakoids, some peripheral light reaction-associated nosine penta- and tetraphosphate (ppGpp), which is a sig- genes showed interesting differential gene expression pat- naling nucleotide that has been shown to impact various terns. LIGHT-HARVESTING-LIKE 3 (LIL3) proteins are local- aspects of chloroplast function (Sugliani et al., 2016) and, ized in the thylakoids (Takahashi et al., 2014) and are importantly, abiotic stress (Takahashi et al., 2004). In associated with chlorophyll biosynthesis (Tanaka et al., Mougeotia sp., a homolog of RSH3 was non-significantly 2010). Although a LIL3 homolog was present in Mougeotia induced by 2.0-fold (adjusted P = 0.07; likely plastid/mito- sp. and S. pratensis each, only S. pratensis induced LIL3 chondrion dual localized with an ATP2 score of 0.649). In (4.2-fold), albeit not significantly (corrected P = 0.077). It is S. pratensis, an RSH3 homolog was significantly 7.9-fold nonetheless tempting to speculate that this is tied to heat- down-regulated (adjusted P < 0.01; likely plastid/mitochon- induced phenotypic alterations (Figure 1). FERRITIN (FER) drion dual localized with an ATP2 score of 0.677); one homologs, potentially targeted to the chloroplast (GTP homolog of RSH1 found in Spirogyra was non-significantly score of 0.664 and 0.503 for Mougeotia sp. and S. praten- reduced by 4.0-fold (adjusted P = 0.07; partial sequence). sis, respectively) were significantly (adjusted P < 0.05) RSH-based synthesis of ppGpp was also detected in the down-regulated by 3.9-fold in Mougeotia sp. and up-regu- chlorophyte C. reinhardtii (Kasai et al., 2002). This suggests lated by 13.5-fold in S. pratensis. Long and Merchant that ppGpp is a signaling molecule used in stress response (2008) reported that the model green alga C. reinhardtii across the green tree of life, including in streptophyte down-regulates FER1 upon high light stress; yet, based on algae. data from rice (Murchie et al., 2005), Arabidopsis (Rossel Streptophyte algae harbor an elaborate set of genes et al., 2002) and barley (Atiena et al., 2004), it was also homologous to those involved in land plant retrograde sig- noted that some land plants up-regulate FER, whereas naling (de Vries et al., 2018; Nishiyama et al., 2018). Many others down-regulate FER under high light stress. The of these proved transcriptionally responsive under stress responsiveness of FER in both directions that we observe (de Vries et al., 2018). Given the importance of retrograde in our two Zygnematophyceae hence adds to this intricate signaling in response to physiological challenges such as picture. abiotic stress (Nott et al., 2006; Chan et al., 2016), we Various pentatricopeptide repeat (PPR) proteins act in screened our transcriptome for 17 components known chloroplast gene expression and RNA metabolism (Sch- from Arabidopsis to be involved in retrograde signaling; mitz-Linneweber and Small, 2008). A likely plastid/mito- we found at least one homolog for ten (13 genes in total) chondrion dual-localized (ATP2 score of 0.566) Mougeotia and six (eight in total) of these proteins in Mougeotia sp. sp. homolog encoding the PPR protein WHAT’S THIS FAC- and S. pratensis, respectively. Upon heat stress, Mougeo- TOR 1 (WTF1) was significantly up-regulated by 7.1-fold tia sp. significantly up-regulated four and down-regulated upon heat (adjusted P < 0.001); S. pratensis down-regu- three of the 13 homologs detected (adjusted P < 0.05); lated its WTF1 homolog non-significantly (adjusted S. pratensis significantly down-regulated two (adjusted P = 0.15) by 4.4-fold. Also, the NARA5 proteins (GENES P < 0.05). Importantly, we detected a GUN1 homolog in NECESSARY FOR THE ACHIEVEMENT OF RUBISCO ACCU- Mougeotia sp., which corroborates our previous inference MULATION 5) play a role in plastid gene expression that GUN1 is part of the plastid–nucleus communication (Ogawa et al., 2009). A homolog of NARA5 was chassis in ZCC streptophyte algae (de Vries et al., 2018;

Nishiyama et al., 2018). Altogether, plastid–nucleus com- In a previous study (de Vries et al., 2018), we noted the munication likely plays a role in zygnematophyceaen responsiveness of AHK to cold stress and thus speculated responses to a variety of stressors. that interconnections between the AHKs, cytokinin and Both Mougeotia sp. and S. pratensis encode histidine stress (and possibly ABA) were already present in the clos- kinase AHK homologs in their transcriptomes. AHKs are est algal relatives of land plants. For example, Zcircumcari- well known for their involvement in the canonical cytokinin natum_DN47719_c0_g1 shows putative orthology to AHK4, signaling pathway of land plants (Kieber and Schaller, has a chase domain, and is 1.9-fold up-regulated upon cold 2018). Additionally, certain AHKs are recognized as being stress. The present data on heat stress further corroborate responsive to a variety of abiotic stressors (Tran et al., this notion. Histidine kinases, potentially even CK-sensing 2007; Jeon et al., 2010; Pham et al., 2012). ones, are hence key candidates for future research into the AHKs that are involved in cytokinin signaling contain a early evolution of the molecular physiology of plants. That CHASE domain (Cyclases/Histidine kinases Associated said, the cytokinin-mediated induction of certain ARR Sensory Extracellular) (Heyl et al., 2007), known to bind homologs (a key feature of cytokinin signaling in seed cytokinins in seed plants and thereby initialize the down- plants) (Brenner and Schmulling,€ 2015) is either missing or stream phosphorelay chain that leads to the phosphoryla- very low in S. pratensis (maximum 1.5-fold) (Figure S5). tion of type B response regulators. For the non-CHASE Additional physiological and reverse genetic studies are domain-containing AHKs, such as AHK1, knowledge on necessary to complete the picture. the ligand remains obscure. Based on a PFAM search, Our previous analyses on the zygnematophyceaen alga we identified CHASE domains in two sequences Z. circumcarninatum highlighted the presence and tran- from Mougeotia sp. (Mousp15483_c1_g3 and Mous- scriptional responsiveness of homologs of the ABA signal- p17224_c0_g1) and in four sequences from S. pratensis ing components of land plants (de Vries et al., 2018); the (Spipra3165_c2_g1, Spipra3977_c2_g2, Spipra3130_c0_g4 detection of a putative PYR/PYL/RCAR ortholog was partic- and Spipra3862_c3_g8). We performed phylogenetic anal- ularly unexpected. Here, we detected a PP2C (best hit ysis of all detected AHK homologs as well as Arabidopsis against Arabidopsis was AtABI2) and one SnRK2 (best hit and P. patens AHK sequences; we recovered an AHK1 against Arabidopsis was AtSnRK2.6/AtOST1) in the tran- clade, with one ortholog in Mougeotia and two to three scriptomes of Mougeotia sp. and S. pratensis, respectively. orthologs in Spirogyra (Figure S4). All AHK homologs Spirogyra pratensis significantly (corrected P < 0.01) with the best BLAST hit against AHK2, 3 and 4 did not down-regulated the homologs of PP2C 5.5-fold and SnRK2 cluster with the expected groups in land plants, which can 2.2-fold. We detected no PYR/PYL/RCAR homolog in either be interpreted as lineage-specific expansions in these alga. Given that PYR/PYL/RCAR homologs are present in streptophyte algae. other Zygnematophyceae (Sun et al., 2019), including the As in Arabidopsis (Brenner and Schmulling,€ 2015), the genome of Mesotaenium endlicherianum (Cheng et al., expression of AHK (homologs) did not significantly change 2019), there are several possible explanations for this, the under the influence of exogenous cytokinin in S. pratensis most likely being that it was simply below the expression (Figure S5). However, a strong differential gene expression cutoff. Yet, (i) not all Zygnematophycae have PYR/PYL/ response occurred under heat stress in our transcriptome RCAR encoded in their genome (as in the case of Spiro- data. In Mougeotia sp., five AHK homologs were signifi- gloea muscicola) (Cheng et al., 2019) and (ii) the PYR/PYL/ cantly (Benjamini–Hochberg corrected P < 0.05) up-regu- RCAR proteins of Zygnematophyceae likely do not function lated (4.2-, 3.0-, 2.8-, 2.7- and 2.7-fold) upon heat stress, as bona fide ABA receptors (Sun et al., 2019). Thus, strep- two of them being CHASE domain-containing ones. In tophyte algal homologs of genes that in land plants are S. pratensis, one CHASE domain-containing AHK homolog known to form the foundation of stress-regulatory net- was significantly (adjusted P < 0.001) 3.2-fold up-regulated works might be regulated through different components, and two were significantly (adjusted P < 0.05) 5.8-fold and and even used for divergent means (see also discussions 22.4-fold down-regulated. Of the down-regulated AHK in Furst-Jansen€ et al., 2020). Nonetheless, the mere pres- homologs, one contained a CHASE domain (Figure 6). ence of such homologs sheds light on the genetic material Both Mougeotia CHASE domain-containing AHK homologs from which the stress-regulatory cascades of land plants (best BLAST hit: AHK4) were significantly up-regulated were molded. (corrected P < 0.05) upon heat (2.8- and 2.7-fold). Also, Pinpointing conserved patterns in two divergent both Spirogyra CHASE domain-containing AHK homologs Zygnematophyceae (best BLAST hit: AHK4) showed induction, although this was not statistically significant. Hence, it is conceivable In the present study, we have investigated two divergent that, in both species, AHK homologs cause a modulation species, both of which thrive in freshwater systems such of phosphorelay signal transduction contributing to the as streams or springs. It is prudent to note that a rapid regulation of stress response. shift from 22°Cto37°C likely does not occur in the

Plastid Mougeotia Spirogyra WTF1 FER RSH NARA5 NPQ1 3 AHG2 - LHCA

LHCB2.1 -2 LHCB2.3 LIL3 -1 ELIP PSBS PSBW/28 PSAK 1 CytosolCtytoso l E E EE CPK / E E EE CRK EEE EE 2≤ E E PP2C CPKPK OST1 ≥3 0 PPPP2C DRS1 AHKs ABI8 SnRKSn CRKRK AHK1 C C C C C AHK2,3,4

G (fold change)] 2 RACK,RGSS PHYB GUN1 GUN2 GUN3 GUN4 Nucleus GUN5 GLK1 EX HY5

STN7 differential expression [log SAL1 PRL1

Figure 6. A cellular signaling framework for the molecular response to heat in Mougeotia sp. and Spirogyra pratensis. Based on a BLASTP approach, we queried the predicted proteins of Mougeotia sp. and S. pratensis against the well-annotated proteome of Arabidopsis thaliana; we screened for homologs of components involved in (i) the light reaction of photosynthesis, (ii) calcium-signaling, (iii) ABA signaling, (iv) AHK- and g-protein-mediated signaling, and (v) retrograde signaling. Components are shown when one of the detected homologs exhibited differential gene expression changes. The cartoon on the left indicates in which cell biological/ physiological context the detected homologs act. The gradient-colored circles on the right visualize the associated gene expression data

(log2[fold changeheat/control], calculated using EDGER) from green (up-regulation) to blue (unchanged) to purple (down-regulation); an orange ring around the circles indicates that a signiﬁcant change in gene expression occurred (Benjamini–Hochberg corrected P < 0.05). A ‘C’ within circles indicates the likely presence of a CHASE domain in the algal AHK homologs; an ‘E’ indicates the likely presence of an EF-hand domain in the algal CPK/CRK homologs. The exact expression values and transcript IDs are provided in Data S5. freshwater environment in which Mougeotia and Spirogyra conserved between streptophyte algae and land plants naturally occur. We have used a very drastic shift in tem- (Lind et al., 2015; Sun et al., 2019). This demonstrates that perature to ensure that we trigger a pronounced gene the PP2Cs are exciting candidates for acting as master expression response; information that is not aimed at illu- switches in stress signaling since before the dawn of land minating (eco-)physiological properties of the organisms plants. That said, during the course of evolution, these reg- but, instead, the responsiveness of genes in a conserved ulatory networks will have been rewired multiple times stress-regulatory response network (see also Furst-Jansen€ (Furst-Jansen€ et al., 2020). Through comparative et al., 2020). Irrespective of the habitat in which these approaches, we can pinpoint which parts of the wires extant algae live, inferences made from the stress-respon- (such as protein–protein interactions, phosphorylation, TF- siveness observed here shed light on the homologous binding, and many more) are conserved. Yet, the mere fact genetic chassis that might have existed in their last com- that the required components (e.g. CDPKs, PP2Cs) are pre- mon ancestor. sent is already instructive; it highlights the genetic poten- We have detected a variety of signaling components tial residing within the Zygnematophyceae (de Vries and (such as those for phosphorelay systems) in Zygnemato- Archibald, 2018). It is this potential that served as a hotbed phyceae that are known to play important roles in the for the evolution of the plant perceptron that is so critical stress response system of land plants. The networks regu- for the molecular stress physiology of land plants. lating plant physiology are interwoven on multiple levels. CONCLUSIONS This is well-illustrated by the interconnections of ABA- and Ca2+-mediated signaling, comprising a convergence that The earliest land plants must have had the means to likely occurs at PP2Cs in land plants (Edel and Kudla, respond to and ultimately overcome terrestrial stressors 2016). Regarding their role in ABA-mediated signal trans- (de Vries and Archibald, 2018; Furst-Jansen€ et al., 2020). duction processes, the protein-protein interactions A key terrestrial stressor is drastic temperature elevation. upstream and downstream of PP2Cs appear to be We have shown that the zygnematophyceaen algae

Mougeotia sp. and S. pratensis have distinct responses to For confocal laser scanning microscopy, algae were incubated heat. Differences in the response were visible phenotypi- for 10 min in liquid freshwater F/2 containing 1% calcoﬂuor white cally, as well as at the molecular level (metabolome, tran- (Fluorescent Brightener 28; Sigma-Aldrich, St Louis, MO, USA). Specimens were washed with and mounted in liquid freshwater scriptome). These differences likely reﬂect that the two F/2. Imaging of the samples was done on a LSM710 system (Carl algae diverged from a common ancestor about 500 mya Zeiss). ZEN software (Carl Zeiss) was used for the process of image (Morris et al., 2018). That said, their shared responses to capturing. Final images were generated using FIJI (Schindelin heat stress are instructive: they reveal a set of genes et al., 2012). encoding proteins that are known as regulators of the RNA extraction and sequencing stress response in land plants. The responsiveness of these regulators to heat not only tells us about how both embry- Algal material was scraped off the plates and immediately homog- ophytes and their algal relatives deal with elevated temper- enized in a Tenbroek tissue homogenizer. RNA was extracted using the Trizol reagent (Thermo Fisher, Walthm, MA, USA) in atures, but also pinpoints those components that form the accordance with the manufacturer’s instructions. For Mougeotia perceptron of land plants (Scheres and van der Putten, sp., we isolated RNA from six control and three heat samples; for 2017). Our data suggest that the perceptron of the earliest S. pratensis, we isolated RNA from three control samples and land plants featured the responsiveness of a relay system three heat samples. RNA samples were treated with DNAse I of calcium and histidine kinases that may have tied envi- (Thermo Fisher); we assessed RNA quality using a formamide agarose gel and quantity using a Nanodrop spectrometer (Thermo ronmental inputs (and their effects on cell and plastid Fisher). RNA of adequate quality was shipped on dry ice to Gen- physiology) to the appropriate responses. Genetic dissec- ome Quebec (Montreal, Canada). After an additional round of tion of the cornerstones in this relay system will highlight quality assessment using a Bioanalyzer (Agilent Technologies Inc., the degree of functional conservation. The most promising Santa Clara, CA, USA), libraries were constructed using the NEB candidates for such an approach have been outlined in the mRNA stranded Library preparation kit (New England Biolabs, Beverly, MA, USA) and sequenced on the NovaSeq 6000 platform present study. (Illumina). In total, we obtained 459 187 218 reads (92.76 Gbp).

EXPERIMENTAL PROCEDURES RNA-seq analysis Culturing and treatments All raw RNA-seq read data were quality assessed using FASTQC, version 0.11.7 (FASTQC, 2018), and trimmed using TRIMMOMATIC, ver- Cultures of Mougeotia sp. (MZCH240; originally collected in sion 0.36 (Bolger et al., 2014), with the settings: ILLUMINACLIP: Byron’s Pool, Trumpington, Cambridge, UK, and deposited by TruSeq3-PE- 2.fa:2:30:10:2:TRUE HEADCROP:10 TRAILING:3 SLI- E. A. George in 1949) and Spirogyra pratensis (MZCH10213; origi- DINGWINDOW:4:20 MINLEN:36. All trimmed data were uploaded nally deposited by M. A. Allen in 1958) were provided by the to the Sequence Read Archive (SRA; https://www.ncbi.nlm.nih. Microalgae and Zygnematophyceae Collection, Hamburg, Ger- gov/sra) (accession numbers SRR9083681 to SRR9083701). many (Schwartzenberg et al., 2013). We used 7-day-old cultures of Trimmed data were quality assessed again (Bolger et al., 2014). Mougeotia sp. and 12-day-old Spirogyra pratensis for every exper- For each species, we pooled all the trimmed read data (to obtain iment. The algae were grown on modified freshwater F/2 (Guil- the best possible assembly for Mougeotia sp., we added addi- lard, 1975) with 1% agar at 22°C. An LED light source provided tional reads from the same isolate that are however not part of photosynthetically active radiance at 120 µmol quanta mÀ2 secÀ1 this study and the downstream differential analyses) and com- under a 12:12 h light/dark photocycle. Heat stress treatments (in a puted a de novo assembly using the TRINITY pipeline (Haas et al., 37°C culture room) were performed at the same light regime and 2013), which included various isoforms and transcripts with low photoperiod. All experiments were carried out 6 h after the light abundance. We, hence, next performed a series of steps to refine was turned on; for the heat-treated algae this means that, after and filter the data. 6 days (in the case of Mougeotia sp.) and 11 days (Spirogyra Using RSEM (RNA-Seq by Expectation Maximization) (Li and pratensis)at22°C, the algae experienced 6 h light, 12 h darkness Dewey, 2011) we mapped the trimmed reads onto the contigs and and 6 h light at 37°C. calculated expression values as TPM that we normalized using the trimmed mean of M values (TMM) procedure (Robinson and Osh- Microscopy lack, 2010). Making use of our biological replicates, we calculated differential gene expression changes comparing heat versus con- Light microscopy was performed using Nomarski intereference trol using EDGER (Robinson et al., 2010). The TRINITY pipeline clus- contrast on an Axioplan II system (Carl Zeiss, Oberkochen, Ger- ters de novo assembled transcripts into genes and isoforms; here, many) with an Axiocam HRC colour camera (Zeiss). Specimens we worked only with the ‘genes’ datasets and filtered out lowly were mounted in liquid freshwater F/2. For each individual, fully expressed genes by retaining only those that had an average TPM imaged cell (i.e. where a whole cell was in the field of view), mea- over all the libraries that were higher than the number of libraries surements of cell as well as plastid lengths and widths were taken sequenced. The highest expressed isoform was chosen as the rep- using FIJI (Schindelin et al., 2012). R STUDIO (RStudio Team, 2016) resentative for each gene. Querying all the representative was used to plot the data and perform the statistics. A Shapiro– sequences against the NCBI nr database (https://www.ncbi.nlm. Wilk test (Shapiro and Wilk, 1965) was used to determine normal- nih.gov/protein/) via DIAMOND (Buchfink et al., 2015), we assigned a ity of the data. If normally distributed, equal variance was tested taxonomic affiliation to our transcripts and retained only those and a two-sample t-test or Welch’s two sample t-test (Gosset, sequences with the best hit against Chloroplastida. We thus 1908) was performed accordingly. If not normally distributed, a obtained 4955 contigs for Mougeotia sp. and 3950 contigs for Mann–Whitney U-test was performed (Mann and Whitney, 1947).

Spirogyra pratensis (all sequence data have been deposited under K. nitens (Hori et al., 2014), C. braunii (Nishiyama et al., 2018), the NCBI BioProject PRJNA543475). Next, we in silico translated M. polymorpha (Bowman et al., 2017), P. patens (Lang et al., our transcripts into proteins using the EMBOSS getorf tool (Rice 2018), S. moellendorffii (Banks et al., 2011), Picea abies (Nystedt et al., 2000) and a custom perl script that evaluated all possible et al., 2013) and A. thaliana TAIR10 (Berardini et al., 2015). protein sequences and selected the best option; this process was Sequences were aligned using MAFFT (Katoh et al., 2002; Katoh and based on the BLASTX alignment obtained when querying the tran- Standley, 2013) and maximum likelihood phylogenies were com- scripts against a database of well-curated Chloroplastida gen- puted using IQ-TREE (Nguyen et al., 2015). omes. Any work with the sequence data was based on the protein sequences. Quantative reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of cytokinin treated Metabolite extraction and metabolomics using liquid S. pratensis chromatography-tandem mass spectrometry One-week-old S. pratensis cultures were harvested and washed Algal material was scraped off the agar plates and immediately with fresh medium. Approximately 0.2 g fresh weight of algal fila- frozen in liquid nitrogen. The fresh weight of the frozen tissue was ments was suspended in 20 ml of medium and treated with differ- determined. For extraction, we transferred the frozen tissue into a ent cytokinins for the indicated times. The total RNA was isolated 4°C cold MeOH:ddH2O mixture (7:3 v/v) and homogenized, on ice, as described in Onate-S~ anchez and Vicente-Carbajosa (2008) and in a Tenbroek tissue homogenizer. Debris was removed from the the concentration and purity of the RNA were measured spec- homogenate by (cooled) centrifugation. The supernatant was trophotometrically, RNA samples were treated with RNase free dried using a vacuum centrifuge (SPD111V SpeedVac; Thermo DNase I (Thermo Scientific) and cDNA synthesis was performed Electron Corporation, Waltham, MA, USA) and resuspended in using the RevertAid Reverse Transcriptase (Thermo Scientific). 1 ml of buffer containing 95% acetonitrile, 5% 20 mM ammonium Real-time PCR was performed using a StepOnePlusTM Real-Time carbonate (pH 9.8) and 1.25 µM metabolomics amino acid mix PCR System (#4376600; Applied Biosystems, Foster City, CA, standard (PN MSK-A2-1.2; Cambridge Isotope Laboratories, USA). SYBR Fast qPCR master mix (#KM4102; KAPA Biosystems, Tewksbury, MA, USA), the latter used as internal standard. Then, Wilmington, MA, USA) was used in accordance with the manufac- 100 µl of each sample were pooled together for quality controls turer’s instructions. A test for contamination with genomic DNA (QCs). All samples including QCs, where then analyzed by liquid was carried out by PCR using the SpUBI-F/R primers on control chromatography tandem mass spectrometry (LC-MS/MS) using samples prepared without reverse transcriptase. The temperature selected reaction monitoring (SRM) (also known as multiple reac- profile was set as: 3 min of initial denaturation at 95°C; then 40 tion monitoring) acquisitions on an Agilent 1100 liquid chro- cycles of 30 sec at 95°C, 15 sec at 58°C and 30 sec at 72°C. Data matograph coupled in-line to a QTRAP 5500 (Sciex, Framingham, were acquired at 72°C. A melt curve was run at the end of the 40 MA, USA) triple-quadrupole linear ion trap tandem mass spec- cycles to test for a homogenous PCR product. Amplification effi- trometer. Chromatographic separation was achieved by injecting ciency was determined using a standard curve and expressions of 50 µl of sample on an XBridgeâ Amide 3.5 µm particle, target genes were calculated by the 2ÀDDCt method (Livak and Sch- 1.0 9 50 mm column (PN 186004871; Waters, Milford, MA, USA) mittgen, 2001) relative to expression of Ubiquitin gene as refer- heated to 37°C across an 8 min gradient starting at 5% 20 mM ence. The primers sequences were: SpUBI-F 5’-GAGTCCACCCTT ammonium carbonate (pH 9.8) and 95% acetonitrile, ending at CACCTTGT-3’; SpUBI-R 5’-CTTGGGCCTCAATTGATTGGT-3’; 98% 20 mM ammonium carbonate pH 9.8 and 2% acetonitrile (for SpCHK1-F 5’-CTGCAAGAGACAGGACGAAT-3’; SpCHK1-R 5’- details on the gradient, see Data S4). The LC flow rate was set at CAAAGGCTTGGTGAGAGTCC-3’; SpCHK2-F 5’-TGCACCGATCG- –1 200 µl min . A 475°C heated-assisted electrospray source (Tur- TATTCTCTT-3’; SpCHK2-R 5’-AAACAGGGAGAAGGGCAACT-3’; boIonSpray; Sciex) was used for ionization; curtain gas was kept SpCHK3-F 5’-GTGCAAACCCTCTGTTGTCC-3’; SpCHK3-R 5’-TTTC at 20 units and gas 1 and 2 at 33 units (arbitrary units in ANALYST, CTCCGTTTGCCAATCC-3’; SpRRA-F 5’-TGCCGTTGCTCCTAACAA version 1.6.2; Sciex). Ionization spray voltage was set to À4.5 and AG-3’; SpRRA-R 5’-TGGAGCTTTCATCTTCAAGCC-3’; SpRRB-F 5’- 5.5 kV for negative and positive ionisation mode, respectively. The ACGAGCGCAGTAATGAAAGG-3’; SpRRB-R 5’-AGCTCAGGTCTG- data acquisition considered 565 transitions for the analysis of GATTGGAC-3’. main central metabolites such as amino acids, sugars, lipids, fatty acids, small dicarboxylic acids and nucleobases; this approach ACKNOWLEDGEMENTS entailed the use of internal standards for all 20 proteinogenic SdV thanks the Killam Trusts for an Izaak Walton Postdoctoral Fel- amino acids. The method was optimized for metabolomics on lowship. JdV thanks the German Research Foundation (DFG) for a human samples. Thus, we analyzed a set of 565 tentative metabo- Research Fellowship (VR132/1-1) and a Return Grant (VR132/3-1); lites (with internal standards for the amino acids), by adapting JdV further thanks the European Research Council for funding and tuning the analytical parameters reported previously (Yuan under the European Union’s Horizon 2020 research and innovation et al., 2012; Li et al., 2017). Each target was identified using a diag- programme (Grant Agreement No. 852725; ERC-StG ‘TerreStriAL’). nostic transition in a particular retention time window as JMA acknowledges funding from the Natural Sciences and Engi- described by Yuan et al. (2012) and Li et al. (2017). Data were cap- neering Research Council of Canada (Discovery grant RGPIN-2014- tured using the ANALYST, VERSION 1.6.2 software (Sciex); peak inte- 05871). gration was performed using SKYLINE VERSION 19.1 (MacLean et al., 2010; Pino et al., 2017). CONFLICT OF INTERESTS

Phylogenetic analysis The authors declare that they have no competing interests.

Sequences for phylogenetic analyses were retrieved from the pro- AUTHOR CONTRIBUTIONS tein data of the in silico translated transcriptomes of Mougeotia sp. and S. pratensis, published transcriptomes of streptophyte JdV designed the study. JdV, SdV and HZ performed algae (Ju et al., 2015; de Vries et al., 2018), and the genomes of experiments. Algae were provided by KvS, JdV, SdV and

AMC carried out the metabolite extraction. AMC and SP Figure S4. Unrooted phylogeny of histidine kinases. A phylogeny generated and processed the mass spectrometry data. was computed based on a maximum likelihood (ML) AMC, SP, KF and DMP analyzed the metabolome. JdV, WAG+F + I+G4 method using IQ-tree (bootstrap values < 50 are not shown) AHK homologs of (a) the land plants Arabidopsis thali- SdV and BAC performed the bioinformatic analyses. JdV ana and Physcomitrella patens and (b) the Zygnematophyceae and SdV performed statistics. JdV and JMA acquired fund- Mougeotia sp. and Spirogyra pratensis. ing for the study. JMA, AMC and MS provided research Figure S5. (a) Relative expression levels of SpCHK1, SpCHK2, infrastructure. JdV wrote the initial draft with JMA and SpCHK3, SpRRA and SpRRB affected by cytokinin (400 nM iP for SdV All authors contributed to writing the final manuscript. 4 h). Three biological replicates were measured with two technical All authors discussed the results. replicates. A ubiquitin gene in S. pratensis served as the endogenous control, and gene expression was normalized to levels of DATA AVAILABILITY STATEMENT untreated samples (0 M). (b) Relative expression levels of SpCHK1, SpCHK2 and SpRRA affected by different types of cytokinin (1 lM All sequence data have been deposited under the NCBI iP, tZ and cZ for 2 h). BioProject PRJNA543475. The Transcriptome Shotgun Data S1. Sequencing statistics. Assembly project for Spirogyra pratensis MZCH-SVCK Data S2. For the two species Mougeotia sp. (Mousp) and Spir- 10213 has been deposited at DDBJ/EMBL/GenBank under ogyra pratensis (Spipra) all transcripts for ‘genes’ (as defined by TRINITY) that were retained after filtering are shown. The columns the accession GICF00000000. The version described in this under ‘TPM’ show the expression levels for all biological repli- paper is the first version, GICF01000000. The Transcrip- cates (control and heat) in trimmed mean of M values (TMM)-nor- tome Shotgun Assembly project for Mougeotia sp. MZCH- malized TPM values. Based on in silico-translation of the SVCK 240 has been deposited at DDBJ/EMBL/GenBank representative transcript, a BLASTp was performed against pro- under the accession GHUK00000000. The version described tein data from the genomes of Arabidopsis thaliana (TAIR10; At), Physcomitrella patens (V3; Pp), Chara braunii (Cb) and Kleb- in this paper is the first version, GHUK01000000. The corre- sormidium nitens (Kn). The column to the right of the respective sponding reads are deposited in the NCBI Sequence Read best BLASTp hits indicates whether this hit was recovered recipro- Archive (SRA) under the accessions SRX5858892 to cally. The last two columns show the results of the differential SRX5858912. Additional light micrographs regarding the gene expression analysis (heat versus control) in log2(fold change) quantifcation of the stress phenotypes are deposited at and the Benjamini–Hochberg-corrected P values. Zenodo under http://doi.org/10.5281/zenodo.3769748. Data S3. Relative metabolite abundance normalized on fresh- weight of the samples. On the top, the M/Z transition (precur- SUPPORTING INFORMATION sor ? product; black) and the tentative metabolites (dark grey) are given. On the left, the biological replicates, the species, and the Additional Supporting Information may be found in the online ver- condition under which the sample was taken (control or after 24 h sion of this article. of heat stress) are labeled. All data are based on technical dupli- cates. Figure S1. Phylogeny of CAD-related sequences. Sequences were Data S4. Top: gradient of solvent system A, 5% acetonitrile. Of identified based on a BLASTP/TBLASTN search with CAD sequences 20 mM ammonium carbonate pH: 9.8, and solvent system B, 100% from Arabidopsis thaliana as query against a sequence database acetonitrile. Bottom: pump details. of genomic and transcriptomic data from bryophytes and streptophyte algae. Additional sequences that were identified in de Vries Data S5. The differential gene expression data plotted in Figure 6 et al. (2017) were also included. Transcriptomic data from this for the two Zygnematophyceae, Mougeotia sp. (Mousp) and Spir- study and de Vries et al. (2018) were filtered and only sequences ogyra pratensis (Spipra). For each transcript, the best BLASTP hit with a TPM ≥ 1 in at least one condition were included. The phy- against protein data from the genomes of Arabidopsis thaliana logeny is based on full-length sequences, which were aligned (TAIR10; At) is shown. The last two columns show the results of using MAFFT (option G-INS-I). Annotation for CAD and CAD-like the differential gene expression analyses (heat versus control) in – sequences is based on de Vries et al., 2017 and annotation of the log2(fold change) and the Benjamini Hochberg-corrected P values. other sequences is based on a BLASTP search using the sequences as query and the genome of the Arabidopsis thaliana TAIR10 REFERENCES release as the database. The e-value cut-off for all BLASTP searches was 10. SOR-like sequences were used as an outgroup. Ahlfors, R., Lang, S., Overmyer, K. et al. (2004) Arabidopsis RADICAL- Figure S2. A bar diagram showing the differential gene expression INDUCED CELL DEATH1 belongs to the WWE protein-protein interaction domain protein family and modulates abscisic acid, ethylene, and methyl (as log [fold change ]) of the 207 concordantly down-reg- 2 heat/control jasmonate responses. Plant Cell, 16, 1925–1937. ulated, shared orthologs depicted in Figure 4B. The diagram is Al-Babili, S. and Bouwmeester, H.J. (2015) Strigolactones, a novel carote- sorted by the between species average log2(fc) values for each noid-derived plant hormone. Annu. Rev. Plant Biol. 66, 161–186. given ortholog. Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Figure S3. Up-regulated homologs of Arabidopsis thaliana pro- Basic local alignment search tool. J. Mol. Biol. 215, 403–410. teins that are associated with cytoskeletal organization detected in Atiena, S.G., Facciolo, P., Perrotta, G., Dalfino, G., Zschiesche, W., Hum- the Zygnematophyceae Mougeotia sp. and Spirogyra pratensis. beck, K., Stanca, A.M. and Cattivelli, L. (2004) Large scale analysis of transcripts abundance in barley subjected to several single and com- Protein domains found in the (in silico translated) protein candi- bined abiotic stress conditions. Plant Sci. 167, 1359–1365. dates were predicted via INTERPROSCAN. On the left, gradient-colored Avendano-V~ azquez, A.O., Cordoba, E., Llamas, E. et al. (2014) An uncharac- circles indicate the associated gene expression data (log2[fold terized apocarotenoid-derived signal generated in f-carotene desaturase change heat/control], calculated using edgeR) from green (up-reg- mutants regulates leaf development and the expression of chloroplast ulation) to blue (unchanged). and nuclear genes in Arabidopsis. Plant Cell, 26, 2524–2537.

Azimzadeh, J., Nacry, P., Christodoulidou, A., Drevensek, S., Camilleri, C., Deng, Y., Humbert, S., Liu, J.X., Srivastava, R., Rothstein, S.J. and Howell, Amiour, N., Parcy, F., Pastuglia, M. and Bouchez, D. (2008) Arabidopsis S.H. (2011) Heat induces the splicing by IRE1 of a mRNA encoding a tran- TONNEAU1 proteins are essential for preprophase band formation and scription factor involved in the unfolded protein response in Arabidopsis. interact with centrin. Plant Cell, 20, 2146–2159. Proc. Natl Acad. Sci. USA, 108, 7247–7252. Banks, J.A., Nishiyama, T., Hasebe, M. et al. (2011) The Selaginella genome Dubrovina, A.S., Kiselev, K.V., Khristenko, V.S. and Aleynova, O.A. (2017) identifies genetic changes associated with the evolution of vascular The calcium-dependent protein kinase gene VaCPK29 is involved in plants. Science, 332, 960–963. grapevine responses to heat and osmotic stresses. Plant Growth Regul. Becker, B. and Marin, B. (2009) Streptophyte algae and the origin of embry- 82,79–89. ophytes. Ann. Bot. 103, 999–1004. Eberhard, S., Finazzi, G. and Wollman, F.-A. (2008) The dynamics of photo- Berardini, T.Z., Reiser, L., Li, D., Mezheritsky, Y., Muller, R., Strait, E. and synthesis. Annu. Rev. Genet. 42, 463–515. Huala, E. (2015) The Arabidopsis information resource: making and min- Edel, K.H. and Kudla, J. (2015) Increasing complexity and versatility: how ing the “gold standard” annotated reference plant genome. Genesis, 53, the calcium signaling toolkit was shaped during plant land colonization. 474–485. Cell Calcium, 57, 231–246. Berry, J. and Bjorkman,€ O. (1980) Photosynthetic response and adaptation Edel, K.H. and Kudla, J. (2016) Integration of calcium and ABA signaling. to temperature in higher plants. Annu. Rev. Plant. Physiol. 31, 491–543. Curr. Opin. Plant Biol. 33,83–91. Bolger, A.M., Lohse, M. and Usadel, B. (2014) Trimmomatic: a flexible trim- Edel, K.H., Marchadier, E., Brownlee, C., Kudla, J. and Hetherington, A.M. mer for Illumina sequence data. Bioinformatics, 30, 2114–2120. (2017) The evolution of calcium-based signalling in plants. Curr. Biol. 27, Bowman, J.L., Kohchi, T., Yamato, K.T. et al. (2017) Insights into land plant R667–R679. evolution garnered from the Marchantia polymorpha genome. Cell, 171, Eklund, D.M., Kanei, M., Flores-Sandoval, E. et al. (2018) An evolutionarily 287–304. conserved abscisic acid signaling pathway regulates dormancy in the liv- Bowman, J.L., Briginshaw, L.N., Fisher, T.J. and Flores-Sandoval, E. (2019) erwort Marchantia polymorpha. Curr. Biol. 28, 3691–3699.e3. Something ancient and something neofunctionalized – evolution of land Estavillo, G.M., Crisp, P.A., Pornsiriwong, W. et al. (2011) Evidence for a plant hormone signaling pathways. Curr. Opin. Plant Biol. 47,64–77. SAL1-PAP chloroplast retrograde pathway that functions in drought and Brenner, W.G. and Schmulling,€ T. (2015) Summarizing and exploring data high light signaling in Arabidopsis. Plant Cell, 23, 3992–4012. of a decade of cytokinin-related transcriptomics. Front. Plant Sci. 6, 29. Falcone, D.L., Ogas, J.P. and Somerville, C.R. (2004) Regulation of mem- https://doi.org/10.3389/fpls.2015.00029. brane fatty acid composition by temperature in mutants of Arabidopsis Brocard-Gifford, I., Lynch, T.J., Garcia, M.E., Malhotra, B. and Finkelstein, with alterations in membrane lipid composition. BMC Plant Biol. 4, 17. R.R. (2004) The Arabidopsis thaliana ABSCISIC ACID-INSENSITIVE8 locus FASTQC (2018) A Quality Control Tool for High Throughput Sequence Data. encodes a novel protein mediating abscisic acid and sugar responses Available at www.bioinformatics.babraham.ac.uk/projects/fastqc. essential for growth. Plan Cell, 16, 406–421. Accessed September 15, 2018. Buchfink, B., Xie, C. and Huson, D.H. (2015) Fast and sensitive protein align- Feder, M.E. and Hofmann, G.E. (1999) Heat-shock proteins, molecular chap- ment using DIAMOND. Nat. Methods, 12,59–60. erones, and the stress response: evolutionary and ecological physiology. Campbell, R., Ducreux, L.J.M., Morris, W.L., Morris, J.A., Suttle, J.C., Ram- Annu. Rev. Physiol. 61, 243–282. say, G., Bryan, G.J., Hedley, P.E. and Taylor, M.A. (2010) The metabolic Feraru, E., Feraru, M.I., Barbez, E., Waidmann, S., Sun, L., Gaidora, A. and and developmental roles of carotenoid cleavage Dioxygenase4 from Kleine-Vehn, J. (2019) PILS6 is a temperature-sensitive regulator of potato. Plant Physiol. 154, 656–664. nuclear auxin input and organ growth in Arabidopsis thaliana. Proc. Natl Chan, K.X., Phua, S.Y., Crisp, P., McQuinn, R. and Pogson, B.J. (2016) Learn- Acad. Sci. USA, 116, 3893–3898. ing the language of the chloroplast: retrograde signaling and beyond. Fernandez, A.P. and Strand, A. (2008) Retrograde signaling and plant stress: Annu. Rev. Plant Biol. 67,25–53. plastid signals initiate cellular stress responses. Curr. Opin. Plant Biol. Cheng, S.-H., Willmann, M.R., Chen, H.-C. and Sheen, J. (2002) Calcium sig- 11, 509–513. naling through protein kinases. The Arabidopsis Calcium-Dependent Pro- Fernandez-Bautista, N., Fernandez-Calvino, L., Munoz,~ A., Toribio, R., Mock, tein Kinase gene family. Plant Physiol. 129, 469–485. H.P. and Castellano, M.M. (2018) HOP family plays a major role in long- Cheng, S., Xian, W., Fu, Y. et al. (2019) Genomes of subaerial Zygnemato- term acquired thermotolerance in Arabidopsis. Plant Cell Environ. 41, phyceae provide insights into land plant evolution. Cell, 179, 1057– 1852–1869. 1067.e14. Ferro, M., Brugiere, S., Salvi, D. et al. (2010) AT_CHLORO, a comprehensive Chinnusamy, V., Zhu, J. and Zhu, J.-K. (2007) Cold stress regulation of gene chloroplast proteome database with subplastidial localization and expression in plants. Trends Plant Sci. 12, 444–451. curated information on envelope proteins. Mol. Cell. Proteomics, 9, Chong, J., Yamamoto, M. and Xia, J. (2019) MetaboAnalystR 2.0: from raw 1063–1084. spectra to biological insights. Metabolites, 9, 57. Finka, A., Cuendet, A.F.H., Maathuis, F.J.M., Saidi, Y. and Goloubinoff, P. Davletova, S., Mesz aros, T., Miskolczi, P., Oberschall, A., Tor€ ok,€ K., Magyar, (2012) Plasma membrane cyclic nucleotide gated calcium channels con- Z., Dudits, D. and Deak, M. (2001) Auxin and heat shock activation of a trol land plant thermal sensing and acquired thermotolerance. Plant Cell, novel member of the calmodulin like domain protein kinase gene family 24, 3333–3348. in cultured alfalfa cells. J. Exp. Bot. 52, 215–221. Foyer, C.H., Lelandais, M. and Kunert, K.J. (1994) Photooxidative stress in de Vries, J., Stanton, A., Archibald, J.M. and Gould, S.B. (2016) Strepto- plants. Physiol Plant. 92, 696–717. phyte terrestrialization in light of plastid evolution. Trends Plant Sci. 21, Frandsen, G.I., Mundy, J. and Tzen, J. (2001) Oil bodies and their associated 467–476. proteins, oleosin and caleosin. Physiol Plant. 112, 301–307. de Vries, J., Curtis, B.A., Gould, S.B. and Archibald, J.M. (2018) Embryo- Furst-Jansen,€ J.M.R., de Vries, S. and de Vries, J. (2020) Evo-physio: on phyte stress signaling evolved in the algal progenitors of land plants. stress responses and the earliest land plants. J. Exp. Bot. eraa007, Proc. Natl Acad. Sci. USA, 115, E3471–E3480. https://doi.org/10.1093/jxb/eraa007. de Vries, J., de Vries, S., Slamovits, C.H., Rose, L.E. and Archibald, J.M. Fujibe, T., Saji, H., Arakawa, K., Yabe, N., Takeuchi, Y. and Yamamoto, K.T. (2017) How embryophytic is the biosynthesis of phenylpropanoids (2004) A methyl viologen-resistant mutant of Arabidopsis, which is allelic and their derivatives in streptophyte algae? Plant Cell Physiol. 58, to ozone-sensitive rcd1, is tolerant to supplemental ultraviolet-B irradia- 934–945. tion. Plant Physiol. 134, 275–285. de Vries, J. and Archibald, J.M. (2018) Plant evolution: landmarks on the Furihata, T., Maruyama, K., Fujita, Y., Umezawa, T., Yoshida, R., Shinozaki, path to terrestrial life. New Phytol. 217, 1428–1434. K. and Yamaguchi-Shinozaki, K. (2006) Abscisic acid-dependent multisite Delwiche, C.F. (2016) The genomes of charophyte green algae. In Advances phosphorylation regulates the activity of a transcription activator AREB1. in Botanical Research: Genomes and Evolution of Charophytes, Bryo- Proc. Natl Acad. Sci. USA, 103, 1988–1993. phytes, Lycophytes and Ferns (Rensing, S.A., ed). London, UK: Elsevier, Fuss, J., Liegmann, O., Krause, K. and Rensing, S.A. (2013) Green targeting pp. 255–270. predictor and ambiguous targeting predictor 2: The pitfalls of plant pro- Delwiche, C.F. and Cooper, E.D. (2015) The evolutionary origin of a terres- tein targeting prediction and of transient protein expression in heterolo- trial flora. Curr. Biol. 25, R899–R910. gous systems. New Phytol. 200, 1022–1033.

Garcıa-Cerdan, J.G., Kovacs, L., Toth, T., Kere€ıche, S., Aseeva, E., Boekema, Jahns, P. and Holzwarth, A.R. (2012) The role of the xanthophyll cycle and E.J., Mamedov, F., Funk, C. and Schroder,€ W.P. (2011) The PsbW protein of lutein in photoprotection of photosystem II. BBA – Bioenerget. 1817, stabilizes the supramolecular organization of photosystem II in higher 182–193. plants. Plant J. 65, 368–381. Jeon, J., Kim, N.Y., Kim, S. et al. (2010) A subset of cytokinin two-compo- Gerotto, C. and Morosinotto, T. (2013) Evolution of photoprotection mecha- nent signaling system plays a role in cold temperature stress response in nisms upon land colonization: evidence of PSBS- dependent NPQ in late Arabidopsis. J. Biol. Chem. 285, 23371–23386. Streptophyte algae. Physiol. Plant. 149, 583–598. Jiao, C., Sørensen, I., Sun, X. et al. (2019) The genome of the charophyte Gil, P., Dewey, E., Friml, J., Zhao, Y., Snowden, K.C., Putterill, J., Palme, K., alga Penium margaritaceum bears footprints of the evolutionary origins Estelle, M. and Chory, J. (2001) BIG: a calossin-like protein required for of land plants. bioRxiv. doi: https://doi.org/10.1101/835561. polar auxin transport in Arabidopsis. Genes Dev. 15, 1985–1997. Ju, C., Van de Poel, B., Cooper, E.D., Thierer, J.H., Gibbons, T.R., Delwiche, Gong, M., van der Luit, A.H., Knight, M.R. and Trewavas, A.J. (1998) Heat- C.F. and Chang, C. (2015) Conservation of ethylene as a plant hormone shock-induced changes in intracellular Ca2+ level in tobacco seedlings in over 450 million years of evolution. Nat. Plants, 1, 14004. relation to thermotolerance. Plant Physiol. 116, 429–437. Kacprzak, S.M., Mochizuki, N., Naranjo, B., Xu, D., Leister, D., Kleine, T., Gontcharov, A.A. and Melkonian, M. (2010) Molecular phylogeny and revi- Okamoto, H. and Terry, M.J. (2019) Plastid-to-nucleus retrograde sig- sion of the genus Netrium (Zygnematophyceae, Streptophyta): Nucleo- nalling during chloroplast biogenesis does not require ABI4. Plant Phys- taenium gen. nov. J. Phycol. 46, 346–362. iol. 179,18–23. Gosset, W.S. (1908) The probable error of a mean. Biometrika, 6,1–25. Kanehisa, M., Sato, Y. and Morishima, K. (2016) BlastKOALA and Ghost- Guillard, R.R.L. (1975) – Culture of phytoplankton for feeding marine inverte- KOALA: KEGG tools for functional characterization of genome and meta- brates. In Culture of marine invertebrate animals. (Smith, W.L. and Chan- genome sequences. J. Mol. Biol. 428, 726–731. ley, M.H., eds). New York: Plenum Book Publ. Corp, pp. 29–60. Kaplan, F., Kopka, J., Haskell, D.W., Zhao, W., Schiller, K.C., Gatzke, N., Guy, C., Kaplan, F., Kopka, J., Selbig, J. and Hincha, D.K. (2008) Metabolo- Sung, D.Y. and Guy, C.L. (2004) Exploring the temperature-stress meta- mics of temperature stress. Physiol. Plant. 132, 220–235. bolome of Arabidopsis. Plant Physiol. 136, 4159–4168. Haas, B.J., Papanicolaou, A. and Yassour, M. et al. (2013) De novo transcript Kaplan, B., Davydov, O., Knight, H., Galon, Y., Knight, M.R., Fluhr, R. and sequence reconstruction from RNA-seq using the Trinity platform for ref- Fromm, H. (2006) Rapid transcriptome changes induced by cytosolic erence generation and analysis. Nat. Protoc. 8, 1494–1512. Ca2+ transients reveal ABRE-related sequences as Ca2+-responsive cis Hagman, A., Shi, L.X., Rintamaki,€ E., Andersson, B. and Schroder,€ W.P. elements in Arabidopsis. Plant Cell, 18, 2733–2748. (1997) The nuclear-encoded PsbW protein subunit of photosystem II Kasai, K., Usami, S. and Yamada, T. (2002) A RelA–SpoT homolog (Cr-RSH) undergoes light-induced proteolysis. Biochemistry, 36, 12666–12671. identified in Chlamydomonas reinhardtii generates stringent factor Harholt, J., Moestrup, Ø. and Ulvskov, P. (2016) Why plants were terrestrial in vivo and localizes to chloroplasts in vitro. Nucleic Acids Res. 30, 4985– from the beginning. Trends Plant Sci. 21,96–101. 4992. Harmon, A.C., Gribskov, M. and Harper, J.F. (2000) CDPKs – a kinase for Katoh, K., Misawa, K., Kuma, K.-I. and Miyata, T. (2002) MAFFT: a novel every Ca2+ signal? Trends Plant Sci. 5, 154–159. method for rapid multiple sequence alignment based on fast Fourier Harper, J.F., Huang, J.-F. and Lloyd, S.J. (1994) Genetic Identification of an transform. Nucleic Acids Res. 30, 3059–3066. Autoinhibitor in CDPK, a Protein Kinase with a Calmodulin-like Domain. Katoh, K. and Standley, D.M. (2013) MAFFT multiple sequence alignment Biochemistry, 33, 7267–7277. software version 7: improvements in performance and usability. Mol. Hauser, F., Waadt, R. and Schroeder, J.I. (2011) Evolution of abscisic acid Biol. Evol. 30, 772–780. synthesis and signaling mechanisms. Curr. Biol. 21, R346–R355. Kieber, J.J. and Schaller, G.E. (2018) Cytokinin signaling in plant develop- Hemme, D., Veyel, D., Muhlhaus,€ T. et al. (2014) Systems-wide analysis of ment. Development, 145, https://doi.org/10.1242/dev.149344. acclimation responses to long-term heat stress and recovery in the pho- Kim, J.H. and Kim, W.T. (2013) The Arabidopsis RING E3 ubiquitin ligase tosynthetic model organism Chlamydomonas reinhardtii. Plant Cell, 26, AtAIRP3/LOG2 participates in positive regulation of high-salt and drought 4270–4297. stress responses. Plant Physiol. 162, 1733–1749. Herburger, K., Xin, A. and Holzinger, A. (2019) Homogalacturonan accumu- Kim, D., Cho, Y.-H., Ryu, H., Kim, Y., Kim, T.-H. and Hwang, I. (2013) BLH1 lation in cell walls of the green alga Zygnema sp. (Charophyta) increases and KNAT3 modulate ABA responses during germination and early seed- desiccation resistance. Front. Plant Sci. 10, 540. ling development in Arabidopsis. Plant J. 75, 755–766. Hernandez-Verdeja, T. and Strand, A. (2018) Retrograde signals navigate Kinnersley, A.M. and Turano, F.J. (2000) Gamma aminobutyric acid (GABA) the path to chloroplast development. Plant Physiol. 176, 967–976. and plant responses to stress. Crit. Rev. Plant Sci. 19, 479–509. Heyl, A., Wulfetange, K., Pils, B., Nielsen, N., Romanov, G.A. and Klein, J.D. and Ferguson, I.B. (1987) Effect of high temperature on calcium Schmulling,€ T. (2007) Evolutionary proteomics identifies amino acids uptake by suspension-cultured pear fruit cells. Plant Physiol. 84, 153–156. essential for ligand-binding of the cytokinin receptor CHASE domain. Kleine, T. and Leister, D. (2016) Retrograde signaling: organelles go net- BMC Evol. Biol. 7, 62. working. BBA Bioenerget. 1857, 1313–1325. Hildebrandt, T.M., Nesi, A.N., Araujo, W.L. and Braun, H.-P. (2015) Amino Kotak, S., Larkindale, J., Lee, U., von Koskull-Doring,€ P., Vierling, E. and acid catabolism in plants. Mol. Plant, 8, 1563–1579. Scharf, K.-D. (2007) Complexity of the heat stress response in plants. Holzinger, A. and Karsten, U. (2013) Desiccation stress and tolerance in Curr. Opin. Plant Biol. 10, 310–316. green algae: consequences for ultrastructure, physiological, and molecu- Kudla, J., Becker, D., Grill, E., Hedrich, R., Hippler, M., Kummer, U., Par- lar mechanisms. Front. Plant Sci. 4, 327. niske, M., Romeis, T. and Schumacher, K. (2018) Advances and current Hori, K., Maruyama, F., Fujisawa, T. et al. (2014) Klebsormidium flaccidum challenges in calcium signaling. New Phytol. 218, 414–431. genome reveals primary factors for plant terrestrial adaptation. Nat. Lang, D., Ullrich, K.K., Murat, F. et al. (2018) The Physcomitrella patens Commun. 5, 3978. chromosome-scale assembly reveals moss genome structure and evolu- Hou, X., Rivers, J., Leon, P., McQuinn, R.P. and Pogson, B.J. (2016) Synthe- tion. Plant J. 93, 515–533. sis and function of apocarotenoid signals in plants. Trends Plant Sci. 21, Larkindale, J. and Knight, M.R. (2002) Protection against heat stress-induced 792–803. oxidative damage in Arabidopsis involves calcium, abscisic acid, ethy- Hrabak, E.M., Chan, C.W.M., Gribskov, M. et al. (2003) The Arabidopsis lene, and salicylic acid. Plant Physiol. 128, 682–695. CDPK-SnRK superfamily of protein kinases. Plant Physiol. 132, 666–680. Lecointre, G. and Le Guyader, H. (2006) The Tree of Life: A Phylogenetic Hutin, C., Nussaume, L., Moise, N., Moya, I., Kloppstech, K. and Havaux, M. Classification. Harvard University Press. (2003) Early light-induced proteins protect Arabidopsis from photooxida- Leebens-Mack, J.H., Barker, M.S., Carpenter, E.J. et al. (2019) One thousand tive stress. Proc. Natl Acad. Sci. USA, 100, 4921–4926. plant transcriptomes and the phylogenomics of green plants. Nature, Jaspers, P., Overmyer, K., Wrzaczek, M., Vainonen, J.P., Blomster, T., Sal- 574, 679–685. ojarvi,€ J., Reddy, R.A. and Kangasjarvi,€ J. (2010) The RST and PARP-like Lenzoni, G. and Knight, M.R. (2018) Increases in absolute temperature stim- domain containing SRO protein family: analysis of protein structure, ulate free calcium concentration elevations in the chloroplast. Plant Cell function and conservation in land plants. BMC Genomics, 11, 170. Physiol. 60, 538–548.

Lewis, L.A. and McCourt, R.M. (2004) Green algae and the origin of land Morris, J.L., Puttick, M.N., Clark, J.W. et al. (2018) The timescale of early plants. Am. J. Bot. 91, 1535–1556. land plant evolution. Proc. Natl Acad. Sci. USA, 115, E2274–E2283. Li, B. and Dewey, C.N. (2011) RSEM: Accurate transcript quantification from Murata, N. and Los, D.A. (1997) Membrane fluidity and temperature percep- RNA-Seq data with or without a reference genome. BMC Bioinformatics, tion. Plant Physiol. 115, 875–879. 12, 323. Murchie, E.H., Hubbart, S., Peng, S. and Horton, P. (2005) Acclimation of Li, Z., Yu, Z., Peng, Z. and Huang, B. (2016) Metabolic pathways regulated photosynthesis to high irradiance in rice: gene expression and interac- by c-aminobutyric acid (GABA) contributing to heat tolerance in creeping tions with leaf development. J. Exp. Bot. 56, 449–460. bentgrass (Agrostis stolonifera). Sci. Rep. 6, 30338. Mutte, S.K., Kato, H., Rothfels, C., Melkonian, M., Wong, G.-K.-S. and Wei- Li, F.-W., Melkonian, M., Rothfels, C.J., Villarreal, J.C., Stevenson, D.W., jers, D. (2018) Origin and evolution of the nuclear auxin response sys- Graham, S.W., Wong, G.-K.-S., Pryer, K.M. and Mathews, S. (2015) Phy- tem. eLife, 7, e33399. tochrome diversity in green plants and the origin of canonical plant phy- Nakamoto, H. and Vıgh, L. (2007) The small heat shock proteins and their tochromes. Nat. Commun. 6, 7852. clients. Cell. Mol. Life Sci. 64, 294–306. Li, K., Naviaux, J.C., Bright, T., Wang, L. and Naviaux, R.K. (2017) A robust, Nambara, E. and Marion-Poll, A. (2005) Abscisic acid biosynthesis and cata- single-injection method for targeted, broad-spectrum plasma metabolo- bolism. Annu. Rev. Plant Biol. 56, 165–185. mics. Metabolomics, 13, 122. Narayanan, S., Tamura, P.J., Roth, M.R., Prasad, P.V.V. and Welti, R. (2016) Lima, A., Lima, S., Wong, J.H., Phillips, R.S., Buchanan, B.B. and Luan, S. Wheat leaf lipids during heat stress: I. High day and night temperatures (2006) A redox-active FKBP-type immunophilin functions in accumulation result in major lipid alterations. Plant Cell Environ. 39, 787–803. of the photosystem II supercomplex in Arabidopsis thaliana. Proc. Natl Nguyen, L.-T., Schmidt, H.A., von Haeseler, A. and Minh, B.Q. (2015) IQ- Acad. Sci. USA, 103, 12631–12636. TREE: a fast and effective stochastic algorithm for estimating maximum Lin, Y.-P., Lee, T.-Y., Tanaka, A. and Charng, Y.-Y. (2014) Analysis of an Ara- likelihood phylogenies. Mol. Biol. Evol. 32, 268–274. bidopsis heat-sensitive mutant reveals that chlorophyll synthase is Nishimura, N., Kitahata, N., Seki, M., Narusaka, Y., Narusaka, M., Kuromori, involved in reutilization of chlorophyllide during chlorophyll turnover. T., Asami, T., Shinozaki, K. and Hirayama, T. (2005) Analysis of ABA Plant J. 80,14–26. Hypersensitive Germination2 revealed the pivotal functions of PARN in Lind, C., Dreyer, I., Lopez-Sanjurjo, E.J. et al. (2015) Stomatal guard cells co- stress response in Arabidopsis. Plant J. 44, 972–984. opted an ancient ABA-dependent desiccation survival system to regulate Nishimura, N., Sarkeshik, A., Nito, K. et al. (2010) PYR/PYL/RCAR family stomatal closure. Curr. Biol. 25, 928–935. members are major in-vivo ABI1 protein phosphatase 2C-interacting pro- Lindquist, S. and Craig, E.A. (1988) The heat-shock proteins. Annu. Rev. teins in Arabidopsis. Plant J. 61, 290–299. Genet. 22, 631–677. Nishiyama, T., Sakayama, H., de Vries, J. et al. (2018) The Chara genome: Liu, H.-T., Gao, F., Li, G.-L., Han, J.-L., Liu, D.-L., Sun, D.-Y. and Zhou, R.-G. secondary complexity and implications for plant terrestrialization. Cell, (2008) The calmodulin-binding protein kinase 3 is part of heat-shock sig- 174, 448–464.e24. nal transduction in Arabidopsis thaliana. Plant J. 55, 760–773. Nott, A., Jung, H.-S., Koussevitzky, S. and Chory, J. (2006) Plastid-to-nu- Liu, Z.-B., Wang, J.-M., Yang, F.-X. et al. (2014) A novel membrane-bound cleus retrograde signaling. Annu. Rev. Plant Biol. 57, 739–759. E3 ubiquitin ligase enhances the thermal resistance in plants. Plant Bio- Nystedt, B., Street, N.R., Wetterbom, A. et al. (2013) The Norway spruce tech. J. 12,93–104. genome sequence and conifer genome evolution. Nature, 497, 579–584. Livak, K.J. and Schmittgen, T.D. (2001) Analysis of relative gene expression Ogawa, T., Nishimura, K., Aoki, T., Takase, H., Tomizawa, K.-I., Ashida, H. ÀDDC data using Real-Time quantitative PCR and the 2 T method. Methods, and Yokota, A. (2009) A phosphofructokinase B-type carbohydrate kinase 25, 402–408. family protein, NARA5, for massive expressions of plastid-encoded pho- Long, J.C. and Merchant, S.S. (2008) Photo-oxidative stress impacts the tosynthetic genes in Arabidopsis. Plant Physiol. 151, 114–128. expression of genes encoding iron metabolism components in chlamy- Ohama, N., Sato, H., Shinozaki, K. and Yamaguchi-Shinozaki, K. (2017) domonas. Photochem. Photobiol. 84, 1395–1403. Transcriptional regulatory network of plant heat stress response. Trends Lynch, T., Erickson, B.J. and Finkelstein, R.R. (2012) Direct interactions of Plant Sci. 22,53–65. ABA-insensitive(ABI)-clade protein phosphatase(PP)2Cs with calcium-de- Ohtaka, K., Hori, K., Kanno, Y., Seo, M. and Ohta, H. (2017) Primitive auxin pendent protein kinases and ABA response element-binding bZIPs may response without TIR1 and Aux/IAA in the charophyte alga Klebsormid- contribute to turning off ABA response. Plant Mol. Biol. 80, 647–658. ium nitens. Plant Physiol. 174, 1621–1632. Ma, Y., Szostkiewicz, I., Korte, A., Moes, D., Yang, Y., Christmann, A. and Onate-S~ anchez, L. and Vicente-Carbajosa, J. (2008) DNA-free RNA isolation Grill, E. (2009) Regulators of PP2C phosphatase activity function as absci- protocols for Arabidopsis thaliana, including seeds and siliques. BMC sic acid sensors. Science, 324, 1064–1068. Res. Notes, 1, 93. MacLean, B., Tomazela, D.M., Shulman, N., Chambers, M., Finney, G.L., Fre- Pagant, S., Bichet, A., Sugimoto, K., Lerouxel, O., Desprez, T., McCann, M., wen, B., Kern, R., Tabb, D.L., Liebler, D.C. and MacCoss, M.J. (2010) Sky- Lerouge, P., Vernhettes, S. and Hofte,€ H. (2002) KOBITO1 encodes a line: an open source document editor for creating and analyzing targeted novel plasma membrane protein necessary for normal synthesis of cellu- proteomics experiments. Bioinformatics, 26, 966–968. lose during cell expansion in Arabidopsis. Plant Cell, 14, 2001–2013. Mann, H.B. and Whitney, D.R. (1947) On a test of whether one of two ran- Page, M.T., Kacprzak, S.M., Mochizuki, N., Okamoto, H., Smith, A.G. and dom variables is stochastically larger than the other. Ann. Math. Statist. Terry, M.J. (2017) Seedlings lacking the PTM protein do not show a gen- 1,50–60. omes uncoupled (gun) mutant phenotype. Plant Physiol. 174,21–26. Maurel, C., Boursiac, Y., Luu, D.-T., Santoni, V., Shahzad, Z. and Verdoucq, Park, S.-Y., Fung, P., Nishimura, N. et al. (2009) Abscisic acid inhibits type L. (2015) Aquaporins in plants. Physiol. Rev. 95, 1321–1358. 2C protein phosphatases via the PYR/PYL family of START proteins. McCurdy, D.W. and Harmon, A.C. (1992) Calcium-dependent protein kinase Science, 324, 1068–1071. in the green alga Chara. Planta, 188,54–61. Perez-Mart ın, M., Perez-P erez, M.E., Lemaire, S.D. and Crespo, J.L. (2014) Meiri, D. and Breiman, A. (2009) Arabidopsis ROF1 (FKBP62) modulates Oxidative stress contributes to autophagy induction in response to endo- thermotolerance by interacting with HSP90.1 and affecting the accumula- plasmic reticulum stress in Chlamydomonas reinhardtii. Plant Physiol. tion of HsfA2-regulated sHSPs. Plant J. 59, 387–399. 166, 997–1008. Mittler, R., Finka, A. and Goloubinoff, P. (2012) How do plants feel the heat? Pham, J., Liu, J., Bennett, M.H., Mansfield, J.W. and Desikan, R. (2012) Ara- Trends Biochem. Sci. 37, 118–125. bidopsis histidine kinase 5 regulates salt sensitivity and resistance Montane, M.H., Dreyer, S., Triantaphylides, C. and Kloppstech, K. (1997) against bacterial and fungal infection. New Phytol. 194, 168–180. Early light-inducible proteins during long-term acclimation of barley to Pichrtova, M., Arc, E., Stoggl,€ W., Kranner, I., Hajek, T., Hackl, H. and Holzin- photooxidative stress caused by light and cold: High level of accumula- ger, A. (2016) Formation of lipid bodies and changes in fatty acid compo- tion by posttranscriptional regulation. Planta, 202, 293–302. sition upon pre-akinete formation in Arctic and Antarctic Zygnema Monte, I., Kneeshaw, S., Franco-Zorrilla, J.M., Chini, A., Zamarreno,~ A.M., (Zygnematophyceae, Streptophyta) strains. FEMS Microbiol. Ecol. 92, Garcıa-Mina, J.M. and Solano, R. (2020) An ancient COI1-independent fiw096. function for reactive electrophilic oxylipins in thermotolerance. Curr. Pino, L.K., Searle, B.C., Bollinger, J.G., Nunn, B., MacLean, B. and MacCoss, Biol. 30, 962–971.e3. M.J. (2017) The Skyline ecosystem: Informatics for quantitative mass

spectrometry proteomics. Mass Spectrom. Rev. 39(3), 229–244. https:// von Schwartzenberg, K., Bornfleth, S., Lindner, A.-C. and Hanelt, D. (2013) doi.org/10.1002/mas.21540. The Microalgae and Zygnematophyceae Collection Hamburg (MZCH) – Puttick, M.N., Morris, J.L., Williams, T.A. et al. (2018) The interrelationships living cultures for research on rare streptophytic algae. Algol. Stud. 142, of land plants and the nature of the ancestral embryophyte. Curr. Biol. 77–107. 28, 733–745. Seok, M.S., You, Y.N., Park, H.J., Lee, S.S., Aigen, F., Luan, S., Ahn, J.C. Quevillon, E., Silventoinen, V., Pillai, S., Harte, N., Mulder, N., Apweiler, R. and Cho, H.S. (2014) AtFKBP16-1, a chloroplast lumenal immunophilin, and Lopez, R. (2005) InterProScan: protein domains identifier. Nucleic mediates response to photosynthetic stress by regulating PsaL stability. Acids Res. 33, W116–W120. Physiol. Plant. 150, 620–631. Ramel, F., Birtic, S., Ginies, C., Soubigou-Taconnat, L., Triantaphylides, C. Shapiro, S.S. and Wilk, M.B. (1965) An analysis of variance test for normal- and Havaux, M. (2012) Carotenoid oxidation products are stress signals ity (complete samples). Biometrika, 52, 591–611. that mediate gene responses to singlet oxygen in plants. Proc. Natl Acad. Sharkey, T.D. (2005) Effects of moderate heat stress on photosynthesis: Sci. USA, 109, 5535–5540. importance of thylakoid reactions, rubisco deactivation, reactive oxygen Ramesh, S.A., Tyerman, S.D. and Xu, B. et al. (2015) GABA signalling modu- species, and thermotolerance provided by isoprene. Plant Cell Environ. lates plant growth by directly regulating the activity of plant-specific 28, 269–277. anion transporters. Nat. Commun. 6, 7879. Shimizu, T., Mochizuki, N., Nagatani, A. et al. (2019) GUN1 regulates tetra- Reddy, A.S.N., Ali, G.S., Celesnik, H. and Day, I.S. (2012) Coping with stres- pyrrole biosynthesis. bioRxiv. https://doi.org/10.1101/532036. ses: roles of calcium- and calcium/calmodulin-regulated gene expres- Soon, F.F., Ng, L.M., Zhou, X.E., et al. (2012) Molecular mimicry regulates sion. Plant Cell, 23, 2010–2032. ABA signaling by SnRK2 kinases and PP2C phosphatases. Science, 335, Rensing, S.A. (2018) Great moments in evolution: the conquest of land by 85–88. plants. Curr. Opin. Plant Biol. 42,49–54. Song, L., Chen, Z. and Larkin, R.M. (2018) The genomes uncoupled mutants Rice, P., Longden, L. and Bleasby, A. (2000) EMBOSS: the European molecu- are more sensitive to norflurazon than wild type. Plant Physiol. 178, 965– lar biology open software suite. Trends Genet. 16, 276–277. 971. Rippin, M., Becker, B. and Holzinger, A. (2017) Enhanced desiccation toler- Stael, S., Wurzinger, B., Mair, A., Mehlmer, N., Vothknecht, U.C. and Teige, ance in mature cultures of the streptophytic green alga Zygnema circum- M. (2011) Plant organellar calcium signalling: an emerging field. J. Exp. carinatum revealed by transcriptomics. Plant Cell Physiol. 58, 2067–2084. Bot. 63, 1525–1542. Rippin, M., Pichrtova, M., Arc, E., Kranner, I., Becker, B. and Holzinger, A. Su, Z., Tang, Y., Ritchey, L.E., Tack, D.C., Zhu, M., Bevilacqua, P.C. and Ass- (2019) Metatranscriptomic and metabolite profiling reveals vertical mann, S.M. (2018) Genome-wide RNA structurome reprogramming by heterogeneity within a Zygnema green algal mat from Svalbard (High acute heat shock globally regulates mRNA abundance. Proc. Natl Acad. Arctic). Environ. Microbiol. 21, 4283–4299. Sci. USA, 115, 12170–12175. Rizhsky, L., Liang, H., Shuman, J., Shulaev, V., Davletova, S. and Mittler, R. Sugliani, M., Abdelkefi, H., Ke, H., Bouveret, E., Robaglia, C., Caffarri, S. and (2004) When defense pathways collide. The response of Arabidopsis to a Field, B. (2016) An ancient bacterial signaling pathway regulates chloro- combination of drought and heat stress. Plant Physiol. 134, 1683–1696. plast function to influence growth and development in Arabidopsis. Plant Robinson, M.D. and Oshlack, A. (2010) A scaling normalization method for Cell, 28, 661–679. differential expression analysis of RNA-seq data. Genome Biol. 11, R25. Sun, A.-Z. and Guo, F.-Q. (2016) Chloroplast retrograde regulation of heat Robinson, M.D., McCarthy, D.J. and Smyth, G.K. (2010) edgeR: A biocon- stress responses in plants. Front. Plant Sci. 7, 398. ductor package for differential expression analysis of digital gene expres- Sun, Y., Harpazi, B., Wijerathna-Yapa, A., Merilo, E., de Vries, J., Michaeli, sion data. Bioinformatics, 26, 139–140. D., Gal, M., Cuming, A., Kollist, H. and Mosquna, A. (2019) A ligand-inde- Rolland, F., Baena-Gonzales, E. and Sheen, J. (2006) Sugar sensing and sig- pendent origin of abscisic acid perception. Proc: Natl Acad. Sci USA, 116 naling in plants: conserved and novel mechanisms. Annu. Rev. Plant (49), 24892–24899. https://doi.org/10.1073/pnas.1914480116. Biol. 57, 675–709. Sussmilch, F.C., Roelfsema, M.R.G. and Hedrich, R. (2019) On the origins of Rossel, J.B., Wilson, I.W. and Pogson, B.J. (2002) Global changes in gene osmotically driven stomatal movements. New Phytol. 331, 582–587. expression in response to high light in Arabidopsis. Plant Physiol. 130, Takahashi, K., Kasai, K. and Ochi, K. (2004) Identification of the bacterial 1109–1120. alarmone guanosine 50-diphosphate 30-diphosphate (ppGpp) in plants. RStudio Team. (2016) RStudio: Integrated Development for R. Boston, MA: Proc. Natl Acad. Sci. USA, 101, 4320–4324. RStudio, Inc. http://www.rstudio.com/ Takahashi, K., Takabayashi, A., Tanaka, A. and Tanaka, R. (2014) Func- Rubio, A., Rambla, J.L., Santaella, M., Gomez, M.D., Orzaez, D., Granell, A. tional analysis of light-harvesting-like protein 3 (LIL3) and its light-har- and Gomez-G omez, L. (2008) Cytosolic and plastoglobule-targeted caro- vesting chlorophyll-binding motif in Arabidopsis. J. Biol. Chem. 289, tenoid dioxygenases from Crocus sativus are both involved in b-ionone 987–999. release. J. Biol. Chem. 283, 24816–24825. Tanaka, R., Rothbart, M., Oka, S. et al. (2010) LIL3, a light-harvesting-like Rubio, S., Rodrigues, A., Saez, A. et al. (2009) Triple loss of function of pro- protein, plays an essential role in chlorophyll and tocopherol biosynthe- tein phosphatases type 2C leads to partial constitutive response to sis. Proc. Natl Acad. Sci. USA, 107, 16721–16725. endogenous abscisic acid. Plant Physiol. 150, 1345–1355. Terashima, M., Specht, M. and Hippler, M. (2011) The chloroplast proteome: Ruhfel, B.R., Gitzendanner, M.A., Soltis, P.S., Soltis, D.E. and Burleigh, J.G. a survey from the Chlamydomonas reinhardtii perspective with a focus (2014) From algae to angiosperms-inferring the phylogeny of green on distinctive features. Curr. Genet. 57, 151–168. plants (Viridiplantae) from 360 plastid genomes. BMC Evol. Biol. 14, 23. Tewari, A.K. and Tripathy, B.C. (1998) Temperature-stress-induced impair- Saidi, Y., Finka, A., Muriset, M., Bromberg, Z., Weiss, Y.G., Maathuis, F.J.M. ment of chlorophyll biosynthetic reactions in cucumber and wheat. Plant and Goloubinoff, P. (2009) The heat shock response in moss plants is Physiol. 117, 851–858. regulated by specific calcium-permeable channels in the plasma mem- Timmis, J.N., Ayliffe, M.A., Huang, C.Y. and Martin, W. (2004) Endosymbi- brane. Plant Cell, 21, 2829–2843. otic gene transfer: organelle genomes forge eukaryotic chromosomes. Scheres, B. and van der Putten, W.H. (2017) The plant perceptron connects Nat. Rev. Genet. 5, 123–135. environment to development. Nature, 543, 337–345. Tran, L.S.P., Urao, T., Qin, F., Maruyama, K., Kakimoto, T., Shinozaki, K. Schindelin, J., Arganda-Carreras, I., Frise, E. et al. (2012) Fiji: an open-source and Yamaguchi-Shinozaki, K. (2007) Functional analysis of AHK1/ATHK1 platform for biological-image analysis. Nat. Methods, 9, 676–682. and cytokinin receptor histidine kinases in response to abscisic acid, Schmitz-Linneweber, C. and Small, I. (2008) Pentatricopeptide repeat pro- drought, and salt stress in Arabidopsis. Proc. Natl Acad. Sci. USA, 104, teins: a socket set for organelle gene expression. Trends Plant Sci. 13, 20623–20628. 663–670. Turmel, M., Otis, C. and Lemieux, C. (2006) The chloroplast genome Schroda, M., Hemme, D. and Muhlhaus,€ T. (2015) The Chlamydomonas heat sequence of Chara vulgaris sheds new light into the closest green algal stress response. Plant J. 82, 466–480. relatives of land plants. Mol. Biol. Evol. 23, 1324–1338. Schulz, P., Herde, M. and Romeis, T. (2013) Calcium-dependent protein Umezawa, T., Nakashima, K., Miyakawa, T. et al. (2010) Molecular basis of kinases: hubs in plant stress signaling and development. Plant Physiol. the core regulatory network in ABA responses: sensing, signaling and 163, 523–530. transport. Plant Cell Physiol. 51, 1821–1839.

Umezawa, T., Sugiyama, N., Mizoguchi, M., Hayashi, S., Myouga, F., Yam- Thellungiella, a close relative of Arabidopsis. Plant Physiol. 140, 1437– aguchi-Shinozaki, K., Ishihama, Y., Hirayama, T. and Shinozakia, K. 1450. (2009) Type 2C protein phosphatases directly regulate abscisic acid-acti- Wu, G.-Z., Chalvin, C., Hoelscher, M., Meyer, E.H., Wu, X.N. and Bock, R. vated protein kinases in Arabidopsis. Proc. Natl Acad. Sci. USA, 41, (2018) Control of retrograde signaling by rapid turnover of GENOMES 17588–17593. UNCOUPLED1. Plant Physiol. 176, 2472–2495. Vlad, F., Rubio, S., Rodrigues, A., Sirichandra, C., Belin, C., Robert, N., Xiao, Y., Savchenko, T., Baidoo, E.E.K. et al. (2012) Retrograde signaling by Leung, J., Rodriguez, P.L., Lauriere, C. and Merlot, S. (2009) Protein the plastidial metabolite MEcPP regulates expression of nuclear stress- phosphatases 2C regulate the activation of the Snf1-related kinase OST1 response genes. Cell, 149, 1525–1535. by abscisic acid in Arabidopsis. Plant Cell, 21, 3170–3184. Xu, X.Z., Wes, P.D., Chen, H., Li, H.S., Yu, M., Morgan, S., Liu, Y. and Mon- Vogel, J.T., Tan, B.-C., McCarthy, D.R. and Klee, H.J. (2008) The Carotenoid tell, C. (1998) Retinal targets for calmodulin include proteins implicated Cleavage Dioxygenase 1 enzyme has broad substrate specificity, cleaving in synaptic transmission. J. Biol. Chem. 273, 31297–31307. multiple carotenoids at two different bond positions. J. Biol. Chem. 283, Yuan, M., Breitkopf, S.B., Yang, X. and Asara, J.M. (2012) A positive/nega- 11364–11373. tive ion-switching, targeted mass spectrometry-based metabolomics Vogt, T. (2010) Phenylpropanoid biosynthesis. Mol. Plant, 3,2–20. platform for bodily fluids, cells, and fresh and fixed tissue. Nat. Protoc. 7, Wang, L., Guo, Y., Jia, L. et al. (2014) Hydrogen peroxide acts upstream of 872–881. nitric oxide in the heat shock pathway in Arabidopsis seedlings. Plant Zhao, X., Huang, J. and Chory, J. (2018) genome uncoupled1 mutants are Physiol. 164, 2184–2196. hypersensitive to norflurazon and lincomycin. Plant Physiol. 178, 960– Wickett, N.J., Mirarab, S., Nguyen, N. et al. (2014) Phylotranscriptomic anal- 964. ysis of the origin and early diversification of land plants. Proc. Natl Acad. Zhao, C., Wang, Y., Chan, K.X. et al. (2019) Evolution of chloroplast retro- Sci. USA, 111, E4859–E4868. grade signaling facilitates green plant adaptation to land. Proc. Natl Wodniok, S., Brinkmann, H., Glockner,€ G., Heidel, A.J., Philippe, H., Melko- Acad. Sci. USA, 116, 5015–5020. nian, M. and Becker, B. (2011) Origin of land plants: do conjugating Zhu, J.-K. (2016) Abiotic stress signaling and responses in plants. Cell, 167, green algae hold the key? BMC Evol. Biol. 11, 104. 313–324. Wong, C.E., Li, Y., Labbe, A. et al. (2006) Transcriptional profiling implicates novel interactions between abiotic stress and hormonal responses in