Phytochrome A is required for light-inhibited germination of Aethionema arabicum seed

23 Phytochrome, seed germination, light response, Aethionema arabicum , photoblastic 24 phenotype, adaptation 25 26 ORCIDs 27 Zsuzsanna Merai ORCID 0000-0002-2048-1628 28 Fei Xu ORCID 0009-0005-1482-4435 29 Anita Hajdu ORCID 0000-0002-0837-7529 30 László Kozma-Bognár ORCID 0000-0002-8289-193X 31 Liam Dolan ORCID 0000-0003-1206-7096 32 33 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 2 Summary 34 • The germination of most seeds is influenced by the duration, intensity, and quality of light. 35 The seeds of the model plant Arabidopsis are positive photoblastic and require light to 36 germinate. The germination of negative photoblastic seeds is inhibited by white light. The 37 molecular mechanisms that regulate negative photoblastic germination are unknown due to 38 the lack of a suitable model plant. 39 • We identified an accession with negative photoblastic germination in Aethionema arabicum 40 that grows in semi-arid natural habitats. In a forward genetic screen, we identified a mutant – 41 koyash2 (koy2) – that is defective in negative photoblastic germination. There is a nonsense 42 mutation in the gene encoding the phytochrome A photoreceptor in the koy2 mutant. 43 • Here we show that phytochrome A is r equired for negative photoblastic germination. The 44 defective negative photoblastic phenotype of the koy2 mutant is the result of defective 45 inhibition of germination by the phytochrome A mediated high-irradiance response. 46 • This is the first example of phytochro me A-mediated response controlling negative 47 photoblastic seed germination in white, red, far-red, and blue light. We speculate that 48 genetically encoded variation in phytochrome A-mediated germination responses is 49 responsible for local adaptation of Ae. arabicum throughout the Irano-Turanian region. 50 51

52 Natural selection acts on genetic variation that arises in wild populations. This genetic 53 variation can control the development of diverse life history traits that may be adaptive. 54 Aethionema arabicum (Brassicaceae) is an annual plant that grows in the Mediterranean 55 ecosystems of Southeastern Europe and the Irano-Turanian region. There is considerable 56 variation in life history traits among Aethionema species and different Ae. arabicum 57 accessions isolated from different parts of its natural range. For example, life cycles may be 58 annual or perennial (Mohammadin et al., 2017). Fruits may be dimorphic and the ratio 59 between the two morphs varies between closely related Aethionema species and between 60 different Ae. arabicum accessions (Lenser et al., 2016; Mohammadin et al., 2017; Arshad et 61 al., 2019; Mérai et al., 2023). The dimorphic fruit morph ratio is modulated by environmental 62 factors and impacts patterns of seed dispersal (Lenser et al., 2018). These variations in life 63 history are genetically controlled and therefore likely adaptations to local environments. 64 Seed germination characteristics impact the life history of plants, and while they differ 65 between species, there is also variation within species. White light modulates seed 66 germination; positively photoblastic seeds require a light pulse for germination, while white 67 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 3 light represses germination in negatively photoblastic seeds. Light-neutral seeds germinate 68 in both darkness and light. Photoblastic responses are white light-specific and do not include 69 wavelength-specific seed germination responses (Takaki, 2001). An Ae. arabicum accession 70 from Turkey germinates in all light conditions (light-neutral) upon imbibition (Mérai et al., 71 2019). However, other accessions of Ae. arabicum are negatively photoblastic and darkness 72 stimulates germination, while white light represses germination (Mérai et al., 2019). This 73 variation may reflect adaptations to different environments. The negatively photoblastic trait 74 is likely adaptive and could hypothetically promote germination in conditions where the seed 75 is not directly exposed to light. Seedlings would be less susceptible to the desiccation 76 caused by direct exposure to the sun during early stages of establishment when root growth 77 occurs but before photosynthesis is necessary. The germination of negatively photoblastic 78 seeds may be modulated by day length (Mérai et al., 2019, 2023, 2024). For example, the 79 seeds of a negatively photoblastic accession from Cyprus (CYP) do not germinate in white 80 light in long day conditions. However, seeds of the CYP accession germinate in white light in 81 short day conditions. This promotes the germination and seedling establishment during the 82 spring season when conditions are optimal for germination and seedling establishment, 83 whereas it inhibits germination during the summer period when the conditions are deleterious 84 for seedling establishment (Mérai et al., 2023, 2024). This indicates that some negatively 85 photoblastic seeds can sense differences in day length to repress germination in one 86 condition and promote germination in another. 87 The mechanism that promotes seed germination by light– positive photoblasty – is well 88 characterized in a variety of species including Arabidopsis thaliana (Brassicaceae). In A. 89 thaliana, light stimulates the production of gibberellic acid (GA) and reduces the level of the 90 germination repressor, abscisic acid (ABA), in seed. Light is perceived in seed by the 91 phytochrome photoreceptors and increases the expression of genes encoding enzymes that 92 promote GA synthesis (GA3OX) and represses the expression of genes encoding enzymes 93 that metabolize GA (GA2OX) (Seo et al., 2006). Light also reduces the expression of genes 94 encoding proteins involved in ABA synthesis (NCED5, NCED6). Light signalling modulates 95 the balance between the germination-promoting activity of GA and the germination-96 repressing activity of ABA (Seo et al., 2009). 97 Phytochromes are red and far-red absorbing photoreceptors in plants that exist in 98 interconvertible forms; the red-light absorbing Pr (inactive) and the far-red light absorbing Pfr 99 (active) form. Multiple genes encode phytochromes in most angiosperms. For example, there 100 are five different phytochrome proteins – phyA, phyB, phyC, phyD, phyE – in A. thaliana. 101 phyA and phyB are the dominant photoreceptors that control seed germination in A. thaliana. 102 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 4 However, phyE is also involved and its function can be observed in phy1 phyb double 103 mutants (Hennig et al., 2002). 104 While the absorption spectra of all A. thaliana phytochromes are identical, the specificity in 105 light responses can be explained by the difference in the kinetics of the interconversion 106 between the Pr and Pfr forms, cellular localization, and specific molecular interactions. The 107 active Pfr form of phyA is light-labile and imported into the nucleus upon illumination and 108 subsequently degraded (Somers et al., 1991). PhyB-E on the other hand are light-stable. The 109 physiological responses are initiated by phyA, in low levels of Pfr that occur when plants are 110 grown in high levels of far-red light or in low intensities of white light. Under these conditions, 111 the active nuclear pool of phyA can be maintained by the constant nuclear import of a small 112 Pfr amount that replaces the degraded fraction (Casal, Candia and Sellaro, 2014). 113 Phytochrome acts in discrete responses depending on the fluence level and wavelength of 114 light (Casal, Sánchez and Botto, 1998). The very low fluence response (VLFR) regulates 115 germination by the phy A photoreceptor (Shinomura et al., 1996). VLFR is active when the 116 Pr:Pfr ratio is high and the relative levels of Pfr are very low. For example, VLFR-activated 117 germination is induced by red light between 1-100 nmol m-2 total fluence or low intensity far-118 red light between 0.5 and 10 µmol m-2 total fluence in which as little as 0.01% of 119 phytochrome is in the active Pfr form (Shinomura et al., 1996). The low fluence response 120 (LFR) photoreversibly regulates germination at light intensities between 10-1000 µmol m-2 121 total fluence through the phyB photoreceptor; red light induces, and far-red light inhibits 122 germination (Shinomura et al., 1996). The high intensity response (HIR) regulates hypocotyl 123 and stem elongation and anthocyanin production but does not regulate seed germination of 124 A. thaliana (Casal and Sánchez, 1998). The high irradiance response requires circa 60 µmol 125 m-2 s-1 I and is mediated by phyA (Shinomura, Uchida and Furuya, 2000). Germination of Ae. 126 arabicum seed is promoted by the VLFR and inhibited by the HIR but it is unknown if the low 127 fluence response (LFR) also controls germination (Mérai et al., 2023). Identifying which 128 phytochrome is involved in conferring germination and through which mechanisms it 129 operates is an outstanding question in Ae. arabicum seed biology. 130 Phytochrome A and B promote germination in positively photoblastic A. thaliana seed 131 (Shinomura et al., 1994; Lee et al., 2012). Less is known about the mechanism of light-132 repressed seed germination in negatively photoblastic plants such as the negatively 133 photoblastic accessions of Ae. arabicum. Although it is known that phytochrome is required 134 for negative photoblasty, the identity of the phytochrome(s) that is active during negative 135 photoblasty in Ae. Aethionema is unknown (Mérai et al., 2023). To identify genes required for 136 negatively photoblastic seed germination in Ae. arabicum, we carried out a genetic screen in 137 the Cyprus (CYP) accession that is negatively photoblastic: it germinates only in the dark but 138 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 5 does not germinate in the light. We screened for mutants that germinated in the light, 139 reasoning that the genes required for light-repressed germination would be defective in these 140 mutants. We identified Koyash (koy2) mutants that germinated in the light and carried a 141 mutation in the phytochrome A gene. We show that phyA (KOY2) is required for light-induced 142 repression of germination in this negatively photoblastic accession. However, phyA activity is 143 required for two antagonistic responses in Ae. arabicum (CYP) seed. It promotes light-144 induced germination through VLFR while it represses germination in red and far-red HIR. 145 These data indicate that the same photoreceptor is active in promoting or repressing 146 germination in Ae. arabicum through the VLFR and HIR, respectively. These data also 147 suggest that the divergence of negative photoblasty from a positive photoblastic state in A. 148 arabicum involved regulatory changes downstream of the photoreceptor. 149 150

151 152 Plant and seed material 153 Experiments were conducted with Aethionema arabicum (L.) Andrz. ex DC. accessions CYP 154 (obtained from Eric Schranz, Wageningen). Wild type and koy2 plants were propagated for 155 seed material under 16 h light/19°C and 8 h dark/16°C diurnal cycles, under ~300 μ mol m-2 s-156 1 light intensity. Wild type and koy2 plants were randomly distributed on the shelves. Seeds 157 were harvested upon full maturation and stored in darkness at 50% humidity and 24°C for at 158 least three months, except for the experiments in Figure 4B and 4C where semi-dormant 159 seed batches eight weeks after harvest were used. 160 161 Plant chambers and light source 162 The mutant screen and white light germination assays were carried out in a Percival plant 163 growth chamber equipped with fluorescent white light tubes (Philips). Red, far-red, and blue 164 light applications were performed using customer-designed LED light sources, using LEDs 165 from OptoSupply (www.optosupply.cn). The spectral properties and the LED types are as in 166 (Mérai et al., 2023). Light spectra and intensity were measured by LED meter MK350S 167 (UPRtek). 168 169 Aethionema arabicum genome and annotations 170 Aethionema arabicum genome version 3.0, gene models, cDNA, and protein annotations 171 (version 3.1) were obtained from Ae. arabicum database (https://plantcode.cup.uni-172 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 6 freiburg.de/aetar_db/) (Fernandez-Pozo et al., 2021). Gene accession numbers used in this 173 study are listed in Supplementary Table S1. 174 175 Forward genetic screening 176 For genetic screening, mutant seed bank and screening conditions were used as described 177 in (Mérai et al., 2023). 20-30 seeds each from M2 or M3 seed batches were tested from 178 approximately 1320 lines originating from independent mutagenized seeds. Seed 179 germination assays were carried out in Petri dishes with wet filter paper at 14° C in a Percival 180 growth chamber. Seeds were kept under continuous 160 µmol m-2 s-1 white light and scored 181 for germination after 7 days. The germinating seeds were kept and propagated. One 182 selected line with a stable mutant phenotype in two following generation was named as koy2. 183 The koy2 mutant line was backcrossed with the wild type and the segregation of the 184 germination phenotype was determined in the progeny. As the original mutant line is 185 hyposensitive and germinates with ~80% at 160 µmol m-2 s-1, in contrast to 0% of the wild 186 type, we expected the germination rate in the F2 backcross generation in case of a recessive 187 mutation to be lower than 25%. The observed germination of F2 seed population under light 188 was ~17% and matched this expectation. Given the long hypocotyl phenotype in the koy2 189 mutant under far-red light, the gene encoding PHYA (Aa31LG1G5460) was amplified with 190 primers 5`ATGTCAGGAGCTAGGCCGAGTC and 5`TTTGTTTGCAGCTGCAAGTTCAG and 191 sequenced throughout its entire full length. For genotyping the koy2 mutant, the gene region 192 was amplified using primers 5`TATAAGTTAGCTGCTAAAGCGA and 193 5`AGATGATCTTGGATGCATTCTTC, followed by Sanger sequencing of the amplicons. 194 Germination assays 195 Germination assays were carried out in Petri dishes on wet filter paper using distilled water 196 supplemented with 0.1 v/v % PPMTM (Plant Preservative Media, 5-Chloro-2-methyl-3(2H)-197 isothiazolone, Plant Cell Technology). 20-30 seeds were used for one replicate and three 198 biological replicates were used from different seed batches. Germination in darkness was 199 tested by transferring seeds onto wet plates in complete darkness without green safety light 200 and subsequent wrapping the plates in two layers of aluminum foil and placing them in a dark 201 box. All germination assays were done at 14° C and scored after 7 days. For light pulse 202 experiments, 100 µmol m-2 s-1 red light and 10 µmol m-2 s-1 far-red light was used for 10 min 203 (Figure 4B and C). 204 Hypocotyl assay 205 Hypocotyl assays were carried out as described in (Mérai et al., 2023). 206 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 7 RNA extraction and quantitative RT-PCR 207 To obtain RNA samples, wild type and koy2 seeds were exposed to continuous light 208 exposure for 24 h or kept in darkness. 80 µmol m-2 s-1 red light, 1 µmol m-2 s-1 far-red light, or 209 100 µmol m-2 s-1 blue light was used. Only seeds with intact seed coat were collected under 210 safety green light. RNA extraction, cDNA synthesis, and RT-qPCR were performed as 211 described (Mérai et al., 2019, 2023), using the primer pairs listed in Supplemental Table S1. 212 The geometric mean of Ae. arabicum putative orthologues of POLYUBIQUITIN10 213 (AearUBQ10, Aa3LG9G835) and ANAPHASE-PROMOTING COMPLEX2 (AearAPC2, 214 Aa31LG10G13720) was used for normalization. For each gene, the expression levels were 215 presented as fold change relative to the level of the dark samples in wild type seeds, where 216 the average expression was set to one. Statistical analysis was done using the SATQPCR 217 tool (Rancurel et al., 2019). Error bars represent standard deviation of three biological 218 replicates. Asterisks indicate significant differences from the wild type or koy2 dark level with 219 P-values as *P < 0.05, **P < 0.01, and ***P < 0.001 calculated with the Tukey test. 220 Western blotting 221 To obtain protein samples, wild type and koy2 seeds were germinated in darkness at 14° C 222 for 6 days. Seedlings were transferred under red (100 µmol m-2 s-1), far-red (1 µmol m-2 s-1), 223 or blue (100 µmol m-2 s-1) illumination for 24 h or kept in darkness. For positive control 224 samples, A. thaliana wild type Wassilewskija ecotype seeds were exposed to 4 h light to 225 induce germination, followed by 6 days darkness at 22° C. Total protein extraction and 226 Western blot analysis were done as described (Kevei et al., 2007). 50 µg protein per lane 227 was analyzed. For the detection of phyA proteins, a primary polyclonal rabbit antibody (Wolf 228 et al., 2011), 1:2400 dilution) and a horse radish peroxidase(HRP)-conjugated secondary 229 anti-rabbit antibody (Dako, #: P039901-2, 1:3000 dilution) wwere used. Actin levels were 230 determined using a monoclonal anti-actin primary antibody (Sigma, #: A0480, clone 10-B3, 231 1:10000 dilution) and an HRP-conjugated secondary anti-mouse antibody (Thermo Fisher 232 Scientific, #: 31431, 1:10000 dilution). Chemiluminescent signals were detected and 233 quantified as described (Hajdu et al., 2018). 234

235 koy2 mutant seeds are hyposensitive for light-inhibited germination 236 Seed of the Cyprus accession of Aethionema arabicum germinate in darkness, while 237 increasing light intensity gradually leads to complete inhibition of germination (Mérai et al., 238 2019, 2023). We performed a forward genetic screen to identify mutants that germinate 239 despite light exposure, conditions where wild type seeds are fully inhibited (Mérai et al., 240 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 8 2023). 20-30 seeds of each mutant line generated by γ-radiation were plated under 120-241 160 µmol m-2 s-1 continuous white light and germination was scored after 6 days. Mutated 242 lines with germinating seeds were kept and propagated for two generations to validate the 243 stable inheritance of the phenotype. 100% of the M3 seed generation of one line germinated 244 at 96 µmol m-2 s-1 compared to 12.5% germination of the wild type (Figure 1A, B). Exposure 245 to higher light intensities (166 and 213 µmol m-2 s-1) resulted in complete inhibition of 246 germination in wild type. However, 71.6% and 8% of the mutant seed germinated in 166 and 247 213 µmol m-2 s-1 light, respectively (Figure 1A). We concluded that seeds of this line are 248 hyposensitive for the inhibition of germination by light. The line was named as koyash2 249 (koy2) after the god of sun in Turkic mythology. 250 The hyposensitive phenotype of koy2 seeds suggested that the mutation affects either the 251 light perception or the phytohormone metabolism. To discriminate between these 252 alternatives, we compared hypocotyl elongation under monochromatic red, far-red, and blue 253 light in wild type and koy2 mutants. We hypothesized that if the mutation affects light 254 perception, the hypocotyl shortening response of the mutant seedlings will significantly differ 255 from the wild type shortening response. Seeds were plated and kept in darkness for two days 256 to induce germination. Germinating seeds were then illuminated with increasing light 257 intensities of red, far-red, and blue light, for five days. The hypocotyl shortening response 258 was similar in the koy2 mutant and the wild type under red and blue light; both koy2 and wild 259 type elongated less in increasing light (Figure 1C). This indicates that hypocotyl elongation in 260 red and blue light is not defective in the koy2 mutant. However, the length of koy2 seedlings 261 in far-red light was the same as in darkness, indicating that the koy2 mutant is non-262 responsive to far-red light (Figure 1C). 263 In angiosperms, phytochrome A is the far-red photoreceptor; complete loss-of-function 264 mutants of Arabidopsis PHYA develops long hypocotyls under far-red light. Therefore, we 265 hypothesized that koy2 is a phya mutant in Aethionema. The sequencing of PHYA cDNA in 266 koy2 revealed a single C to A nucleotide change at the position of 822 nucleotides, resulting 267 in a predicted premature termination codon after amino acid 273 (Figure 1D). Seedlings with 268 long hypocotyls grown under far-red light were homozygous for the koy2 mutant allele while 269 seedlings with short hypocotyls were heterozygotes or carried the wild-type PHYA allele 270 (Supplementary Table S2). The predicted truncated phyA protein lacks the chromophore 271 attachment site that is essential for light perception. Therefore, we conclude that koy2 is a 272 complete loss of function phya mutant allele (Figure 1D). To test this hypothesis, we 273 measured the steady state levels of CHALCONE SYNTHASE (CHS) mRNA that 274 accumulates at higher steady state levels in seeds and seedlings upon red, blue and far-red 275 l ight, than in the dark (Mérai et al., 2023). In wild-type seeds, CHS mRNA levels were higher 276 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 9 in red, far-red, and blue light than in dark-grown controls. However, steady state CHS mRNA 277 levels in koy2 seeds were identical in far-red light compared to those kept in the dark. 278 Furthermore, the steady state CHS mRNA levels in koy2 seeds were significantly higher in 279 red and blue light than in darkness, as in wild type. These data indicate that the induction of 280 CHS expression by far-red light is defective in koy2 while CHS induction in red and blue light 281 is independent of KOY2 function. These data are consistent with the hypothesis that 282 KOY2/phyA are required for the far-red responses in Ae. arabicum. 283 Light reduces steady state levels of PHYA mRNA and promotes phyA protein degradation by 284 the 26S proteosome (Somers et al., 1991; Cantón and Quail, 1999). To test if the 285 transcriptional and post-translational repression of PhyA also occurs in the wild type and the 286 koy2 mutant of Ae. arabicum, we measured levels of PhyA mRNA and protein. Steady state 287 levels of PHYA mRNA level were highest in seeds incubated in darkness and significantly 288 lower in red, blue, or far-red light-illuminated wild-type seeds (Figure 1F). The mRNA level of 289 koy2 mutant PHYA in the darkness was 15% the levels of wild type (Figure 1F). 290 PHYA mRNA levels were lower in red light in koy2 than in darkness, indicating that the phyB-291 mediated repression of PHYA gene expression is functional in the koy2 seed (Figure 1F). 292 The PHYA mRNA level did not change significantly in far-red light in koy2 seed, further 293 confirming that koy2 is non-responsive to far-red light. The PHYA mRNA level was 294 unchanged in koy2 in blue light, suggesting the autorepression of PHYA expression under 295 blue light (Figure 1F). 296 phyA protein is a light-labile photoreceptor quickly degraded in all light conditions in A. 297 thaliana. We set out to test if phyA is light-labile in Ae. arabicum. phyA protein was 298 detectable in dark-grown wild type Ae. arabicum seedlings. However, the protein was 299 undetectable after 4 days of constant red or blue illumination and less abundant under far-300 red light than in dar kness (Figure 1G). These data demonstrate that phyA protein is 301 degraded upon exposure to light. Furthermore, phyA protein was not detected in the koy2 302 mutant seedlings in darkness, indicating that phytochrome protein does not accumulate in 303 the koy2 mutant (Figure 1G). 304 Taken together, these data indicate that the koy2 mutant is a phya complete loss of function 305 mutant that is hyposensitive to the light-inhibited germination of Ae. arabicum seed. 306 307 Loss-of-function phyA (koy2) mutant seeds germinate in red, far-red, and blue light 308 Red, blue, and far-red light inhibit germination of wild type Ae. arabicum in an intensity-309 dependent manner; higher light intensities lead to progressive reduction in germination 310 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 10 (Mérai et al., 2019, 2023). To determine the wavelength-specific action of phyA in 311 germination inhibition, the germination of wild type and koy2 seed germination was 312 measured under increasing intensities of red, far-red, and blue light illumination. 100% of 313 koy2 seeds germinated under 300 µmol m2 s1 red light, while no wild type seed germinated at 314 162 µmol m2 s1 and above (Figure 2B). These data are consistent with the hypothesis that 315 red light repression of germination requires phyA (KOY2) activity. Over 95% of koy2 seeds 316 germinated in far-red light, while wild type did not germinate, even at very low light intensities 317 (0.03 µmol m2 s1) (Figure 2A), consistent with the hypothesis that far-red light repression of 318 germination requires phyA (KOY2) activity. koy2 seed germination was hyposensitive for 319 blue light; 72% of koy2 mutant seed germinated in blue light while 0% of wild type seed 320 germinated at 122 µmol m2 s1 (Figure 2C), consistent with the hypothesis that blue light 321 repression of germination requires phyA (KOY2) activity. These results indicate that phyA 322 mediates the inhibition of seed germination by red, far-red, and blue light in Ae. arabicum. 323 324 phyA (KOY2) promotes the expression of genes encoding proteins that reduce GA 325 levels and increase ABA levels in seeds in red, far-red, and blue light 326 Red light 327 Red light represses Ae. arabicum seed germination (Mérai et al., 2019, 2023). Phytochromes 328 regulate germination through a signalling cascade that leads to the transcriptional regulation 329 of the genes encoding enzymes of GA and ABA metabolism (Seo et al., 2006). To test the 330 role of phyA (KOY2) in GA metabolism during the repression of germination by red light, we 331 compared the steady state level of mRNAs for key enzymes for GA synthesis (AearGA3ox1) 332 and GA degradation (AearGA2ox3). Red light-exposure of wild-type plants increased the 333 steady state levels of AearGA2ox3 50-fold compared to dark-treated wild type. Since 334 AearGA2OX3 catabolizes GA, the higher levels of AearGA2ox3 mRNA levels likely reduce 335 GA levels and contribute to the repression of germination by light (Figure 3). The light-336 induced increase in steady state levels of AearGA2ox3 mRNA was much lower in koy2 337 mutants than in wild type seed; red light increased steady state levels 4.8 times in koy2 338 compared to 50 times in the wild type. These data demonstrate that phyA (KOY2) is required 339 for the increase in expression of AearGA2ox3 during red light-mediated repression of 340 germination. To test if phyA (KOY2)-mediated red light signalling represses AearGA3ox1 341 expression, the steady state levels of AearGA3ox1mRNAs were compared in wild type and 342 koy2 mutants. In wild type, steady state levels of AearGA3ox1 mRNA were lower in red light-343 treated plants than in darkness, consistent with the hypothesis that red light represses 344 germination through repression of GA synthesis. Furthermore, steady state levels of 345 AearGA3ox1 mRNA were higher in koy2 mutants than in wild type. These data suggest that 346 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 11 red light repression of AearGA3ox1 is phytochrome dependent. Taken together, these data 347 suggest that phyA (KOY2)-mediated red light signalling promotes the expression of 348 AearGA2OX3, which catabolises GA and represses the expression of AearGA3OX1, 349 required for GA synthesis, during red light-induced repression of germination in A. arabicum 350 seed. 351 ABA represses seed germination (Ali et al., 2022). To test the role of phyA (KOY2) in ABA-352 mediated repression of germination, we measured the steady state levels of mRNAs for 353 proteins involved in ABA synthesis, AearNCED5, AearNCED6, and in ABA degradation 354 (AearCYP707A2) in red light. AearNCED5 and AearCYP707A2 mRNA levels were higher in 355 red light than in darkness, while AearNCED6 level was lower in red light compared to 356 darkness. This higher level of AearNCED5 in red light-treated seeds is consistent with the 357 hypothesis that red light stimulates the production of ABA. Furthermore, AearNCED5, 358 AearNCED6, and AearCYP707A2 mRNA levels were almost identical in wild type and koy2 359 under red light. These data indicate that the levels of genes encoding enzymes for ABA 360 metabolism do not require phyA (KOY2) activity. 361 The data suggest that phyA (KOY2) represses GA synthesis and induces GA degradation 362 during red light-mediated repression of germination. 363 Far-red light 364 Far-red light represses Ae. arabicum germination (Mérai et al., 2019, 2023). To determine 365 the effect of far-red light on the expression of genes that promote germination (AearGA3ox1) 366 and repress germination (AearGA2ox3, AearNCED5, and AearNCED6), we compared the 367 corresponding steady state mRNA levels in seed grown in far-red light and darkness. In wild 368 type, steady state levels of AearGA3ox1 were lower in far-red light than in darkness while 369 steady state levels of AearGA2ox3, AearNCED5, and AearNCED6 were higher in far-red 370 light than in darkness (Figure 3). These data suggest that far-red light represses GA 3 71 synthesis and promotes GA degradation and ABA synthesis (Merai et al., 2023). However, 372 the steady state levels of AearGA3ox1, AearGA2ox3, AearNCED5, and AearNCED6 mRNAs 373 in far-red light and darkness in the koy2 mutant were indistinguishable, indicating that phyA 374 (KOY2) is required for their expressional changes in far-red light. These data indicate that 375 phyA is required for the repression of AearGA3ox1 and induction of AearGA2ox3, 376 AearNCED5, and AearNCED6 in far-red light. Together this would reduce GA levels and 377 increase ABA levels in far-red light compared to controls in the wild-type seeds. 378 Blue light 379 Blue light represses Ae. arabicum germination (Mérai et al., 2019, 2023). To test the function 380 of phyA in blue light repression of germination, steady state mRNA levels of AearGA3ox1, 381 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 12 AearGA2ox3, AearNCED5, AearNCED6, and AearCYP707A2 were measured in wild type 382 and koy2 seed grown in 100 µmol m2 s1 blue illumination in which no wild type seed 383 germinated and approximately 80% of koy2 seeds germinated. Steady state levels of 384 AearGA3ox1, AearGA2ox3, AearNCED5, and AearCYP707A2 were higher in wild-type 385 plants grown in blue light than in the dark. In koy2 seed, the difference in mRNA levels of 386 AearGA2ox3 and AearNCED5 mRNAs between blue light-treated and dark-grown plants was 387 much less than in the wild type. For example, in wild type, AearGA2ox3 mRNA levels were 388 82-fold higher in blue light than in darkness but only 2.96-fold higher in blue light than in 389 darkness in koy2 mutant seed (Figure 3). These data indicate that phyA is required for the 390 induction of the genes that repress germination – AearGA2ox3 and AearNCED5 – in blue 391 light-treated Ae. arabicum seeds. The expression of AearGA3ox1 and AearCYP707A2 was 392 identical between blue light-treated wild type and koy2 mutants, indicating that their 393 expression is not modulated by phyA. Taken together, the expression of negative regulators 394 of seed germination – AearGA2ox3 and AearNCED5 – is induced in wild type in red, far-red, 395 and blue light conditions. Their expression is lower in koy2 mutant than in wild type in each of 396 the light conditions. This indicates that phyA (KOY2) activity promotes the expression of the 397 germination repressors AearGA2OX3 and AearNCED5 in red, far-red, and blue light in Ae. 398 arabicum. 399 400 phyA (KOY2) is required for the high irradiance response-repression of germination in 401 non-dormant seed 402 Ae. arabicum seed germination does not require light, because 100% of ripened non-403 dormant seeds germinate in darkness. While the inhibitory effects of light on germination can 404 be tested on after-ripened, non-dormant seed, the promotion of germination by light can be 405 tested by evaluating the effects of light exposure on semi-dormant seed, a state that is 406 reached if seeds are stored for 8 weeks after harvest. 407 Ae. arabicum seed germination is repressed by the high irradiance response (Mérai et al., 4 08 2023). To test if germination inhibition by high irradiance response is mediated by phyA, we 409 compared the germination of wild type and koy2 seeds under constant light where the high 410 irradiance response is active and represses germination. Illumination with continuous red or 411 far-red light inhibited germination of wild type seed. Zero % of wild type germinated when 412 imbibed seed was illuminated in 392 mmol m-2 h-1 continuous red or 0.072 mmol m-2 h-1 413 continuous far-red light, confirming that the high irradiance response repressed germination 414 (Figure 4A). By contrast, 95.7% of koy2 seeds incubated in 917 mmol m-2 h-1 continuous red, 415 and 100% koy2 seed incubated in 0.6 mmol m-2 h-1 far-red light germinated (Figure 4A). 416 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 13 These data are consistent with the hypothesis that phyA (KOY2) is required for the high 417 irradiance repression of germination. 418 A defining characteristic of the high irradiance response is the requirement for continuous 419 illumination to trigger a physiological response; the high irradiance response is not induced if 420 light application is interrupted by intermittent dark pulses (intermittent light). If the high 421 irradiance response represses germination, we hypothesized that (i) the germination 422 percentage of wild type would be higher in intermittent light than under continuous light and 423 (ii) koy2 mutants would be similar in intermittent light, because HIR would be inactive in both 424 genotypes. With intermittent red (red light with intermittent darkness) or far-red (far-red with 425 intermittent darkness) exposure, wild type seeds germinated (100% in red and 95% in far-426 red) while 0% of wild type seed grown in continuous red or far-red light germinated. Second, 427 similar germination rates were found in wild type (100% in red and 95% in far-red) and koy2 428 mutants (100% in red and far-red). These data are consistent with the hypothesis that the 429 high irradiance response is defective in koy2 mutants. 430 The defective germination of koy2 mutants indicates that phyA (KOY2) activity is required for 431 the repression of seed germination by the red and far-red high irradiance response in Ae. 432 arabicum. 433 434 PhyA is required for VLFR-promoted germination of semi-dormant seed 435 The very low fluence response promotes germination in Ae. arabicum (Mérai et al., 2023). 436 Since VLFR promotes germination, it can be evaluated in semi-dormant seed 8 weeks after 437 harvest. To test if phyA is required for the very low fluence response control of seed 438 germination, we determined the phenotype of koy2 mutants in conditions where a light pulse 439 24 h after imbibition activates this response. 28% of semi-dormant wild type seed germinated 440 in the dark (optimal germination conditions) (Figure 4C). A far-red light pulse 24 h after 441 imbibition increased the germination to 87%. This confirms that the very low fluence 442 response promotes seed germination in wild type Ae. arabicum seed (Figure 4C) (Mérai et 443 al., 2023). By contrast, koy2 germination rate was the same in seed incubated in darkness 444 (40%) and in darkness supplemented with a red light pulse after 24 h (40%) (Figure 4C). The 445 lack of germination induction after a red light pulse at 24 h in koy2 mutant seed indicates that 446 phyA is required for the very low fluence response that promotes germination in semi 447 dorm ant Ae. arabicum seed. 448 449 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 14 The low fluence response does not control germination of semi dormant seed but 450 phyA (KOY2) promotes germination after a pulse of far-red light 451 The role of the phytochrome-mediated low fluence response in germination in Ae. arabicum 452 is unknown. We set out to test if the low fluence response controls germination. A defining 453 characteristic of the low fluence response is reversibility; red light pulses induce the low 454 fluence response while far-red pulse inactivates the response. We hypothesized that if the 455 low fluence response controls germination, the germination rate of semi-dormant seed would 456 be higher after a red pulse and lower after far-red pulse than in darkness. 457 First, we tested the effect of red and far-red pulses on non-dormant seed. Over 95% of after-458 ripened, non-dormant Ae. arabicum seeds germinated in darkness, indicating that light is not 459 required for germination of seeds at this stage of maturity. Similar levels of germination were 460 observed when a red pulse or far-red pulse was followed by darkness (Figure 4B). This 461 indicates that a short light pulse does not impact germination in non-dormant seed 462 (Figure 4B). These data suggest that the low fluence response does not operate in non-463 dormant seed to promote germination. 464 To determine if the low fluence response controls germination of semi-dormant Ae. 465 arabicum, seed, we measured germination of seed batches approximately 8 weeks after 466 harvest. The maximum seed germination ranged from 20 to 50% when wild-type seeds were 467 plated in the dark (optimal) conditions. To test if pulses of red or far-red light impacted 468 germination, semi-dormant seed were exposed to light pulses upon imbibition followed by 469 dark incubation. The germination of wild type semi-dormant seeds was not significantly 470 different in darkness and in applications where a red pulse was followed by darkness 471 (Figure 4B). This indicates that the low fluence response does not control the germination of 472 semi-dormant wild-type seed. 473 If the low fluence response does not control germination in Ae. arabicum, we hypothesized 474 that the germination of semi-dormant seed would be identical in wild type and koy2 (phya) 475 mutants. To test this hypothesis, we compared the germination of wild type and koy2 semi-476 dormant seed in conditions where the low fluence response would be active if present. The 477 germination rates of wild type and koy2 mutant seed was similar when incubated in darkness 478 or after receiving a red pulse. This indicates that the low fluence response is not activated by 479 red light in either wild type or the koy2 mutant. These data further support the hypothesis that 480 the low fluence response does not control seed germination in Ae. arabicum. 481 Next, we compared wild type semi-dormant seed germination in darkness with germination in 482 darkness after a pulse of far-red light. 40% of wild type germinated in darkness while 31% 483 germinated after a far-red light pulse. This indicates that far-red light at imbibition does not 484 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 15 significantly affect germination in wild type semi-dormant seed. However, germination was 485 significantly lower after a far-red pulse in the koy2 mutant (11.9%) than in darkness (51.8%) 486 (Figure 4B). These data indicate that phyA promotes germination induced by a pulse of far-487 red light at imbibition in semi-dormant seeds of Ae. arabicum. 488 Taken together, these data provide no evidence for the low fluence response in wild type 489 seeds. However, phyA (KOY2) is required for the stimulation of germination by far-red light. 490 491

492 We report the discovery that phyA is required for negative photoblastic germination in Ae. 493 arabicum (CYP accession) seeds. We identified a mutant (koy2) in Ae. arabicum (CYP 494 accession) in a forward genetic screen that is defective in light-inhibited germination. koy2 495 mutant seeds are less responsive to white and blue light than wild type, and unlike wild type 496 do not respond to red and far-red light. The mutation in the koy2 mutant encodes a 497 premature stop codon in the gene encoding the phyA, the far-red photoreceptor in Ae. 498 arabicum. The germination phenotype of the koy2 mutant demonstrates that phyA is required 499 for the repression of germination by white, red, far-red, and blue light in Ae. arabicum. This is 500 the first detailed description of a role for phyA in the control of seed germination in a 501 negatively photoblastic species, where white light strongly represses germination. 502 Phytochromes act through three distinct action modes, HIR, VLFR, and LFR. We 503 demonstrate that both the HIR and VLFR operate during phyA-modulated germination in Ae 504 arabicum. We show that phyA-mediated HIR promotes negative photoblasty in Ae. arabicum. 505 Both red and far-red light-mediated HIR repress germination. While there are numerous 506 examples of far-red light-mediated HIR in plants, this is only the second example of red light-507 induced HIR controlling germination. Previously, Appenroth (2006) demonstrated that red-508 light-induced HIR-repressed germination by ~50% (Appenroth et al., 2006). In Ae. arabicum, 509 the HIR effect is much stronger; 100% of seeds do not germinate when imbibed in red light 510 HIR conditions. 511 HIR and VLFR function antagonistically in photoblastic seed germination in Ae. arabicum. 512 Red and far-red HIR represses germination while far-red VLFR promotes germination. This 513 antagonism between HIR and VLFR has also been demonstrated in the weed Datura ferox 514 where far-red partially represses germination through HIR and induced germination through 515 VLFR (Arana et al., 2007). 516 We demonstrate that phytochrome-mediated LFR does not act in Ae. arabicum to control 517 seed germination. Phytochrome-mediated LFR controls germination in many species 518 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 16 (Borthwick et al., 1952; Shropshire et al., 1961; Negbi and Koller, 1964). This action mode 519 has been characterized in detail in A. thaliana and Lactuca sativa (lettuce) (Hartmann, 1966; 520 Shinomura et al., 1996). In these species, red light promotes germination, and far-red light 521 represses germination. This red light promotion and far-red light repression of germination is 522 a characteristic of the LFR action mode and is mediated by phytochrome B, which is present 523 in mature seed. Exposure to red light increases the Pfr:Pr ratio and activates the germination 524 program through a PIF-dependent mechanism that increases GA levels and decreases ABA. 525 Exposure to far-red light on the other hand, lowers the Pfr:Pr ratio and results in a decreased 526 GA:ABA ratio. We demonstrated that the steady state mRNA level of the genes encoding the 527 enzyme involved in GA degradation (AearGA2ox3) and ABA biosynthesis (AearNCED5) are 528 much higher in both red and far-red light than in darkness. This would be expected to result 529 in a decreased GA:ABA ratio, which would inhibit germination, in both red and far-red light. 530 This is consistent with the hypothesis that red light does not promote germination through 531 LFR in Ae. arabicum. 532 We demonstrated that phytochrome-mediated HIR repress germination and phytochrome-533 mediated LFR has no function in germination control in the CYP accession of Ae. arabicum. 534 It has been proposed that the photoblastic mode – negative or positive – of seeds is 535 determined by the HIR in negative photoblastic seed and LFR in positive photoblastic seed. 536 There are many examples where LFR is involved in positive photoblastic seed germination 537 (Shinomura et al., 1996; Appenroth et al., 2006). However, there are no examples where the 538 presence of HIR and lack of LFR has been shown to be involved in negative photoblastic 539 seed. Our demonstration that red and far-red light HIR are required for negative photoblastic 540 germination supports this hypothesis. We provide the first complete set of evidence that 541 supports the assertion that the distinct phytochrome action modes define seed germination 542 behavior. 543 The photoblastic phenotype is an adaptive seed trait that has likely evolved as an adaptation 544 for seedling establishment in the natural habitat. The Cyprus accession of Ae. arabicum 545 grows in an exposed habitat with high illumination. Seedling establishment is poor in 546 summer, when light levels and temperature are high and little water is available. Therefore, 547 avoiding germination during the long days of summer may be adaptive. We speculate that 548 phyA-mediated HIR represses germination during the summer in this accession of Ae. 5 49 arabicum. Other accessions of Ae. arabicum are light neutral – seeds germinate in both 550 darkness and light. We speculate that genetically encoded variations in phytochrome A-551 mediated light responses are responsible for local adaptation of Ae. arabicum throughout the 552 Irano-Turanian region. 553 554 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 17

555 We thank Eric M. Schranz for providing Aethionema seed stocks. We also thank the staff of 556 the Vienna BioCenter Core Facilities GmbH (VBCF), a member of Vienna BioCenter (VBC), 557 Austria, especially the Plant Sciences Facility for the growth of the plants, the Molecular 558 Biology Unit for providing multiple reagents, the Vienna Covid-19 Detection Initiative (VCDI) 559 for generating a safe work environment during the pandemic and the VBC Child Care Center. 560 We thank Nicole Lettner for the technical support. We thank Ortrun Mittelsten Scheid for the 561 fruitful discussions and advices throughout the project. 562 Funding 563 This research was funded by the Austrian Science Fund (FWF) I3979-B25/ DOI: 564 10.55776/I3979. For open access, the author has applied a CC BY public copyright license 565 to any Author Accepted Manuscript version arising from this submission. It was additionally 566 supported by the National Research, Development and Innovation Office (Hungary), grant 567 number AN-128740 (LKB), K-134567 (LKB) and PD-138963 (AH). LD is funded by the 568 Austrian Academy of Sciences, and advanced grant from the European Research Council 569 (project number 787613). 570 Author contributions 571 ZM planned and designed the research; ZM, FX, and AH performed the experiments; ZM, 572 FX, LKB, and LD analyzed and interpreted the data; ZM and LD wrote the paper. All authors 573 approved the submitted version. 574 Declaration of interest 575 The authors declare no competing interests. 576 577 Figure legends 578 Figure 1. Identification of koy2 as a phya null mutant. (A) Germination response of wild type 579 (WT) and koy2 mutant seeds under darkness (at light intensity 0) and continuous white light. 580 Maximal germination was scored after 7 days. Error bars indicate the standard deviation of 581 three biological replicates. (B) Images of wild type (WT) and koy2 seeds after being kept in 582 darkness or under 96 µmol m-2 s-1 white light for 7 days. (C) Hypocotyl elongation test of wild 583 type (WT) and koy2 seedlings. Seeds were kept in darkness for two days to induce 584 germination, then illuminated with red, far-red, or blue light for 5 days during seedling growth. 585 At least 12 hypocotyls were measured for each point. Error bars indicate the standard 586 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 18 deviation. (D) Identified mutation in the coding sequence of the gene encoding phyA 587 (Aa31LG1G5460). Protein domain structure of wild type phyA protein and the truncated 588 protein in koy2 lacking the chromophore binding site. (E-F) Relative expression of AearCHS 589 (E) and AearPHYA (F) in wild type (WT) and koy2 seeds, incubated in darkness, under red, 590 far-red, or blue light for 24 h. Expression level is normalized to the level in wild type (WT) in 591 darkness, which is set to 1. `Fold change` numbers indicate the fold change of expression in 592 red, far-red, or blue exposure of the koy2 mutant relative to the level in koy2 seeds incubated 593 in darkness. Error bars indicate the standard deviation of three biological replicates. 594 Asterisks indicate significant differences from the wild type or koy2 dark level with P-values 595 as *P < 0.05, **P < 0.01, and ***P < 0.001 calculated with the Tukey test. 596 (G) Accumulation of phyA protein (indicated with arrow) in wild type (WT) and koy2 597 seedlings. 598 Figure 2. Germination response of wild-type (WT) and koy2 mutant seeds under far-red (A), 599 red (B), and blue (C) light. Maximal germination was scored after 7 days. Error bars indicate 600 the standard deviation of three biological replicates. 601 Figure 3. Light-induced gene expression of genes encoding GA and ABA synthesis and 602 catabolic enzymes. Relative expression of selected genes in wild type (WT) and koy2 seeds 603 incubated in darkness, or seeds incubated under 80 µmol m-2 s-1 red light, 1 µmol m-2 s-1 far-604 red light, or 100 µmol m-2 s-1 blue light for 24 h. Expression level is normalized to the level in 605 wild type (WT) in darkness, which is set to 1. `Fold change` numbers indicate the fold change 606 of expression in red, far-red, or blue exposure of the koy2 mutant relative to the level in koy2 607 seeds incubated in darkness. Error bars indicate the standard deviation of three biological 608 replicates. Asterisks indicate significant differences from the wild type or koy2 dark level with 609 P-values as *P < 0.05, **P < 0.01, and ***P < 0.001 calculated with the Tukey test. 610 Figure 4. Phytochrome action modes in Ae. arabicum seeds. (A) Total germination of wild 611 type and koy2 seeds under constant red (Rc, left panel) or far-red light (FRc, right panel) or 612 15 min light pulses (Rp, dashed lines, left panel, or FRp, dashed lines, right panel) 613 intermitted with 45 min dark periods. Dots indicate the average of three biological replicates. 614 Maximal germination was scored after 6 days. (B) Total germination of wild type non-615 dormant (left panel) or semi-dormant (right panel) and koy2 semi-dormant seeds. Seeds 616 were exposed to 10 min of 100 µmol m-2 s-1 red light pulse (Rp) or 10 µmol m-2 s-1 far-red 617 pulse (FRp) followed by 6 days incubation in darkness. `Dark` indicates that seeds were kept 618 in darkness without any light pulses. Error bars indicate the standard deviation of three 619 biological replicates. (C) Total germination of semi-dormant wild type and koy2 seeds. Seeds 620 were imbibed in complete darkness. 24 h after imbibition, seeds were illuminated for 10 min 621 with 10 µmol m-2 s-1 far-red (FRp) or kept in darkness. Germination was scored 6 days after 622 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 19 the imbibition. (B-C) Asterisks indicate significant differences in pairwise comparisons within 623 one genotype with P-values as *P < 0.05, **P < 0.01, and ***P < 0.001 calculated with the 624 Students t-test. (D) schematic representation of the phytochrome action modes in the 625 positive photoblastic Arabidopsis seeds and the negative photoblastic seeds in Ae. arabicum 626 (CYP accession). Green arrows indicate germination induction and red arrows indicate 627 germination inhibition. 628 629

630 Ali, F. et al. (2022) ‘Updated role of ABA in seed maturation, dormancy, and germination’, 631 Journal of Advanced Research, 35, pp. 199–214. Available at: 632 https://doi.org/10.1016/j.jare.2021.03.011. 633 Appenroth, K. et al. (2006) ‘Tomato seed germination: regulation of different response modes 634 by phytochrome B2 and phytochrome A’, Plant, Cell & Environment, 29(4), pp. 701–709. 635 Available at: https://doi.org/10.1111/j.1365-3040.2005.01455.x. 636 Arana, M.V. et al. (2007) ‘The very-low-fluence and high-irradiance responses of the 637 phytochromes have antagonistic effects on germination, mannan-degrading activities, and 638 DfGA3ox transcript levels in Datura ferox seeds’, Journal of Experimental Botany, 58(14), pp. 639 3997–4004. Available at: https://doi.org/10.1093/jxb/erm256. 640 Arshad, W. et al. (2019) ‘Dispersal biophysics and adaptive significance of dimorphic 641 diaspores in the annual Aethionema arabicum (Brassicaceae)’, The New Phytologist, 221(3), 642 pp. 1434–1446. Available at: https://doi.org/10.1111/nph.15490. 643 Borthwick, H.A. et al. (1952) ‘A Reversible Photoreaction Controlling Seed Germination’, 644 Proceedings of the National Academy of Sciences of the United States of America, 38(8), pp. 645 662–666. Available at: https://doi.org/10.1073/pnas.38.8.662. 646 Cantón, F.R. and Quail, P.H. (1999) ‘Both phyA and phyB mediate light-imposed repression 647 of PHYA gene expression in Arabidopsis’, Plant Physiology, 121(4), pp. 1207–1216. 648 Available at: https://doi.org/10.1104/pp.121.4.1207. 649 Casal, J.J., Candia, A.N. and Sellaro, R. (2014) ‘Light perception and signalling by 650 phytochrome A’, Journal of Experimental Botany, 65(11), pp. 2835–2845. Available at: 651 https://doi.org/10.1093/jxb/ert379. 652 Casal, J.J. and Sánchez, R.A. (1998) ‘Phytochromes and seed germination’, Seed Science 653 Research, 8(3), pp. 317–329. Available at: https://doi.org/10.1017/S0960258500004256. 654 Casal, J.J., Sánchez, R.A. and Botto, J.F. (1998) ‘Modes of action of phytochromes’, Journal 655 of Experimental Botany, 49(319), pp. 127–138. Available at: 656 https://doi.org/10.1093/jxb/49.319.127. 657 Fernandez-Pozo, N. et al. (2021) ‘Aethionema arabicum genome annotation using PacBio 658 full-length transcripts provides a valuable resource for seed dormancy and Brassicaceae 659 evolution research’, The Plant Journal: For Cell and Molecular Biology, 106(1), pp. 275–293. 660 Available at: https://doi.org/10.1111/tpj.15161. 661 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 20 Hajdu, A. et al. (2018) ‘ELONGATED HYPOCOTYL 5 mediates blue light signalling to the 662 Arabidopsis circadian clock’, The Plant Journal: For Cell and Molecular Biology, 96(6), pp. 663 1242–1254. Available at: https://doi.org/10.1111/tpj.14106. 664 Hartmann, K.M. (1966) ‘A General Hypothesis to Interpret “High Energy Phenomena” of 665 Photomorphogenesis on the Basis of Phytochrome’, Photochemistry and Photobiology, 5(5–666 6), pp. 349–365. Available at: https://doi.org/10.1111/j.1751-1097.1966.tb05937.x. 667 Hennig, L. et al. (2002) ‘Phytochrome E Controls Light-Induced Germination of Arabidopsis’, 668 Plant Physiology, 128(1), pp. 194–200. Available at: https://doi.org/10.1104/pp.010559. 669 Kevei, E. et al. (2007) ‘Arabidopsis thaliana circadian clock is regulated by the small GTPase 670 LIP1’, Current biology: CB, 17(17), pp. 1456–1464. Available at: 671 https://doi.org/10.1016/j.cub.2007.07.018. 672 Lee, K.P. et al. (2012) ‘Spatially and genetically distinct control of seed germination by 673 phytochromes A and B’, Genes & Development, 26(17), pp. 1984–1996. Available at: 674 https://doi.org/10.1101/gad.194266.112. 675 Lenser, T. et al. (2016) ‘Developmental Control and Plasticity of Fruit and Seed Dimorphism 676 in Aethionema arabicum’, Plant Physiology, 172(3), pp. 1691–1707. Available at: 677 https://doi.org/10.1104/pp.16.00838. 678 Lenser, T. et al. (2018) ‘When the BRANCHED network bears fruit: how carpic dominance 679 causes fruit dimorphism in Aethionema’, The Plant Journal: For Cell and Molecular Biology, 680 94(2), pp. 352–371. Available at: https://doi.org/10.1111/tpj.13861. 681 Mérai, Z. et al. (2019) ‘Aethionema arabicum: a novel model plant to study the light control of 682 seed germination’, Journal of Experimental Botany, 70(12), pp. 3313–3328. Available at: 683 https://doi.org/10.1093/jxb/erz146. 684 Mérai, Z. et al. (2023) ‘Phytochromes mediate germination inhibition under red, far-red, and 685 white light in Aethionema arabicum’, Plant Physiology, 192(2), pp. 1584–1602. Available at: 686 https://doi.org/10.1093/plphys/kiad138. 687 Mérai, Z. et al. (2024) ‘Long days induce adaptive secondary dormancy in the seeds of the 688 Mediterranean plant Aethionema arabicum’, Current biology: CB, 34(13), pp. 2893-2906.e3. 689 Available at: https://doi.org/10.1016/j.cub.2024.05.043. 690 Mohammadin, S. et al. (2017) ‘Anatolian origins and diversification of A ethionema, the sister 691 lineage of the core Brassicaceae’, American Journal of Botany, 104(7), pp. 1042–1054. 692 Available at: https://doi.org/10.3732/ajb.1700091. 693 Negbi, M. and Koller, D. (1964) ‘Dual Action of White Light in the Photocontrol of 694 Germination of Oryzopsis miliacea12’, Plant Physiology, 39(2), pp. 247–253. Available at: 695 https://doi.org/10.1104/pp.39.2.247. 696 Rancurel, C. et al. (2019) ‘SATQPCR: Website for statistical analysis of real-time quantitative 697 PCR data’, Molecular and Cellular Probes, 46, p. 101418. Available at: 698 https://doi.org/10.1016/j.mcp.2019.07.001. 699 Seo, M. et al. (2006) ‘Regulation of hormone metabolism in Arabidopsis seeds: phytochrome 700 regulation of abscisic acid metabolism and abscisic acid regulation of gibberellin 701 metabolism’, The Plant Journal, 48(3), pp. 354–366. Available at: 702 https://doi.org/10.1111/j.1365-313X.2006.02881.x. 703 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 21 Seo, M. et al. (2009) ‘Interaction of light and hormone signals in germinating seeds’, Plant 704 Molecular Biology, 69(4), pp. 463–472. Available at: https://doi.org/10.1007/s11103-008-705 9429-y. 706 Shinomura, T. et al. (1994) ‘The Induction of Seed Germination in Arabidopsis thaliana Is 707 Regulated Principally by Phytochrome B and Secondarily by Phytochrome A’, Plant 708 Physiology, 104(2), pp. 363–371. Available at: https://doi.org/10.1104/pp.104.2.363. 709 Shinomura, T. et al. (1996) ‘Action spectra for phytochrome A- and B-specific photoinduction 710 of seed germination in Arabidopsis thaliana’, Proceedings of the National Academy of 711 Sciences of the United States of America, 93(15), pp. 8129–8133. Available at: 712 https://doi.org/10.1073/pnas.93.15.8129. 713 Shinomura, T., Uchida, K. and Furuya, M. (2000) ‘Elementary processes of photoperception 714 by phytochrome A for high-irradiance response of hypocotyl elongation in Arabidopsis’, Plant 715 Physiology, 122(1), pp. 147–156. Available at: https://doi.org/10.1104/pp.122.1.147. 716 Shropshire, W., Klein, W.H. and Elstad, V.B. (1961) ‘Action spectra of photomorphogenic 717 induction and photoactivation of germination in Arabidopsis thaliana’, Plant and Cell 718 Physiology, 2(1), pp. 63–69. Available at: 719 https://doi.org/10.1093/oxfordjournals.pcp.a077664. 720 Somers, D.E. et al. (1991) ‘The hy3 Long Hypocotyl Mutant of Arabidopsis Is Deficient in 721 Phytochrome B’, The Plant Cell, 3(12), pp. 1263–1274. Available at: 722 https://doi.org/10.1105/tpc.3.12.1263. 723 Takaki, M. (2001) ‘New proposal of classification of seeds based on forms of phytochrome 724 instead of photoblastism’, Revista Brasileira de Fisiologia Vegetal, 13, pp. 104–108. 725 Available at: https://doi.org/10.1590/S0103-31312001000100011. 726 Wolf, I. et al. (2011) ‘Light-regulated nuclear import and degradation of Arabidopsis 727 phytochrome-A N-terminal fragments’, Plant & Cell Physiology, 52(2), pp. 361–372. Available 728 at: https://doi.org/10.1093/pcp/pcq194. 729 730 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 22 Supporting Information 731 Supporting Table S1 List of primers used for quantitative RT-PCR analysis. 732 733 Supporting Table S2. Cosegregation of long hypocotyl phenotype under far-red 734 with the mutation identified in koy2. 735 736 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 0 20 40 60 80 100 0 100 200 300 Const. WHITE WT koy-2 Germination rate (%) Fluence rate (µmol m-2 s-1) A B WT koy2 * 822 TGC>TGA PHYA cDNA 3372 nt 1122 aa chromophore attachment C323 nPAS GAF PHY PAS1 PAS2 HKRD 82 194 409 587 631 735 765 877 phyA protein nPASkoy2 phyA *273 aa Fluence rate (µmol m-2 s-1) 0 0,5 1 1,5 2 2,5 0,1 1 10 100 WT koy-2 Hypocotyl length (mm) 0 5 10 15 20 25 0 0,5 1 1,5 2 2,5 0,001 0,01 0,1 1 WT koy-2 Hypocotyl length (mm) 0 5 10 15 20 25 RED FAR-RED Fluence rate (µmol m-2 s-1) C D 0 0,5 1 1,5 2 2,5 0,5 5 50 WT koy-2 BLUE Fluence rate (µmol m-2 s-1) α-phyA α-actin WT seedling Dark Red Blue Far-red Dark F 0 0,2 0,4 0,6 0,8 1 1,2 1,4 WT koy-2 PHYA Dark Red Far-red Blue 0 5 10 15 WT koy-2 CHS*** *** *** ***ns ** *** **** ns ** ns Fold change: 0.26|1.15|1.1 Fold change: 6.1|1.2|17.2 E G Fig. 1 Identification of koy2 as a phya null mutant DarkLight Rel. expression Rel. expression Aa31LG1G5460 koy2 koy2 koy2 koy2 koy2 koy2 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint C A B 0 20 40 60 80 100 1 10 100 WT koy-2 Germination rate (%) 0 7 500 Const. RED Fluence rate (µmol m-2 s-1) 0 20 40 60 80 100 1 10 100 WT koy-2 500 Const. BLUE 0 20 40 60 80 100 0,001 0,01 0,1 1 WT koy2 Germination rate (%) Fluence rate (µmol m-2 s-1) Const. FAR-RED Germination rate (%) Fluence rate (µmol m-2 s-1) Fig. 2 Germination response of koy2 mutant koy2 koy2 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 0 0,5 1 1,5 2 2,5 3 3,5 WT koy-2 GA3ox1 0 50 100 150 200 WT koy-2 GA2ox3 Dark Red Far-red Blue 0 1 2 3 4 5 6 WT koy-2 NCED5 0 5 10 15 20 25 30 WT koy-2 NCED6 0 0,5 1 1,5 2 2,5 3 WT koy-2 CYP707A2 *****ns ** ns ns ********* *ns ** *ns ** ****** *** * **** ns ** ns *ns *** *ns * Fold change: 3.67|0.6 |1.4 Fold change: 4.79|1.9 |2.96 Fold change: 1.7 |1.1 |1.82 Fold change: 0.4 |20.3|0.3 0.19|1.27|1.66 Fold change: 1.9 |0.8 |0.76 Fig. 3 Light induced gene expression of genes encoding GA and ABA synthesis and catabolic enzymes. Rel. expression Rel. expression Rel. expression Rel. exrpressionRel. expression koy2koy2 koy2 koy2 koy2 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint 0% 20% 40% 60% 80% 100% 0 500 1000 Red HIR Rc WT Rc koy-2 Rp WT Rp koy-2 Germination rate (%) Light intensity per hour (mmol m-2h-1) 0% 20% 40% 60% 80% 100% 0,01 0,1 1 10 Far-red HIR FRc WT FRc koy-2 FRp WT FRp koy-2 Light intensity per hour (mmol m-2h-1) Rc Rp 6 d 15minR/45minD FRc FRp 6 d 15minFR/45minD A B C Germination rate (%) Fig. 4 Phytochrome action modes in Aethionema seed 0 20 40 60 80 100 Dark Red Far-red WT, non-dormant semi-dormant imbibition 6 d Rp Dark Rp FRp 6 d FRp 6 d LFR LFR ***n.s. * Germination rate (%) Germination rate (%) 0 20 40 60 80 100 WT koy-2 Dark Red Far-red n.s. Dark Rp FRp Dark Rp FRp WT koy2 0 20 40 60 80 100 WT koy-2 Dark FRp** n.s. VLFR 24 hrs imbibition 5 d FRp Dark FRp semi-dormant 6 d Germination rate (%) WT koy2 Arabidopsis (Col-0 accession) Aethionema (CYP accession) PHYA PHYB PHYA VLFR LFR VLFR R-HIR FR-HIR D koy2 koy2 koy2koy2 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 28, 2025. ; https://doi.org/10.1101/2025.02.26.640300doi: bioRxiv preprint