Efficient CRISPR–Cas genome editing in brown algae

44 Brown algae・ CRISPR-Cas・ genome editing・ RNP-transfection 45 Research topic(s) 46 CP: molecular biology 47

48 Brown algae represent a lineage of marine photosynthetic eukaryotes that evolved complex 49 multicellularity independently of plants and animals 1 through mechanisms that remain 50 largely unexplored2–5. Ubiquitous in coastal environments, they play key ecological roles, 51 serve as a natural source of bioactive compounds for industrial applications, and constitute 52 an important food resource6. However, the lack of efficient and broadly accessible genome 53 editing protocols has hindered both fundamental and applied research. 54 Previous efforts to deliver Clustered Regularly Interspaced Short Palindromic Repeats 55 (CRISPR)–Cas9 ribonucleoproteins (RNPs) in brown algae employing biolistic and/or 56 microinjection methods have successfully yielded genomic edits 7,8. However, these labour-57 intensive approaches rely on specialized equipment, require extensive technical training, 58 and are limited by relatively low editing efficiencies; for instance, biolistic delivery has been 59 associated with a false-positive rate of approximately 50%7. 60 Here, we present a genome editing method based on Polyethylene glycol (PEG)-mediated 61 RNP-transfection that achieves unprecedented efficiency with negligible false-positive rate 62 across four brown algae: Ectocarpus sp. 7 and Scytosiphon promiscuus , widely used 63 systems for fundamental research9–12, Laminaria digitata, a laminariacean kelp of ecological 64 and economic importance, and Undaria pinnatifida (wakame), an alariacean kelp and one 65 of the most commercially cultivated brown algae worldwide6. 66 .CC-BY-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 3

67 PEG-mediated RNP Ectocarpus gamete transfection allows highly-efficient 68 genome editing 69 To develop a simple and highly efficient genome editing method in Ectocarpus sp., we 70 exploited its ability to regenerate haploid partheno-sporophytes (PSPs) from unfertilized 71 gametes via parthenogenesis (Fig. 1A). Notably, Ectocarpus gametes remain cell wall-free 72 for approximately two hours following release from gametangia in the absence of 73 fertilization3, providing a transient window during which CRISPR–Cas RNPs can be 74 efficiently transfected. 75 To transfect Cas12-RNPs we employed PEG-mediated RNP transfection, previously 76 successful in the green alga Ulva prolifera 14. While mature male Ectocarpus gametophytes 77 (GAs) release abundant swimming gamete 15,16 ( Fig. 1 A) and were therefore used to 78 optimize the protocol, mutations can be easily backcrossed into wild-type females 15,16 79 provided fertility is not affected. Because the optimal growth temperature of this alga is 80 14°C, we used an engineered Cas12 (Alt-R L.b. Cas12a [Cpf1], IDT) version with increased 81 temperature tolerance and optimal for systems requiring lower culture temperatures. We 82 isolated male gametes from cultivated algae and transfected them with Cas12-RNPs using 83 40% w/v PEG (Fig. 1B, see methods). As selection marker, we used Cas12 loaded with a 84 APT locus crRNA (Ec-28_000520, Table S1) to generate apt loss-of-function mutations 85 which enable survival in selective medium supplemented with 2-fluoroadenine (2-FA)7. 86 Initially, we tested the effect of an overnight heat-shock at 22 ° C in darkness after PEG 87 transfection ( Fig. S1A) and of different Cas12 to crRNA ratios ( Fig. S1B) with no 88 statistically significant effect on the number of 2-FA resistant obtained ( Fig. S1, Table S2 ). 89 However, the heat-shock step was maintained since it resulted in higher number of 2-FA–90 resistant individuals (Fig. S1A). Additionally, we tested the effect of different PEG molecular 91 weights and observed that PEG8000 resulted in higher number of 2-FA resistant individuals 92 (0.003% of transfection efficiency) averaging 30.3 2-FA resistant individuals per trial i.e. per 93 Petri dish (Fig. 1C, Table S2 ). Data analysis from multiple independent experiments during 94 protocol optimization, revealed that transfection within the first two hours after gamete 95 release significantly improves editing efficiency ( Fig. 1D, Table S2 ), and thus it is 96 recommended to proceed with transfection as soon as gamete release is completed. 97 Given the method’s high efficiency in generating 2-FA resistant individuals, we next tested 98 whether it could produce double mutants targeting both APT and a gene of interest. To 99 achieve this, we designed two Cas12 crRNAs targeting exons 3 and 4 of the IMMEDIATE 100 UPRIGHT ( IMM, Ec-27_002610.1) gene and co-transfected them with the APT RNP 101 complex ( Table S1 ). Mutations in the IMM locus are known to disrupt basal cell 102 development and accelerate upright filament formation in Ectocarpus sp.5 providing a clear 103 developmental phenotype to assess the efficiency of our method (Fig. 1E). Remarkably, we 104 observed this characteristic imm phenotype in 19 individuals out of 59 (32%) of 2-FA 105 resistant individuals following co-transfection. We selected 15 individuals displaying the 106 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 4 imm phenotype and confirmed, employing Sanger sequencing, that all harbored mutations 107 at both the APT and IMM loci (Table S3). In contrast, three 2-FA resistant individuals with a 108 wild-type phenotype showed mutations only at the APT locus ( Table S3). Notably, one of 109 the 15 double mutants carried a 723 bp deletion between the two IMM target sites, while 110 others had smaller indels at individual sites ( Fig. 1F, Table S3 ). While crRNA targeting 111 exon 3 produced mutations in all tested individuals, crRNA targeting exon 4 generated 112 mutations in only 2, suggesting that crRNA or target site susceptibility may differ. 113 Importantly, no false positives were observed since all algae that grew on selective medium 114 and tested, carried mutations in APT, confirming the high precision and efficiency of the co-115 transfection protocol. 116 PEG-mediated RNP transfection results in high-frequency genome edits 117 across multiple brown algal species 118 To test the transferability of the PEG-mediated RNP transfection protocol beyond 119 Ectocarpus, we applied the method to other non-model brown algae. Specifically, we 120 transfected APT-targeting RNPs into S. promiscuus and the kelps L. digitata and U. 121 pinnatifida ( Fig. 2 ), introducing only minor protocol modifications (see Methods). In S. 122 promiscuus, as in Ectocarpus, gametes released from laboratory-cultivated algae were 123 used for PEG transfection (Fig. 2A, B ). For the kelps, we instead used meiospores 124 transformation (Fig. 2C, D), since kelp gametes exhibit limited parthenogenetic capacity. 125 Meiospores develop into haploid gametophytes, allowing direct phenotypic evaluation of 126 mutation effects without the need for backcrossing or generating homozygous lines. 127 The transfection rate in S. promiscuus was approximately 0.25% allowing to retrieve 128 between 107 and 2458 2-FA resistant individuals with no false positives in the all individuals 129 tested ( Table S4). In L. digitata transformation efficiency varied between 0-0.79% and 130 yielded between 0 and 6128 2-FA resistant individuals, depending on the trial (Table S4 ). 131 This discrepancy was likely due to the fact that fresh zoospores - i.e., actively swimming, 132 wall-less cells -were not available in trials 1 and 3, and instead non-motile, possibly cell 133 wall-bearing spores were used ( Table S4 ). In U. pinnatifida the transfection efficiency 134 varied among trials: 0.014-0.021%, which reflects more than 100 2-FA resistant individuals 135 per trials. In L. digitata, genotyped 2FA-resistants exhibited mutations in the APT locus with 136 a false positive rate of 9.3±17.4% among the trials (mean±SD; Table S4 ), while in S. 137 promiscuus and U. pinnatifida, no false positives could be identified. 138 For S. promiscuus , two APT-crRNAs were used simultaneously, causing insertions or 139 deletions ( Fig. 2E ). Most mutants had only one mutation near the crRNA target sites, 140 whereas in two cases, approximately 1.5 kb between the two target sites was entirely 141 deleted (Fig. 2E). Additionally, in one case, an inversion occurred within the region between 142 the two target sites ( Fig. 2E). In L. digitata, where only a single crRNA was used for each 143 PEG transfection, only deletions (-1 to -43 bp) were observed (Fig. 2F). For U. pinnatifida 144 (growing at 20°C) both Cas9 and Cas12 were assayed with each two crRNAs. All were 145 shown to be functional ( Fig. 2G ) with a higher mutation efficiency assessed for Cas12 146 (Table S4 ). Overall, our results support that PEG-mediated Cas12-RNP complex 147 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 5 transfection in gametes or meiospores can be easily scalable and applicable to multiple 148 brown algae species, including kelps of high economic and ecological importance. 149 150 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 6 151 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 7 Figure 1 . Ectocarpus CRISPR-based genome editing. A) Parthenogenetic life cycle of 152 Ectocarpus sp. Gametophyte (GA) produces gametes which parthenogenetically develop 153 into partheno-sporophyte (PSP). B) Schematic diagram of the process of the PEG 154 transfection for Ectocarpus sp . 7 The process begins with gametophytes (GAs) cultured in 155 Petri dishes (step 1), followed by incubation to induce gamete release (step 2). The 156 gametophytes are then transferred using a pipette (step 3), filtered to separate from 157 gametophytic tissue, and centrifuged to concentrate gametes (step 4). After centrifugation, 158 the concentration of gametes is examined using a hemocytometer (step 5). 159 Ribonucleoprotein (RNP) is added to the petri dish (step 6) and mixed with gamete 160 suspension (step 7). Finally, 40% w/v polyethylene glycol (PEG), prepared in seawater, is 161 applied to facilitate transfection and vigorously mixed by pipetting with a wide bore tip (step 162 8). See details in methods section. C) Comparison of the number of 2-FA resistant 163 individuals following treatment with different PEG molecular weights: PEG4000 and 164 PEG8000. Dots depict independent trials (Table S2). The horizontal and vertical lines in the 165 scatter plot represent the mean and standard deviation, respectively. Significantly more 166 resistant individuals were observed with PEG8000 treatment (Exact Wilcoxon rank sum test, 167 p = 0.014). D) Comparison between PEG treatments using gametes collected more than 2 168 hours after release (>2h) and those using freshly released gametes (<2h). A significantly 169 higher number of resistant individuals was observed when fresh gametes were used (exact 170 Wilcoxon rank sum test, p = 0.005). E) Morphological difference of double KO mutant of 171 APT and IMM ( apt;imm) and APT single KO mutant ( apt;IMM). The APT single KO mutant 172 develops prostrate filaments, whereas the double KO mutant develops long upright 173 filaments. F) Schematic representations of the Ectocarpus IMM gene model of WT and 174 mutant individuals. Exon regions are shown as purple boxes, and crRNAs are indicated 175 with arrows. The expected protein products for the WT and each mutant are shown to the 176 right of their respective gene models. The five imperfect tandem repeats of a 38 amino acid 177 cysteine-rich motif, characteristic of the IMM C-terminal region, are represented as white 178 boxes. 179 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 8 180 181 Figure 2 . Scytosiphon and kelp CRISPR-based genome editing. A) Parthenogenetic life 182 cycle of Scytosiphon promiscuus . Macroscopic gametophytes (GA) alternate with 183 microscopic discoidal parthenosporophyte (PSP). B) Suspension of male gametes of S . 184 promiscuus showing accumulation of gametes (asterisk) by negative phototaxis. C) Life 185 cycle of Laminaria digitata and Undaria pinnatifida . In both species, microscopic 186 gametophytes (GA) alternate with macroscopic sporophyte (SP: left, Laminaria sporophyte; 187 right, Undaria sporophyte). D) Released meiospores from Laminaria sori and Undaria 188 sporophylls. E) Schematic representations of the S. promiscuus APT gene model of WT 189 and mutant individuals. See details in Fig. 1. F) Schematic representations of the L. digitata 190 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 9 APT gene model of WT and mutant individuals . G) Schematic representations of the U. 191 pinnatifida APT gene model of WT and mutant individuals following Cas9 transformation 192 (above) and Cas12 transformation (below). H) One-month-old germlings of S. promiscuus 193 APT mutant and WT (right top) in selective medium. 194

195 Our method enabled efficient genome editing in four brown algal species using either 196 gametes or meiospores. Since most brown algae produce parthenogenetic gametes, and 197 even species without such gametes typically generate meiospores, our approach can be 198 applicable to most brown algal species, provided they produce sufficient numbers of wall-199 less zoids. In previous studies using biolistic delivery and microinjection in the model algae 200 Ectocarpus sp 7, only an average of 0.4 and 1 apt mutants were obtained per experiment, 201 respectively7, whereas our method can generate on average 30 individuals per trial. In 202 some instances, even higher numbers have been isolated, which we speculate to be 203 related to the quality and viability of the gametes isolated in each trial. Moreover, with 204 biolistic transformation, fewer than half of the 2-FA resistance individuals carried mutations 205 in the APT gene, indicating a substantial proportion of false positives 7 which would require 206 more intensive screening efforts than our method. In contrast, our method produced tens to 207 thousands of 2-FA resistance individuals per experiment across four different brown algal 208 species, with false positives being absent in most experiments and remaining below 10% at 209 highest. By enabling precise, efficient, and accessible genetic manipulation, our approach 210 opens the possibility to explore the molecular mechanisms underlying brown algal 211 development, physiology, and evolution. Its high efficiency combined with high 212 reproducibility and easy implementation, will enable brown algae studies across a wide 213 community of researchers. The improvement of the protocol therefore represents a major 214 step forward for both fundamental research and biotechnological applications in this unique 215 lineage of complex multicellular marine organisms. 216

of the study 217 The effectiveness of our method may be limited in Fucales and Dictyotales since these 218 groups lack gametes with parthenogenetic capacity 17. Additionally, Fucales do not produce 219 meiospores and thus is not possible to circumvent the lack of parthenogenic gametes while 220 Dictyotales produce meiospores that are large and rich in cytoplasm 17,18 which could 221 potentially interfere with transfection efficiency. These limitations could be circumvented by 222 using brown algae protoplasts for PEG-mediated transfection, as originally reported in 223 tobacco plants 19 with the caveat that protoplast regeneration is time-consuming 20,21. 224 Another possibility would be to transform male gametes and perform a cross immediately 225 after transfection, which would then require further generations to isolate homozygous 226 individuals. While microinjection is labor-intensive and requires specialized equipment, 227 introducing RNPs into vegetative cells or embryos in Fucales and Dictyotales remains an 228 alternative for these brown algal groups. 229 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 10 Acknowledgments 230 We thank Dr. Kensuke Ichihara for valuable advice. We thank Agnes Henschen for help 231 with DNA extractions, and Andrea Belkacemi and Dorothe Koch for assistance with algal 232 cultures. This study was funded by the Max Planck Society, the European Research 233 Council grant 864038 (SMC), the Bettencourt Foundation (SMC) and the Moore Foundation 234 (SMC). M.H. was supported by JSPS KAKENHI Grant number 23K19386, JSPS Overseas 235 Research Fellowships, and Max Planck Partner Groups. 236 Author contributions 237 MH, CM, SMC conceived and designed the experiments. CM, VB, MH and MR developed 238 protocols. CM, VB, MH, AK, KB, RL and MR performed experiments. MH, CM, SMC wrote 239 the manuscript with help of all authors. All authors provided critical feedback and helped 240 shape the research, analysis and manuscript writing. 241 Declaration of interests 242 The authors declare no competing interests. 243 STAR Methods 244 Ectocarpus sp.7 PEG-mediated RNP transfection 245 Ectocarpus sp.7 gene sequences were retrieved from v3 reference genome 22 and 246 CRISPOR23 used to design Cas12 guide RNAS with limited off-targets and highest 247 efficiency scores. Male gametophytes (strain Ec32) were cultured in plastic Petri dishes 248 (150 × 15 mm) (10 individuals per Petri dish) and sterile natural sea water (NSW) enriched 249 with half-strength Provasoli medium 24 (Provasoli enriched sea water: PES) at 14°C in 250 neutral day (12 h: 12 h, light:dark) conditions with LED lighting of 20 μ mol m−2 s−1 photon 251 flux density. Male gametophytes were isolated from individual unilocular sporangia, which in 252 turn were isolated from Ec32 mature partheno-sporophytes (3-4-week old), as previously 253 described25. Mature gametophytes (displaying clear accumulation of mature plurilocular 254 sporangia) are observed usually after 3-4 weeks after culture preparation and clearly visible 255 under a light stereoscope4,25. Gametophytes grown on 50 Petri dishes (500 gametophytes) 256 were collected under a laminar flow hood on a sieve with a 50 μ m mesh and rinsed with 257 NSW at 14°C to remove small filaments and already released gametes. The biomass was 258 concentrated and equally distributed in small Petri dishes (60×15mm) up to 100 259 gametophytes per Petri dish and the excess of water was removed with a 200 μ L pipet. To 260 maintain high humidity levels a few (typically 4-8) drops of NSW were added on the edge of 261 the Petri dish. The dishes were sealed with parafilm, wrapped in aluminium foil to keep 262 darkness conditions and transferred for 3 hours to 14°C. Gamete release was induced by 263 adding 5 mL of 4°C NSW and incubating 5 minutes at room temperature under 40 μ mol m−2 264 s−1 photon flux density. Gametophytes were incubated further 30 minutes at 14°C with LED 265 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 11 lighting of 40 μ mol m −2 s−1 photon flux density to allow for further gamete release. The 266 gametes were then separated from gametophytic tissue by filtering through a 10 μ M cell 267 strainer into a 50 mL conical plastic tube and concentrated by centrifuging in a swing-out 268 rotor (4600 rpm) for 5 minutes at Room Temperature (RT). The majority of the supernatant 269 was removed by pipetting and 200-300 μ L of NSW were left to resuspend the gametes. The 270 gamete suspension was then diluted 1:10 and stained with 1 μ L of lugol solution to count in 271 a hemocytometer. The gametes were then diluted to 1X10 4/μ L in NSW. For PEG-mediated 272 transfection, in a laminar flow hood, 20 μ L of RNP complex mixture (or 20 μ L mock control 273 with IDT buffer) were pipetted into a plastic Petri dish (90×20mm) and 100 μ L of the gamete 274 suspension (106 gametes) were gently mixed by pipetting. This gamete/RNP solution was 275 then mixed with 120 μ L of 40% w/v PEG-8000 (Fisher Scientific, Schwerte, Germany) or 276 PEG-4000 (Sigma, Darmstadt, Germany) solution, filtered through a 0.22 μ M filter by 277 pipetting up and down. PEG pipetting and mixing was carried out vigorously with a wide 278 bore 200 μ L filter tip to ensure proper homogenization. A small degree of bubbling is 279 expected and did not, in our hands, influence transformation efficiency. The gametes were 280 then incubated at RT in darkness for 30 minutes. The 20 μ L of the RNP complex was 281 prepared as following: 1.2 μ L of guide RNA (100 μ M, IDT, Leuven, Belgium), 2.8 μ L of IDTE 282 Buffer (IDT), 8 μ L of Alt-R L.b. Cas12a (Cpf1, 15.6 μ M, IDT), 8 μ L of 2.5X NEB Buffer 3.1 283 (New England Biolabs, Ipswich, MA, USA). For transfections involving multiple RNP 284 complexes, each complex was prepared separately, then combined in the Petri dish, the 285 amount of PEG solution was then adjusted accordingly to maintain a 1:1 ratio 286 (Gamete/RNP:PEG). Following transfection, 20 mL of half-strength PES was added to each 287 plate to stop transfection and enable gamete germination. Dishes were incubated overnight 288 in the dark at 22°C (typically 16 hours), then transferred the next morning to 14°C in neutral 289 day conditions (12 h: 12 h, light:dark). 48h after transformation, 2-fluoroadenine (2-FA, 290 Sigma) was added to the Petri dishes to a final concentration of 20 μ M, and samples were 291 incubated in a in neutral day conditions (12 h: 12 h, light:dark) with LED lighting of 20 μ mol 292 m−2 s−1 photon flux density until the isolation of 2-FA resistant algae and therefore bearing 293 putative mutations in APT and the gene of interest. 294 Six to eight weeks after the transfection, the number of growing germlings in 2-FA 295 supplemented culture medium (putative apt mutants) were counted, and some of them 296 were carefully isolated to 2-FA free half-strength PES using forceps to generate enough 297 biomass for genotyping and validation Genotyping PCR was performed using the Terra 298 PCR Direct Polymerase Mix (Takara) or the Kapa G3 Plant PCR kit (RocheBiosystems) 299 with 2uL of the following suspension: small fragments of algal tissue (approximately 1mm2) 300 grinded in 60 μ L of Nuclease free Water (Ambion). Sanger sequencing was performed by 301 Azenta. Details of gRNA and primers are provided in Table S1. 302 Scytosiphon promiscuus PEG transfection 303 The APT gene of S. promiscuus was identified by aligning DNA sequencings of the 304 Ectocarpus sp.7 APT gene against the S. promiscuus genome 26 using DIAMOND2 27. A 305 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 12 single gene (gene code: mRNA_S-promiscuus_M_contig7.17323.1) on chromosome 28 306 was identified as APT gene of S. promiscuus. 307 Male gametophytes (strain As6m) were cultivated in plastic Petri dishes (90 × 20 mm) and 308 sterile NSW enriched with full strength PESI medium 28 at 10°C in long day (16h:8h, 309 light:dark) conditions with LED lighting of 20 μ mol m-2s-1 photon flux density. Medium was 310 renewed every week until gamete collection. Mature gametophytes release gametes 311 immediately after the medium renewal and thus cultures were closely inspected after each 312 media change. Because S. promiscuus gametes have negative phototaxis, freshly released 313 gametes accumulate on the opposite side of a light source in a Petri dish. The accumulated 314 gametes were collected and diluted to 0.4–10 million per 1 mL with NSW. For PEG 315 transfection, 100 μ L of the gamete suspension was gently mixed with 8 μ L of RNP complex 316 and 108 μ L of 40% w/v PEG-4000 solution in a plastic Petri dish (90 × 20 mm), and then, 317 incubated at room temperature in dark for 30 minutes. To prepare 8 μ L of the RNP 318 complex, two different guide RNAs were mixed as following:1 μ L of guide RNA1 (100 μ M), 319 1 μ L of guide RNA2 (100 μ M), 1.5 μ L of Alt-R L.b. Cas12a (Cpf1, 67 μ M), 3.2 μ L of 2.5X 320 NEB Buffer 3.1, and 1.3 μ L of water. After the transfection, 50 mL of full-strength PESI was 321 added to the Petri dishes and incubated in a 22°C dark condition for 48 hours. Then, 2-FA 322 was added to the Petri dishes (final 2-FA concentration of 20 mM), and the samples were 323 incubated in a 14°C℃ neutral day condition until the isolation of putative APT mutants. One 324 month after the transfection, the number of growing germlings in 2-FA supplemented 325 culture medium (putative apt mutants) were counted, and eight of them were isolated using 326 glass Pasteur pipets, before being genotyped to validate the mutations. Genotyping was 327 performed as in Ectocarpus. Details of gRNA and primers are provided in Table S1. 328 Laminaria digitata PEG transfection 329 The APT gene of Laminaria. digitata was identified by aligning Ectocarpus sp. 7 APT gene 330 against L. digitata genome26, as described above. The APT gene of Laminaria (gene codes: 331 mRNA_L-digitata_M_contig27133.1.1 and mRNA_L-digitata_M_contig6833.2.1) was 332 divided into two contigs (contig27133 and contig6833) because of the fragmented reference 333 genome assembly. 334 Sporophytes of L. digitata were collected at Santec, France on 8 December 2022. The 335 sporophytes were washed with sterile NSW and pat dried to remove water. The sori were 336 excised from the sporophytes and transferred to the laboratory in a chilled transport box. To 337 induce meiospore release, the sori were placed in fresh sterile NSW at room temperature 338 one day after the sampling. However, this did not result in immediate meiospore release. 339 The sori were therefore incubated in sterile NSW in 14°C neutral day conditions until 340 meiospores were released up to two days. Meiospores were collected from three 341 independent sporophyte individuals. Meiospore release resulted in a change in water colour 342 towards brown tones. This brown seawater was transferred to 50 mL tubes and centrifuged 343 at 4000 g for 1 minute. After the centrifugation, the supernatant was removed and fresh 344 sterile seawater was added to the pellet of meiospores and the density of meiospores was 345 adjusted to 0.54–7.7 million per 1 mL. PEG transfection was performed as in S. 346 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 13 promiscuus, except that each guide RNA was used independently. The RNP complex were 347 prepared as following: 1 μ L of guide RNA (100 μ M), 0.75 μ L of Alt-R L.b. Cas12a (Cpf1, 67 348 μ M), 3.2 μ L of 2.5X NEB Buffer 3.1, and 3.05 μ L of water. After the transfection, 50 mL of 349 full-strength PESI was added to the Petri dishes and incubated in a 22°C dark condition for 350 48 hours. Then, 2-FA was added to the Petri dishes to reach a concentration of 20 mM and 351 incubated in a 14°C neutral day condition (green light) until the isolation of putative apt 352 mutants. Some of these putative mutants were isolated and genotyped to confirm the 353 mutations. Genotyping was performed as in Ectocarpus. Details of gRNA and primers are 354 provided in Table S1. 355 Undaria pinnatifida PEG transfection 356 The APT gene of Undaria pinnatifida was identified by aligning Ectocarpus sp. 7 APT gene 357 against the U. pinnatifida genome 29, as described above. The APT gene of Undaria 358 (UNPIN0032CG0760) is located on contig LG22. 359 Mature sporophytes of U. pinnatifida were obtained by crossing the male gametophyte 360 strains Un1f and Un1m, isolated from a single sporophyte (Bizeux, St. Malo, France) 361 cultured in sterile NSW enriched with half-strength PES under controlled laboratory 362 conditions (16–20°C, 16:8 h light:dark) with LED lighting at 10/i2μ mol/i2 m/i2 ²/i2 s/i2 ¹ photon flux 363 density. Spore release was induced by transferring sporophyll tissue to fresh NSW and 364 incubating at room temperature with gentle shaking for 20 minutes. The spore suspension 365 was then collected and its concentration was measured using a haemocytometer. A 366 suspension of approximately 7.5×10/i2 spores per 100 μ L was used for transfection. 367 For PEG transfection, 100 μ L of the spore suspension was gently mixed with 20 μ L of RNP 368 complex and 120 μ L of 40% w/v PEG-8000 (Fisher Scientific, Schwerte, Germany), filtered 369 through a 0.22 /i2μ m syringe filter, in a sterile plastic Petri dish (90 × 20 mm), and then 370 incubated at room temperature in the dark for 30 minutes. 20 /i2μ L of RNP complex was 371 prepared as follows: for Cas9, 4 μ L of crRNA:tracrRNA duplex (100 μ M each, IDT, Leuven, 372 Belgium), annealed in IDT Duplex Buffer, 8 μ L of Cas9 enzyme (15.6 μ M, IDT), and 8 μ L of 373 2.5X NEB 3.1 Buffer (New England Biolabs, Ipswich, MA, USA); for Cas12a, 4 μ L of crRNA 374 (100 μ M, IDT Leuven, Belgium), 8 μ L Alt-R L.b. Cas12a (Cpf1,IDT), and 8 μ L of 2.5X NEB 375 3.1 Buffer. For dual RNP transfections, each complex was prepared separately and 376 combined in the dish prior to PEG addition. 377 Following transfection, 20 mL of sterile NSW supplemented with half-strength PES was 378 added to each dish. The plates were incubated overnight at room temperature in the dark 379 and then transferred the next morning to normal culture conditions at 20°C with a 16:8 380 light:dark cycle. 381 Forty-eight hours after transformation, 2-fluoroadenine (2-FA, Sigma) was added to each 382 plate at a final concentration of 75 /i2μ M. Samples were incubated under standard growth 383 conditions with LED lighting of 10–20 μ mol/i2 m/i2 ²/i2 s/i2 ¹ photon flux density for one week. 384 Surviving 2-FA-resistant gametophytes were isolated and grown individually for further 385 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 14 genotyping to validate APT-targeted mutations using Sanger sequencing as in Ectocarpus. 386 Details of gRNA and APT primers are provided in Table S1. 387 388 Supplementary Figures 389 Figure S1. Protocol optimization for Ectocarpus. A) Scatter plot showing the effect of 390 the heat shock at 22ºC on the number of 2-FA resistant individuals. The horizontal and 391 vertical lines in the scatter plot represent the mean and standard deviation, respectively. No 392 significant effect was detected (exact Wilcoxon rank sum test, p = 0.06977). B) Scatter plot 393 showing the effect of the crRNA:Cas12 ratio (1:1 vs. 4:1) on the number of 2-FA resistant 394 individuals. No significant effect was detected (exact Wilcoxon rank sum test, p = 0.6195). 395 Table Legends 396 Table S1. crRNAs (guide RNAs) and primers used in the present study. 397 Table S2. Result summary of the optimization of PEG condition for Ectocarpus. 398 Table S3. Results summary of IMM mutagenesis for Ectocarpus. 399 Table S4. Result summary of the PEG transfection for Scytosiphon, L. digitata , and U. 400 pinnatifida. 401

402 1. Coelho, S.M., and Cock, J.M. (2020). Brown Algal Model Organisms. Annual Review of 403 Genetics 54, 71–92. https://doi.org/10.1146/annurev-genet-030620-093031. 404 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 15 2. Godfroy, O., Uji, T., Nagasato, C., Lipinska, A.P., Scornet, D., Peters, A.F., Avia, K., 405 Colin, S., Mignerot, L., Motomura, T., et al. (2017). DISTAG/TBCCd1 Is Required for 406 Basal Cell Fate Determination in Ectocarpus. The Plant Cell 29, 3102–3122. 407 https://doi.org/10.1105/tpc.17.00440. 408 3. Godfroy, O., Zheng, M., Yao, H., Henschen, A., Peters, A.F., Scornet, D., Colin, S., 409 Ronchi, P., Hipp, K., Nagasato, C., et al. (2023). The baseless mutant links protein 410 phosphatase 2A with basal cell identity in the brown alga Ectocarpus. Development 411 150, dev201283. https://doi.org/10.1242/dev.201283. 412 4. Coelho, S.M., Godfroy, O., Arun, A., Le Corguillé, G., Peters, A.F., and Cock, J.M. 413 (2011). OUROBOROS is a master regulator of the gametophyte to sporophyte life cycle 414 transition in the brown alga Ectocarpus. Proc. Natl. Acad. Sci. U.S.A. 108, 11518–415 11523. https://doi.org/10.1073/pnas.1102274108. 416 5. Macaisne, N., Liu, F., Scornet, D., Peters, A.F., Lipinska, A., Perrineau, M.-M., Henry, A., 417 Strittmatter, M., Coelho, S.M., and Cock, J.M. (2017). The Ectocarpus IMMEDIATE 418 UPRIGHT gene encodes a member of a novel family of cysteine-rich proteins with an 419 unusual distribution across the eukaryotes. Development 144, 409–418. 420 https://doi.org/10.1242/dev.141523. 421 6. Bringloe, T.T., Starko, S., Wade, R.M., Vieira, C., Kawai, H., Clerck, O.D., Cock, J.M., 422 Coelho, S.M., Destombe, C., Valero, M., et al. (2020). Phylogeny and evolution of the 423 brown algae. Critical Reviews in Plant Sciences 39 , 281–321. 424 https://doi.org/10.1080/07352689.2020.1787679. 425 7. Badis, Y., Scornet, D., Harada, M., Caillard, C., Godfroy, O., Raphalen, M., Gachon, 426 C.M.M., Coelho, S.M., Motomura, T., Nagasato, C., et al. (2021). Targeted CRISPR-427 Cas9-based gene knockouts in the model brown alga Ectocarpus. New Phytologist 231, 428 2077–2091. https://doi.org/10.1111/nph.17525. 429 8. Shen, Y., Motomura, T., Ichihara, K., Matsuda, Y., Yoshimura, K., Kosugi, C., and 430 Nagasato, C. (2023). Application of CRISPR-Cas9 genome editing by microinjection of 431 gametophytes of Saccharina japonica (Laminariales, Phaeophyceae). J Appl Phycol 35, 432 1431–1441. https://doi.org/10.1007/s10811-023-02940-1. 433 9. Hoshino, M., and Kogame, K. (2019). Parthenogenesis is rare in the reproduction of a 434 sexual field population of the isogamous brown alga Scytosiphon (Scytosiphonaceae, 435 Ectocarpales). J. Phycol. 55, 466–472. https://doi.org/10.1111/jpy.12835. 436 10. Hoshino, M., Cossard, G., Haas, F.B., Kane, E.I., Kogame, K., Jomori, T., Wakimoto, T., 437 Glemin, S., and Coelho, S.M. (2024). Parallel loss of sexual reproduction in field 438 populations of a brown alga sheds light on the mechanisms underlying the emergence 439 of asexuality. Nat Ecol Evol 8, 1916–1932. https://doi.org/10.1038/s41559-024-02490-440 w. 441 11. Coelho, S.M. (2024). The brown seaweed Ectocarpus. Nat Methods 21, 363–364. 442 https://doi.org/10.1038/s41592-024-02198-6. 4 43 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 16 12. Cossard, G.G., Godfroy, O., Nehr, Z., Cruaud, C., Cock, J.M., Lipinska, A.P., and 444 Coelho, S.M. (2022). Selection drives convergent gene expression changes during 445 transitions to co-sexuality in haploid sexual systems. Nat Ecol Evol 6, 579–589. 446 https://doi.org/10.1038/s41559-022-01692-4. 447 13. Lipinska, A. (2013). Gamete differentiation /i2 : a genetic, biochemical and evolutionary 448 study using Ectocarpus siliculosus. 449 14. Ichihara, K., Yamazaki, T., and Kawano, S. (2022). Genome editing using a DNA-free 450 clustered regularly interspaced short palindromic repeats-Cas9 system in green 451 seaweed Ulva prolifera. Phycological Research 70, 50–56. 452 https://doi.org/10.1111/pre.12472. 453 15. Luthringer, R., Raphalen, M., Guerra, C., Colin, S., Martinho, C., Zheng, M., Hoshino, 454 M., Badis, Y., Lipinska, A.P., Haas, F.B., et al. (2024). Repeated co-option of HMG-box 455 genes for sex determination in brown algae and animals. Science 383, eadk5466. 456 https://doi.org/10.1126/science.adk5466. 457 16. Luthringer, R., Cormier, A., Ahmed, S., Peters, A.F., Cock, J.M., and Coelho, S.M. 458 (2014). Sexual dimorphism in the brown algae. Perspectives in Phycology, 11–25. 459 https://doi.org/10.1127/2198-011X/2014/0002. 460 17. Heesch, S., Serrano /i2 Serrano, M., Barrera /i2 Redondo, J., Luthringer, R., Peters, A.F., 461 Destombe, C., Cock, J.M., Valero, M., Roze, D., Salamin, N., et al. (2021). Evolution of 462 life cycles and reproductive traits: Insights from the brown algae. J of Evolutionary 463 Biology 34, 992–1009. https://doi.org/10.1111/jeb.13880. 464 18. Bogaert, K.A., Delva, S., and De Clerck, O. (2020). Concise review of the genus 465 Dictyota J.V. Lamouroux. J Appl Phycol 32, 1521–1543. https://doi.org/10.1007/s10811-466 020-02121-4. 467 19. Paszkowski, J., Shillito, R.D., Saul, M., Mandák, V., Hohn, T., Hohn, B., and Potrykus, I. 468 (1984). Direct gene transfer to plants. The EMBO Journal 3, 2717–2722. 469 https://doi.org/10.1002/j.1460-2075.1984.tb02201.x. 470 20. Avila-Peltroche, J., Won, B.Y., and Cho, T.Oh. (2022). An improved protocol for 471 protoplast production, culture, and whole plant regeneration of the commercial brown 472 seaweed Undaria pinnatifida. Algal Research 67, 102851. 473 https://doi.org/10.1016/j.algal.2022.102851. 474 21. Choi, G.C., Avila-Peltroche, J., Won, B.Y., and Cho, T.Oh. (2024). Optimization of 475 protoplast isolation and subsequent regeneration from the economically important 476 brown alga Ecklonia cava using response surface methodology. Algal Research 80, 477 103525. https://doi.org/10.1016/j.algal.2024.103525. 478 22. Cock, J.M., Sterck, L., Rouzé, P., Scornet, D., Allen, A.E., Amoutzias, G., Anthouard, V., 479 Artiguenave, F., Aury, J.-M., Badger, J.H., et al. (2010). The Ectocarpus genome and 480 the independent evolution of multicellularity in brown algae. Nature 465, 617–621. 481 https://doi.org/10.1038/nature09016. 482 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint 17 23. Concordet, J.-P., and Haeussler, M. (2018). CRISPOR: intuitive guide selection for 483 CRISPR/Cas9 genome editing experiments and screens. Nucleic Acids Res 46, W242–484 W245. https://doi.org/10.1093/nar/gky354. 485 24. Starr, R.C., and Zeikus, J.A. (1993). UTEX—THE CULTURE COLLECTION OF ALGAE 486 AT THE UNIVERSITY OF TEXAS AT AUSTIN 1993 LIST OF CULTURES 1. Journal of 487 Phycology 29, 1–106. https://doi.org/10.1111/j.0022-3646.1993.00001.x. 488 25. Coelho, S.M., Scornet, D., Rousvoal, S., Peters, N.T., Dartevelle, L., Peters, A.F., and 489 Cock, J.M. (2012). How to cultivate Ectocarpus. Cold Spring Harb Protoc 2012, 258–490 261. https://doi.org/10.1101/pdb.prot067934. 491 26. Denoeud, F., Godfroy, O., Cruaud, C., Heesch, S., Nehr, Z., Tadrent, N., Couloux, A., 492 Brillet-Guéguen, L., Delage, L., Mckeown, D., et al. (2024). Evolutionary genomics of 493 the emergence of brown algae as key components of coastal ecosystems. Cell 187, 494 6943-6965.e39. https://doi.org/10.1016/j.cell.2024.10.049. 495 27. Buchfink, B., Reuter, K., and Drost, H.-G. (2021). Sensitive protein alignments at tree-496 of-life scale using DIAMOND. Nat Methods 18 , 366–368. 497 https://doi.org/10.1038/s41592-021-01101-x. 498 28. Tatewaki, M. (1966). Formation of a Crustaceous Sporophyte with Unilocular Sporangia 499 in Scytosiphon lomentaria. Phycologia 6, 62–66. https://doi.org/10.2216/i0031-8884-6-1-500 62.1. 501 29. Graf, L., Shin, Y., Yang, J.H., Choi, J.W., Hwang, I.K., Nelson, W., Bhattacharya, D., 502 Viard, F., and Yoon, H.S. (2021). A genome-wide investigation of the effect of farming 503 and human-mediated introduction on the ubiquitous seaweed Undaria pinnatifida. Nat 504 Ecol Evol 5, 360–368. https://doi.org/10.1038/s41559-020-01378-9. 505 506 .CC-BY-ND 4.0 International licenseavailable under a 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 preprint (whichthis version posted July 21, 2025. ; https://doi.org/10.1101/2025.07.21.665871doi: bioRxiv preprint